Management guide, WeOS

Management guide, WeOS
6101-3201
Westermo OS
Management Guide
WeOS
www.westermo.com
© Westermo Teleindustri AB
RedFox Series
Wolverine Series
Lynx Series
Falcon Series
Viper Series
Westermo OS Management Guide
Version 4.21.1-0
Legal information
The contents of this document are provided ”as is”. Except as required by applicable law, no warranties of any kind, either express or implied, including, but not
limited to, the implied warranties of merchantability and fitness for a particular
purpose, are made in relation to the accuracy and reliability or contents of this
document. Westermo reserves the right to revise this document or withdraw it at
any time without prior notice.
Under no circumstances shall Westermo be responsible for any loss of data or
income or any special, incidental, and consequential or indirect damages howsoever caused. More information about Westermo can be found at the following
Internet address: http://www.westermo.com
2
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Contents
Legal information
2
Table of Contents
3
I
Introduction to WeOS and its Management Methods
1 Introduction
1.1 Westermo and its WeOS products .
1.2 Getting Started . . . . . . . . . . . . .
1.3 Introduction to WeOS . . . . . . . . .
1.4 How to read this document . . . . .
1.5 Westermo products running WeOS
10
.
.
.
.
.
11
11
11
12
12
14
2 Quick Start
2.1 Starting the Switch for the First Time . . . . . . . . . . . . . . . . . . .
2.2 Modifying the IP Setting . . . . . . . . . . . . . . . . . . . . . . . . . . .
17
17
18
3 Overview of Management Methods
3.1 When to use the WeConfig tool . . . . . . . . . . . . . . . . . . . . . . .
3.2 When to use the Web . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.3 When to use the CLI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
30
31
31
32
4 Management via Web Interface
4.1 Document Conventions . . . . .
4.2 Logging in . . . . . . . . . . . . .
4.3 Navigation . . . . . . . . . . . . .
4.4 System Overview . . . . . . . .
34
35
36
38
41
5 Management via CLI
© 2017 Westermo Teleindustri AB
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
48
3
Westermo OS Management Guide
Version 4.21.1-0
5.1
5.2
5.3
5.4
Overview of the WeOS CLI
Accessing the CLI . . . . .
Using the CLI . . . . . . . .
General CLI commands . .
hierarchy
. . . . . . .
. . . . . . .
. . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
48
50
54
60
6 WeOS SNMP Support
6.1 Introduction and feature overview . . . . . . . . . . . . . . . . . . . . .
6.2 Managing SNMP via the web interface . . . . . . . . . . . . . . . . . .
6.3 Manage SNMP Settings via the CLI . . . . . . . . . . . . . . . . . . . .
64
64
77
81
II
86
Common Switch Services
7 General Switch Maintenance
7.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7.2 Maintenance via the Web Interface . . . . . . . . . . . . . . . . . . . .
7.3 Maintenance via the CLI . . . . . . . . . . . . . . . . . . . . . . . . . . .
87
87
123
136
8 General System Settings
8.1 Overview of General System Features . . . . . . . . . . . . . . . . . .
8.2 Managing System Settings via Web . . . . . . . . . . . . . . . . . . . .
8.3 Managing System Settings via CLI . . . . . . . . . . . . . . . . . . . . .
166
166
169
174
9 Authentication, Authorisation and Accounting
9.1 Overview over AAA . . . . . . . . . . . . . . . . . .
9.2 Managing AAA via the web . . . . . . . . . . . . .
9.3 Managing AAA via the CLI . . . . . . . . . . . . . .
9.4 Feature Parameters . . . . . . . . . . . . . . . . . .
.
.
.
.
188
189
198
218
237
10 Ethernet Port Management
10.1 Overview of Ethernet Port Management . . . . . . . . . . . . . . . . .
10.2 Managing port settings via the web interface . . . . . . . . . . . . .
10.3 Managing port settings via the CLI . . . . . . . . . . . . . . . . . . . .
238
238
253
257
11 Ethernet Statistics
11.1 Ethernet Statistics Overview . . . . . . . . . . . . . . . . . . . . . . . .
11.2 Statistics via the web interface . . . . . . . . . . . . . . . . . . . . . . .
11.3 Statistics via the CLI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
267
267
274
279
12 SHDSL Port Management
12.1 Overview of SHDSL Port Management . . . . . . . . . . . . . . . . . .
12.2 Managing SHDSL ports via the web interface . . . . . . . . . . . . .
282
282
288
4
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
12.3 Managing SHDSL ports via the CLI . . . . . . . . . . . . . . . . . . . .
296
13 ADSL/VDSL Port Management
13.1 Overview of ADSL/VDSL Port Management . . . . . . . . . . . . . . .
13.2 Managing ADSL/VDSL ports via the web interface . . . . . . . . . .
13.3 Managing ADSL/VDSL ports via the CLI . . . . . . . . . . . . . . . . .
302
302
316
328
14 Power Over Ethernet (PoE)
14.1 Overview of Power over Ethernet (PoE) . . . . . . . . . . . . . . . . .
14.2 Managing PoE via the web interface . . . . . . . . . . . . . . . . . . .
14.3 Managing PoE via the CLI interface . . . . . . . . . . . . . . . . . . . .
333
333
337
341
15 Virtual LAN
15.1 VLAN Properties and Management Features . .
15.2 Port-based network access control . . . . . . . .
15.3 Managing VLAN settings via the web interface
15.4 Managing VLAN settings via the CLI . . . . . . .
.
.
.
.
344
344
355
360
370
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
16 FRNT
16.1 Overview of the FRNT protocol and its features
16.2 FRNT and RSTP coexistence . . . . . . . . . . . . .
16.3 Managing FRNT settings via the web interface .
16.4 Managing FRNT settings via the CLI . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
382
382
386
388
393
17 Ring Coupling and Dual Homing
17.1 Overview . . . . . . . . . . . . . .
17.2 Managing via the Web . . . . .
17.3 Managing via CLI . . . . . . . . .
17.4 Feature Parameters . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
396
397
411
415
425
18 Spanning Tree Protocol - RSTP and STP
18.1 Overview of RSTP/STP features . . . . . . . . . . . . . . . . . . . . . . .
18.2 Managing RSTP via the web interface . . . . . . . . . . . . . . . . . .
18.3 Managing RSTP via the CLI . . . . . . . . . . . . . . . . . . . . . . . . .
426
426
432
436
19 Link Aggregation
19.1 Link Aggregation Support in WeOS . . . . . . . . . . . . . . . . . . . .
19.2 Managing Link Aggregation via the Web . . . . . . . . . . . . . . . . .
19.3 Managing Link Aggregation via CLI . . . . . . . . . . . . . . . . . . . .
441
441
452
456
20 Multicast in Switched Networks
20.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
461
461
© 2017 Westermo Teleindustri AB
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
5
Westermo OS Management Guide
Version 4.21.1-0
20.2 Managing IGMP in the Web Interface . . . . . . . . . . . . . . . . . . .
20.3 Managing IGMP in the CLI . . . . . . . . . . . . . . . . . . . . . . . . . .
21 General Network Settings
21.1 Overview . . . . . . . . . . . . . . . . . . . . .
21.2 Network interfaces . . . . . . . . . . . . . . .
21.3 General IP settings . . . . . . . . . . . . . . .
21.4 Managing network interfaces via the web
21.5 Managing general IP settings via the web
21.6 Managing network interfaces via the CLI
21.7 Managing general IP settings via the CLI
21.8 Feature Parameters . . . . . . . . . . . . . .
467
469
.
.
.
.
.
.
.
.
473
473
474
490
493
500
506
517
533
22 DHCP Server
22.1 Overview of DHCP Server Support in WeOS . . . . . . . . . . . . . . .
22.2 Configuring DHCP Server Settings via the Web . . . . . . . . . . . .
22.3 Configuring DHCP Server Settings via the CLI . . . . . . . . . . . . .
534
535
549
556
23 DHCP Relay Agent
23.1 Overview of DHCP Relay Agent Support . . . . . . . . . . . . . . . . .
23.2 Configuring DHCP Relay Agent via the Web . . . . . . . . . . . . . . .
23.3 Configuring DHCP Relay Agent via the CLI . . . . . . . . . . . . . . .
568
569
580
583
24 Alarm handling, LEDs and Digital
24.1 Alarm handling features . . . . .
24.2 Managing Alarms via the Web .
24.3 Managing Alarms via the CLI . .
24.4 Digital I/O . . . . . . . . . . . . . .
24.5 LEDs . . . . . . . . . . . . . . . . . .
.
.
.
.
.
590
590
603
609
635
637
25 Logging Support
25.1 Logging Support in the web interface . . . . . . . . . . . . . . . . . .
25.2 Managing Logging Support via the CLI . . . . . . . . . . . . . . . . . .
640
641
642
III
644
I/O
. . .
. . .
. . .
. . .
. . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Router/Gateway Services
26 IP Routing in WeOS
26.1 Summary of WeOS Routing and Router Features . . . . . . . . . . .
26.2 Static unicast routes via Web . . . . . . . . . . . . . . . . . . . . . . . .
26.3 Enabling Routing, Managing Static Routing, etc., via CLI . . . . . .
6
645
645
653
656
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
27 Dynamic Routing with OSPF
27.1 Overview of OSPF features . . . . . . . . . . . . . . . . . . . . . . . . .
27.2 OSPF Web . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
27.3 Managing OSPF via the CLI . . . . . . . . . . . . . . . . . . . . . . . . .
658
658
672
676
28 Dynamic Routing with RIP
28.1 Overview of RIP Features . . . . . . . . . . . . . . . . . . . . . . . . . . .
28.2 RIP Web . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
28.3 Managing RIP via the CLI . . . . . . . . . . . . . . . . . . . . . . . . . . .
688
688
694
697
29 IP Multicast Routing
29.1 Summary of WeOS Multicast Routing Features . . . . . . . . . . . . .
29.2 Managing Multicast Routing via Web Interface . . . . . . . . . . . . .
29.3 Managing Multicast Routing via CLI . . . . . . . . . . . . . . . . . . . .
706
706
710
715
30 Virtual Router Redundancy (VRRP)
30.1 Introduction to WeOS VRRP support . . . . . . . . . . . . . . . . . . . .
30.2 Managing VRRP via the web interface . . . . . . . . . . . . . . . . . .
30.3 Managing VRRP via the CLI . . . . . . . . . . . . . . . . . . . . . . . . .
719
720
727
732
31 Firewall Management
31.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
31.2 Firewall Management via the Web Interface . . . . . . . . . . . . . .
31.3 Firewall Management via the CLI . . . . . . . . . . . . . . . . . . . . .
740
741
768
792
IV
806
Virtual Private Networks and Tunnels
32 Overview of WeOS VPN and
32.1 WeOS support for VPNs . .
32.2 Tunneling using PPP . . . .
32.3 Tunneling using GRE . . . .
Tunnel support
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
. . . . . . . . . . . . . . . . . . . . . . . . . .
807
807
808
808
33 PPP Connections
33.1 Overview of PPP Properties and Features . . . . . . . . . . . . . . . .
33.2 Managing PPP settings via the web interface . . . . . . . . . . . . . .
33.3 Managing PPP settings via the CLI . . . . . . . . . . . . . . . . . . . . .
809
810
820
826
34 GRE tunnels
34.1 Overview of GRE tunnel Properties and Management Features . .
34.2 Managing GRE settings via the web interface . . . . . . . . . . . . .
34.3 Managing GRE settings via the CLI . . . . . . . . . . . . . . . . . . . .
837
837
841
843
© 2017 Westermo Teleindustri AB
7
Westermo OS Management Guide
Version 4.21.1-0
35 IPsec VPNs
35.1 Overview of IPsec VPN Management Features
35.2 Managing VPN settings via the web interface
35.3 Managing VPN settings via the CLI . . . . . . .
35.4 Feature Parameters . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
847
848
869
879
896
36 SSL VPN
36.1 Overview of SSL VPN Management Features . . . .
36.2 Managing SSL VPN settings via the web interface
36.3 Managing SSL VPN settings via the CLI . . . . . . .
36.4 Feature Parameters . . . . . . . . . . . . . . . . . . . .
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
897
897
916
922
937
37 WeConnect
37.1 Installing WeConnect via the Web . . . . . . . . . . . . . . . . . . . . .
37.2 Installing WeConnect via the CLI . . . . . . . . . . . . . . . . . . . . . .
37.3 Troubleshooting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
938
940
942
944
V
948
.
.
.
.
.
.
.
.
Serial Port Management and Applications
38 Serial Port Management
38.1 Overview of Serial Port Management . . . . . . . . . . . . . . . . . . .
38.2 Managing serial ports via the web interface . . . . . . . . . . . . . .
38.3 Managing serial ports via the CLI interface . . . . . . . . . . . . . . .
949
950
953
956
39 Serial Over IP
39.1 Overview of Serial Over IP . . . . . . . . . . . . . . . . . . . . . . . . . .
39.2 Managing Serial Over IP via the web interface . . . . . . . . . . . . .
39.3 Managing Serial Over IP via the CLI interface . . . . . . . . . . . . .
962
962
974
981
40 Modbus Gateway
40.1 Managing Modbus Gateway via the web interface . . . . . . . . . .
40.2 Managing Modbus Gateway via the CLI interface . . . . . . . . . . .
997
999
1003
41 MicroLok II Gateway
1012
41.1 Overview of MicroLok Gateway Properties and Management Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1012
41.2 Managing MicroLok Gateway via the web interface . . . . . . . . . . 1017
41.3 Managing MicroLok Gateway via the CLI interface . . . . . . . . . . 1021
8
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
VI
Train Specific Protocols
1028
42 TTDP
1029
42.1 Overview of TTDP Management Features . . . . . . . . . . . . . . . . 1029
42.2 Managing TTDP settings via the CLI . . . . . . . . . . . . . . . . . . . . 1048
VII
Appendixes
1053
Acronyms and abbreviations
1054
References
1057
Index
1062
© 2017 Westermo Teleindustri AB
9
Westermo OS Management Guide
Version 4.21.1-0
Part I
Introduction to WeOS and its
Management Methods
10
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Chapter 1
Introduction
1.1
Westermo and its WeOS products
Westermo provides an extensive set of network products for robust industrial
data communications, managed as well as unmanaged products. Westermo’s
products are found in diverse set of harsh environment applications, and where
robustness and reliability are vital properties.
This guide describes the extensive functionality of managed Westermo products
running the Westermo OS (WeOS).
1.2
Getting Started
Please see www.westermo.com for the latest updated version of this document –
the WeOS Management Guide. There you can also find product User Guides, and
other support information for your product.
The dedicated User Guide of your product includes information on how to get
started with WeOS on your specific product. That is a good place to start if you
wish to do the least possible configuration of your switch (i.e., assign appropriate
IP settings) before putting it into your network infrastructure.
If the User Guide of your specific product lacks a section on how to get started
with WeOS, please visit the chapter 2 (Quick Start) of this document.
© 2017 Westermo Teleindustri AB
11
Westermo OS Management Guide
Version 4.21.1-0
1.3
Introduction to WeOS
Westermo OS (WeOS) is a network operating system delivering an extensive set
of functionality including layer-2 (basic switching, VLAN, IGMP snooping, etc.),
layer-3 (routing, firewall, NAT, etc.), and higher-level services (DHCP, DNS, etc.).
Furthermore, WeOS provides easy management via a Web interface, via the associated WeConfig tool, and via a USB stick. To satisfy even more advanced customer needs, WeOS provides flexible management via a command line interface
(CLI), as well as via SNMP.
WeOS provides two levels of functionality, WeOS Standard and WeOS Extended.
Products running WeOS Standard are outstanding layer-2 switches suitable to
build reliable LAN infrastructures. Products running WeOS Extended extends the
WeOS functionality by adding routing capabilities and a rich set of related higher
level services (NAT, firewall, VPN, etc.).
1.4
How to read this document
This guide is structured in the following parts:
ˆ Part I: This part gives general information on WeOS, and introduces the main
methods to manage a WeOS unit (WeConfig, Web, CLI and SNMP)1 .
The information in Part I applies both to products running WeOS Standard
and WeOS Extended.
– Chapter 1 is this chapter.
– Chapter 2 describes how to get started with your WeOS product.
– Chapters 3 gives an overview of the different ways to manage a WeOS
unit. If you need recommendations of which method to use, please read
chapter 3.
– Chapters 4-5 present the WeOS Web and CLI support. Detailed information for Web and CLI Management is provided in the later parts of the
document.
– Chapters 6 is the main source of information for WeOS SNMP support.
1 For
12
information on how to configure a WeOS unit using a USB memory stick, see Chapter 7.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
ˆ Part II: Each of the chapters in this part covers services and features in
common software levels Standard and Extended.
– Chapter 7 handles general maintenance task (firmware upgrade, configuration file handling, factory reset, etc.) and tools such as ping, traceroute, which be useful when troubleshooting your network.
– Chapters 8-9 handle various general System setttings and AAA services.
– Chapters 10-14 cover management of Ethernet, SHDSL and xDSL (ADSL/VDSL)
ports.
– Chapters 15-20 concern various layer-2 services in WeOS (VLANs, layer2 redundancy (FRNT, RSTP, Link Aggregation), and IGMP Snooping).
– Chapter 21 covers network interface configuration including IP address,
netmask, etc., as well system wide network settings such as default
gateway and DNS.
– Chapters 22-25 handle DHCP services (DHCP Server and Relay), and
status maintenance (Alarm, Digital I/O, Front Panel LEDs, and logging).
ˆ Part III covers WeOS router/gateway services. These features are only applicable for WeOS Extended products.
– Chapters 26-30 describe static and dynamic routing, and VRRP support
in WeOS.
– Chapter 31 concerns NAT and Firewall support.
ˆ Part IV covers WeOS VPN and tunneling services. These features are only
provided for WeOS Extended products.
– Chapter 32 gives an overview to VPN and tunneling services.
– Chapter 33 covers PPP support (PPP over serial port and PPPoE).
– Chapter 34 describes GRE tunneling support.
– Chapters 35-36 present VPN support using IPsec and SSL (OpenVPN).
ˆ Part V contains information on serial port configuration (chapter 38) and
applications. These features apply to WeOS products with serial ports, both
for WeOS Standard and WeOS Extended.
– Chapter 39 describes Serial Over IP and Modem Replacement functionality
– Chapters 40-41 cover Modbus Gateway and Microlok Gateway support.
© 2017 Westermo Teleindustri AB
13
Westermo OS Management Guide
Version 4.21.1-0
ˆ Part VI contains information on train related protocols and train networks.
– Chapter 42 describes WeOS support for the Train Topology Discovery
Protocol (TTDP)[14]. TTDP support is limited to RFR-212-FB products[56]
(Corazon platform).
1.5
Westermo products running WeOS
Below you find the list of Westermo products running WeOS, as well as references
to their respective User Guide:
ˆ Falcon: User Guide [47] (FDV-206-1D1S). (”Basis” platform)
ˆ Lynx: User Guides [52] (Lynx-L110/210) and [48] (Lynx-L106/206-F2G). (”Basis” platform)
ˆ Lynx-DSS: User Guides [49] (L108/208-F2G-S2), [50] (L105/205-S1), and
[51] (L106/206-S2). (”Basis” platform)
ˆ RedFox Industrial (RFI): User Guides [54] (”Corazon” platform) and [53] (”Atlas” platform)
ˆ RedFox Industrial Rack (RFIR): User Guide [55] (”Corazon” platform)
ˆ RedFox Rail (RFR): User Guide [56] (RFR-212-FB (”Corazon” platform), and
RFR-12-FB (”Atlas” platform)).
ˆ Wolverine: User Guides [42] (DDW-142/2422 ), [43] (DDW-142/242-485)3 ,
[44] (DDW-142-BP), [45] (DDW-225) and [46] (DDW-226). (”Basis” platform)
ˆ Viper:
– Viper-12A and Viper-20A (”Coronet” platform): User Guides [59] (Viper112A/212A), [60] (Viper-112A/212A-P8), [61] (Viper-120A/220A), and [62]
(Viper-120A/220A-P8). See the user guides for details on available variants, including number of Gigabit ports.
– Viper-12 (”Basis” platform): [57] (Viper-112/212 and Viper-112/212-T3G)
and [58] (Viper-112/212-P8 and Viper-112/212-T3G-P8).
2 DDW-142
and DDW-242 are also referred to as DDW-x42 in this guide.
and DDW-242-485 are also referred to as DDW-x42-485 in this guide.
3 DDW-142-485
14
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Note
Atlas, Basis, Corazon and Coronet denote HW platforms used by different
products. Products utilising the same HW platform use the same kind of
CPU, and have the same amount of RAM and flash memory.
1.5.1
Product hardware details affecting WeOS functionality
© 2017 Westermo Teleindustri AB
Digital In/Out
USB Port
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X2
X2
X2
X
X
X
X
X
X
X
X
X
X
X
X1
X
X
X
X
X
X
X
X
X
X
X
X
PoE Ports
Console port
X
Bypass Relay
Serial Port(s)
X
xDSL Port
Falcon
FDV-206-1D1S
Lynx
L106/206-F2G
L110/210
Lynx-DSS
All Lynx-DSS models
RedFox Industrial &
RedFox Industrial Rack
All RFI and RFIR models
RedFox Rail
All RFR models
Viper
All ”non-PoE’ models
All ”PoE” models
Wolverine
DDW-x42
DDW-x42-485
DDW-x42-BP
DDW-225
DDW-226
SHDSL Ports
Ethernet Ports
The WeOS functionality described in the Management Guide generally applies to
all Westermo products running WeOS of the appropriate software level (Standard
or Extended). However, where functionality assumes the presence of certain
hardware (such as a USB port), those functions are limited to products including
that hardware. The table below provides a summary of hardware differences affecting the availability of certain WeOS functions. For a more definite description
of hardware specifications you are referred to the dedicated User Guide of each
product (see section 1.5).
X
X1
15
Westermo OS Management Guide
Version 4.21.1-0
1 Bypass Relay is available on DDW-x42-BP, and RedFox Rail models ”RFR-12 FB” and ”RFR212 FB”. See the related User Guides, listed in section 1.5, for more information on bypass relay
functionality.
2 The DDW-x42 SHDSL ports have support for PAF (SHDSL link bonding).
16
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Chapter 2
Quick Start
This section provides a guide to quickly get started with your switch. Only simple
configuration procedures will be covered1 . The steps covered concern:
ˆ Get familiar with the factory default setting
ˆ Configuring an appropriate IP address
2.1
Starting the Switch for the First Time
When booting the switch for the first time the switch will use the factory default
setting.
The factory default setting makes the switch operate as a manageable layer-2
switch, where all Ethernet ports belong to the same virtual LAN (VLAN)2 .
ˆ Manageable: The switch is manageable via any of the Ethernet ports. To
manage the switch via an Ethernet port you need to know the IP address of
the switch (see table 2.1). For switches equipped with a console port, the
switch can as well be managed via that port without knowing the IP address
of the switch.
1 For
more advanced settings, we refer to the remaining chapters of this guide as well as the
online help provided via the Web configuration tool and the Command Line Interface (CLI).
2 On Falcon series of switches, all Ethernet ports belong to the default VLAN (VLAN 1), while the
xDSL port belongs to a separate VLAN (VLAN 1006). That is, by factory default Falcon operates as
a router. See chapter 13 for more details.
© 2017 Westermo Teleindustri AB
17
Westermo OS Management Guide
Version 4.21.1-0
ˆ Single VLAN: By default all ports on the switch will belong to the same VLAN.
Thus, devices connected to different ports of the switch should be able to
communicate with each other right away. For more advanced setups, the
ports of the switch can be grouped into different VLANs. In the factory default setting all ports belong to VLAN 1.
The default IP setting for the switch is as shown in table 2.1.
Primary IP address
Secondary IP address
Address
Netmask
Gateway
Dynamic (DHCP)
192.168.2.200
(Dynamic)
255.255.255.0
(Dynamic)
Disabled
Table 2.1: Factory Default IP settings.
Thus, when you power up your WeOS unit with the factory configuration, you can
connect to it via two addresses:
ˆ The static IP address 192.168.2.200: This address is simplest to use if you
are setting up a single unit.
ˆ A dynamic address assigned by a DHCP server3 (if present): This address
may be simplest to use if you want to connect and configure multiple new
WeOS units simultaneously.
Note
Before you put your switch into your production network you should change
its IP setting according to your network topology. How you change your IP
setting is described in the next section.
2.2
Modifying the IP Setting
The switch can be configured with a static IP setting, or it can get its IP address
dynamically via DHCP. The latter case is useful if you are running a DHCP server
on the same LAN as the switch will be located.
WeOS provides several management tools, which will be presented further in
later chapters of this guide. In this chapter we limit the scope to describe how
these tools can be used to update the IP settings of the switch.
3 In addition, the unit will autoconfigure itself with a link-local address in the 169.254.x.x range,
where ’x’ is in interval 0-255. See section 21.2.6 for more information.
18
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
ˆ WeConfig: is Westermo’s Network configuration management tool (NCM)
made for commissioning and maintenance of components in a network. It
replaces the former Westermo tool known as IPConfig. For further information on WeConfig’s features and how to use the tool, see the WeConfig User
Guide[64].
ˆ Web: Configuration of IP settings via the Web interface is described in section 2.2.1.
ˆ CLI: Configuration of IP settings via the Command Line Interface (CLI) is
described in section 2.2.2.
Hint
If you are not sure what IP address your switch has, use the WeConfig tool,
or the CLI via console method (section 2.2.2.1). If neither of these methods
work, please visit section 7.1.3 for information on how to conduct a factory
reset.
© 2017 Westermo Teleindustri AB
19
Westermo OS Management Guide
Version 4.21.1-0
2.2.1
Using the Web Interface to Update the Switch IP Settings
To configure the IP settings via web your switch is required to be located on the
same IP subnet as your PC.
To Internet or
company Intranet
Switch with default IP setting:
IP address: 192.168.2.200
Netmask: 255.255.255.0
Default gateway: Disabled
Should get the following settings:
IP address: 192.168.55.100
Netmask: 255.255.255.0
Default gateway: 192.168.55.1
WeOS switch
Router
Console
Ethernet ports
Router IP address:
192.168.55.1
PC
Host with Web browser.
PC IP address and netmask known, e.g.,
IP address 192.168.55.35 and netmask 255.255.255.0
In this example the switch shall be assigned the IP address 192.168.55.100, netmask 255.255.255.0 and default gateway 192.168.55.1. To achieve this you must
(temporarily) change the IP address of the PC in order to be able to communicate
with the switch.
The steps to configure the IP settings via the web interface are as follows:
1. Connect your PC to the switch: Connect your PC to the switch as shown in
the figure above.
2. Modifying IP Settings on PC: The IP settings on the PC must be updated to
match the default settings on the switch, i.e., the PC should be assigned an
IP address on the 192.168.2.0/24 network, e.g.,
ˆ PC IP address: 192.168.2.1
ˆ PC Netmask: 255.255.255.0
3. Access switch via web browser: Open your web browser and enter URL
http://192.168.2.200 in the browser’s address field. You will be asked to
enter a username and a password. Use the factory default account settings
shown below:
ˆ Login username: admin
ˆ Password: westermo
20
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
4. Open the Network configuration page: Click on the Configuration topmenu and then on the Network sub-menu and then the Global settings
menu.
5. Configure Default Gateway: Now click the edit icon ( ) in the Global Settings
frame. The following page should appear.
Fill in the appropriate address in the Default Gateway field. In this example,
the default gateway is 192.168.55.1. Click the Apply button. Your switch is
configured with a new default gateway.
6. Open Interface Configuration Page: Click on the Configuration top-menu
and then on the Network sub-menu and then the Interface sub menu. In
© 2017 Westermo Teleindustri AB
21
Westermo OS Management Guide
Version 4.21.1-0
the Interface page, click the edit icon ( ) on the row for the interface
named vlan1. The Interface Configuration Page will appear:
7. Configure Interface IP Settings: Enter the appropriate IP settings for your
switch. In this example we would:
(a) Set IP Address Method to static (radio button).
(b) Set Primary Address to 192.168.55.100 with 255.255.255.0 in the
Netmask field.
(c) Remove Secondary Address (192.168.2.200) using the trash icon ( ).
Click the Apply button and your switch is configured with a new IP address.
8. Reconfigure PC’s IP Settings: As the IP address is changed on the switch,
you cannot reach it from your PC any longer. To access the switch from the
PC, the PC’s IP settings must be changed again. In this case, we assume it
is changed back to its original settings:
ˆ PC IP address: 192.168.55.35
ˆ PC Netmask: 255.255.255.0
ˆ PC Default Gateway: 192.168.55.1
Further management of the switch can be performed via any of the available
management tools - WeConfig, Web, SSH/Telnet/CLI or SNMP.
22
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
2.2.2
Using the CLI to Update the Switch IP Settings
The CLI can be accessed in three ways: via the console port (given that the switch
is equipped with a console port) or via the Ethernet ports using the Secure Shell
(SSH) or the Telnet protocol. Section 2.2.2.1 explains how to access the CLI via
the console port, and how to update the IP settings. Section 2.2.2.2 explains how
to access the CLI via SSH.
Access with Telnet is also possible, but this is not enabled by default on the
switch, and to use it you will first have to access it with one of the other methods
and enable this protocol for management. See Section 8.3.24 (CLI) for information on how to enable the Telnet service on the unit, and then Section 21.4 (Web)
or Section 21.6.6 for information on how to enable Telnet configuration via interface ”vlan1”.
2.2.2.1
Accessing the CLI via the console port
For WeOS switches equipped with a console port, this port can be used to change
IP address of the switch.
1. Connect your PC to the switch: Connect your PC to the switch as shown in
the figure below.
To Internet or
company Intranet
Switch with default IP setting:
IP address: 192.168.2.200
Netmask: 255.255.255.0
Default gateway: Disabled
Should get the following settings:
IP address: 192.168.55.100
Netmask: 255.255.255.0
Default gateway: 192.168.55.1
WeOS switch
Console
Router
Ethernet ports
Router IP address:
192.168.55.1
PC
Host with terminal emulation program.
PC IP address and netmask known, e.g.,
IP address 192.168.55.35 and netmask 255.255.255.0
Important notice for WeOS Switches equipped with a console port
See the User Guide of your specific product (section 1.5) for information
on what Diagnostic Cable to use when connecting to the console port
of your specific product.
© 2017 Westermo Teleindustri AB
23
Westermo OS Management Guide
Version 4.21.1-0
2. Terminal program: To communicate with the switch via the console port, you
need to use a terminal emulation program on your PC, such as Hyperterminal. Ask your system administrator if you need help to install or configure
your terminal emulation program.
The following settings should be used when connecting to the console port:
Console Port Parameter
Setting
Data rate
Data bits
Stop bits
Parity
Flow control
115200 bits/s
8
1
Off
Off
3. Activating the console: When the switch has finished booting, you will be
asked to press the Enter key on your keyboard to activate the console.
4. Logging in: Now you will be asked to enter a username and thereafter a
password. For a switch using the factory default settings, use the following
login username and password:
ˆ Login username: admin
ˆ Password: westermo
Below you see a sample printout when logging in on a WeOS switch. (The
password is not ”echoed” back to the screen.)
Example
example login: admin
Password:
.--.--.--.-----.-----.------.-----.-.--.--------.-----.
_| -__|
_| . . | _ | http://www.westermo.com
| | | | -__|__ --|_
\__/\__/|_____._____| |__| |_____|__| |__|__|__|_____|
[email protected]
Robust Industrial Data Communications -- Made Easy
\\/ Westermo WeOS v4.15.0 4.15.0 -- Jun 16 19:10 CEST 2014
Type: ’help’ for help with commands, ’exit’ to logout or leave a context.
example:/#>
5. Listing IP address: Use the CLI command ”show iface” to list information
about network interfaces.
24
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/#> show iface
Press Ctrl-C or Q(uit) to quit viewer, Space for next page, <CR> for next line.
Interface Name
---------------lo
vlan1
Oper
---UP
UP
Address/Length
MTU
MAC/PtP Address
------------------ ----- --------------------------127.0.0.1/8
16436 N/A
192.168.2.200/24
1500
00:07:7c:10:de:e1
169.254.145.230/16
-----------------------------------------------------------------------------example:/#>
6. Changing IP address and netmask: To change the switch IP addressing mode
(”static” instead of ”DHCP”), set a static address and netmask, and to skip
secondary addresses, use CLI commands ”configure”, ”iface vlan1”,
”inet static”, ”address <IPV4ADDRESS/LEN>”, ”no address secondary”
and ”end” as shown below. This example is based on the setup in step 1,
and configures the switch with an address (192.168.55.100/24) on the same
IP subnet as the PC.
Example
example:/#> configure
example:/config/#> iface vlan1
example:/config/iface-vlan1/#> inet static
example:/config/iface-vlan1/#> address 192.168.55.100/24
example:/config/iface-vlan1/#> no address secondary
Remove all secondary IP addresses, are you sure (y/N)? y
Removing all secondary IPs!
example:/config/iface-vlan1/#> end
example:/config/#> end
Stopping DHCP Clients ...................................... [ OK ]
Configuration activated. Remember "copy run start" to save to flash (NVRAM).
example:/#> show iface
Press Ctrl-C or Q(uit) to quit viewer, Space for next page, <CR> for next line.
Interface Name
Oper Address/Length
MTU
MAC/PtP Address
---------------- ---- ------------------ ----- --------------------------lo
UP
127.0.0.1/8
16436 N/A
vlan1
UP
192.168.55.100/24
1500
00:07:7c:10:de:e1
-----------------------------------------------------------------------------example:/#>
7. Set default gateway IP address: The figure below shows the same network
setup, but with a router attached to the IP subnet.
With this setup you would like to configure a default gateway IP address
to allow management of the switch from outside the local network. This
© 2017 Westermo Teleindustri AB
25
Westermo OS Management Guide
Version 4.21.1-0
can be achieved using CLI commands ”configure”, ”ip”, ”route default
192.168.55.1 <IPADDRESS>”, and ”end” as shown below.
Example
example:/#> configure
example:/config/#> ip
example:/config/ip/#> route default 192.168.55.1
example:/config/ip/#> end
example:/config/#> end
Configuration activated. Remember "copy run start" to save to flash (NVRAM).
example:/#>
8. Save configuration: Although the configuration changes has been activated,
the running configuration must be stored to the startup configuration. Otherwise the changes will be lost if the switch is rebooted.
Example
example:/#> copy running-config startup-config
example:/#>
9. You are now done setting the IP address, subnet mask and default gateway
of your switch. Logout from the CLI using the ”logout” command.
Further management of the switch can be performed via any of the available
management tools - WeConfig, Web, SSH/Telnet/CLI or SNMP.
2.2.2.2
Accessing the CLI via SSH
Configuring the IP settings via SSH/CLI is very similar to configuring them via the
console port. The major differences are:
ˆ The IP address of the PC must (temporarily) be changed in order to be able
to communicate with the switch, i.e., the PC should have an address on
network 192.168.2.0/24, e.g., 192.168.2.1/24.
ˆ After the IP settings have been changed on the switch, the PC is likely to
loose contact with the switch. The PC must therefore change its IP address
again, and login to the switch again in order to copy the running configuration to the startup configuration.
The steps to configure the IP settings via SSH/CLI are as follows:
1. Connect your PC to the switch: Connect your PC to the switch as shown in
the figure below. In this example we assume the switch will get IP address
26
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
192.168.55.100, netmask 255.255.255.0 and default gateway 192.168.55.1.
To Internet or
company Intranet
Switch with default IP setting:
IP address: 192.168.2.200
Netmask: 255.255.255.0
Default gateway: Disabled
Should get the following settings:
IP address: 192.168.55.100
Netmask: 255.255.255.0
Default gateway: 192.168.55.1
WeOS switch
Router
Console
Ethernet ports
Router IP address:
192.168.55.1
PC
Host with SSHv2 client.
PC IP address and netmask known, e.g.,
IP address 192.168.55.35 and netmask 255.255.255.0
2. Modifying IP Settings on PC: The IP settings on the PC must be updated to
match the default settings on the switch, i.e., the PC should be assigned an
IP address on the 192.168.2.0/24 network, e.g.,
ˆ PC IP address: 192.168.2.1
ˆ PC Netmask: 255.255.255.0
ˆ PC Default Gateway: Not needed
3. Connecting and Logging in: When connecting via SSH you will be asked to
enter a username and thereafter a password. For a switch using the factory
default settings, use the following login username and password:
ˆ Login username: admin
ˆ Password: westermo
The procedure to connect may vary slightly depending on what SSH client
you are using. The example below show the connection procedure using
Unix OpenSSH4 . (On Windows one can use Putty5 .)
4 OpenSSH,
5 Putty,
http://www.openssh.com
http://www.chiark.greenend.org.uk/~sgtatham/putty/
© 2017 Westermo Teleindustri AB
27
Westermo OS Management Guide
Version 4.21.1-0
Example
[email protected]:~$ ssh [email protected]
The authenticity of host ’192.168.2.200 (192.168.2.200)’ can’t be established.
RSA key fingerprint is 6d:0c:f3:d3:28:d6:d8:43:bc:69:f8:d0:d6:a2:27:87.
Are you sure you want to continue connecting (yes/no)? yes
Warning: Permanently added ’192.168.2.200’ (RSA) to the list of known hosts.
[email protected]’s password:
.--.--.--.-----.-----.------.-----.-.--.--------.-----.
_| -__|
_| . . | _ | http://www.westermo.com
| | | | -__|__ --|_
\__/\__/|_____._____| |__| |_____|__| |__|__|__|_____|
[email protected]
Robust Industrial Data Communications -- Made Easy
\\/ Westermo WeOS v4.15.0 4.15.0 -- Jun 16 19:10 CEST 2014
Type: ’help’ for help with commands, ’exit’ to logout or leave a context.
example:/#>
4. Changing IP settings: The switch IP settings are changed with the same
commands as described when accessing the CLI via the console port (section 2.2.2.1). In this example we assign IP address, netmask and default
gateway.
Example
example:/#> configure
example:/config/#> iface vlan1
example:/config/iface-vlan1/#> inet static
example:/config/iface-vlan1/#> address 192.168.55.100/24
example:/config/iface-vlan1/#> no address secondary
Remove all secondary IP addresses, are you sure (y/N)? y
Removing all secondary IPs!
example:/config/iface-vlan1/#> end
example:/config/#> ip
example:/config/ip/#> route default 192.168.55.1
example:/config/ip/#> end
example:/config/#> end
The configuration is now changed, but not yet saved to the startup configuration. However, as the IP address is changed, the SSH connection will be
broken.
5. Logging in again to save configuration: To login again, the PC’s IP settings
must be changed again. In this case, we assume it is changed back to its
original settings:
ˆ PC IP address: 192.168.55.35
ˆ PC Netmask: 255.255.255.0
ˆ PC Default Gateway: 192.168.55.1
28
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
We can then login again to copy the running configuration to startup configuration.
Example
[email protected]:~$ ssh [email protected]8.55.100
The authenticity of host ’192.168.55.100 (192.168.55.100)’ can’t be established.
RSA key fingerprint is 6d:0c:f3:d3:28:d6:d8:43:bc:69:f8:d0:d6:a2:27:87.
Are you sure you want to continue connecting (yes/no)? yes
Warning: Permanently added ’192.168.55.100’ (RSA) to the list of known hosts.
[email protected]’s password:
.--.--.--.-----.-----.------.-----.-.--.--------.-----.
_| -__|
_| . . | _ | http://www.westermo.com
| | | | -__|__ --|_
\__/\__/|_____._____| |__| |_____|__| |__|__|__|_____|
[email protected]
Robust Industrial Data Communications -- Made Easy
\\/ Westermo WeOS v4.15.0 4.15.0 -- Jun 16 19:10 CEST 2014
Type: ’help’ for help with commands, ’exit’ to logout or leave a context.
example:/#> copy running-config startup-config
example:/#>
You are now done setting the IP address, subnet mask and default gateway
of your switch. Logout from the CLI using the ”logout” command.
Further management of the switch can be performed via any of the available
management tools - WeConfig, Web, SSH/CLI or SNMP.
© 2017 Westermo Teleindustri AB
29
Westermo OS Management Guide
Version 4.21.1-0
Chapter 3
Overview of Management
Methods
WeOS is managed and monitored using the following tools and interfaces:
ˆ WeConfig: is Westermo’s Network configuration management tool (NCM)
made for commissioning and maintenance of components in a network. It
replaces the former Westermo tool known as IPConfig. For further information on WeConfig’s features and how to use the tool, see the WeConfig User
Guide[64].
ˆ Web: The WeOS Web interface provides management of essential features.
The Web interface should satisfy the needs of all common use cases.
ˆ CLI: The WeOS Command Line Interface is an industry standard CLI, and
provides the most complete management support. The CLI is intended for
advanced users requiring fine grain control of the system.
In addition, WeOS provides device management via SNMP (v1/v2c/v3). A set of
standard MIBs and the WeOS private MIB are supported, as described in chapter 6.
30
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Task
Discover WeOS Devices
Set Device IP Address
Upgrade firmware
Common management tasks
All management tasks
Secure management
WeConfig
Web
CLI
X
X
X
(X)
X
X
X
(X)
X
X
X
X
X
X
SNMP
X
X
X
In the following sections the properties of the WeConfig tool, the Web Interface,
and the CLI are presented further. These sections give information about what
management tool to use for a specific need. For more information on SNMP we
refer to chapter 6.
3.1
When to use the WeConfig tool
The Westermo configuration management tool, WeConfig, is used for basic configuration and maintenance of WeOS products. It is an ideal tool to upgrade
firmware and manage configuration files (backup and restore) of a large set of
WeOS devices. With WeConfig you to scan, discover and draw maps of the WeOS
devices in your network, and you can also conduct some basic configuration of
WeOS units, such as setting the IP address and the default gateway.
For further information on WeConfig’s features and how to use the tool, see the
WeConfig User Guide[64].
3.2
When to use the Web Interface
The Web interface would be the management interface of choice for most users.
The main advantages of the Web Interface are:
ˆ Easy to use: The Web management interface provides an easy to use method
to manage the switch.
ˆ All common features: The web interface includes support for all essential
management features, and should therefore meet the needs of most users.
ˆ Secure management: The web interface can be accessed via regular HTTP
and secure HTTP (HTTPS). Secure management is also possible via the CLI
(SSHv2) and and SNMP (SNMPv3).
© 2017 Westermo Teleindustri AB
31
Westermo OS Management Guide
Version 4.21.1-0
ˆ Discover other Westermo Switches: The Web contains a discovery service
(IPconfig) similar to what WeConfig provides. (Note, you must still be able
to login to one switch in order to make use of this service.)
To use the Web interface, you must know the IP address of your switch. To find
out the switch IP address you may need to use the WeConfig tool1 , but once you
know it you can do the rest of the management via the Web interface.
The Web interface is introduced in chapter 4.
3.3
When to use the Command Line Interface (CLI)
The WeOS CLI aims to serve advanced users. Furthermore, the CLI is the only
management tool which cannot be disabled.
Below we list the situations where the CLI is the most suitable management tool.
ˆ Complete set of management features: The CLI includes all the management features available on the switch. If you cannot accomplish your task
with any of the other management tools, the CLI may provide the feature
you need.
ˆ Discover other Westermo Switches: The CLI contains a discovery service
similar to what WeConfig provides, but more rudimentary.
Note
You must still be able to login to one switch in order to make use of this
service.
ˆ Secure management: To access the CLI you must either have physical access to the switch (console port), or use the Secure Shell (SSHv2) application
to access the CLI remotely. Secure management is also possible via the Web
interface (HTTPS) and SNMP (SNMPv3).
ˆ Configuration scripting: With a CLI it is possible to develop automatic configuration scripts, e.g., using the Expect automation and testing tool. Expect
extensions exist for many common scripting languages (Ruby, Perl, Tcl).
As with the Web interface, you must know the IP address of your switch before
you can access the CLI remotely via SSH (access via the console port is possible
1 For more information about finding the IP address of your switch we refer to the Getting Started
guide in chapter 2.
32
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
without knowing the switch IP address). To find out the switch IP address you may
need to use the WeConfig tool, but once you know it you can do the rest of the
management via SSH/CLI.
The WeOS CLI is introduced in chapter 5.
© 2017 Westermo Teleindustri AB
33
Westermo OS Management Guide
Version 4.21.1-0
Chapter 4
Management via Web Interface
WeOS supports device management via web interface. Both HTTP and HTTPS1
are supported. The design is optimised for style sheet and JavaScript2 capable
web browsers. In addition, the design allows users to access the web interface
and all settings without a style sheet and JavaScript capable browser, but then
with less guidance and support from the user interface.
When using the Web Management Tool you have to be aware of the following:
ˆ Only one user can be logged in at a time (see section 4.2 for more information).
ˆ You are automatically logged out after ten (10) minutes of inactivity (see
section 4.2 for more information).
ˆ When you click Apply on a page, the settings on that page are immediately
activated.
ˆ When you click Apply on a page, all settings are stored in the startup configuration and therefore survive a reboot (see chapter 7 for more information).
Section 4.2 explains how to access the Web Management Tool and section 4.3
describes the web menu hierarchy. In section 4.3 the system overview web pages
are presented. Other pages and settings are described per topic in chapter 8 and
following chapters.
1 For
HTTPS server authentication, a self-signed certificate is used as of WeOS v4.21.1.
is a trademark of Oracle Corporation.
2 JavaScript
34
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
4.1
Document Conventions
Specific conventions for the web part of this document.
Button Text
Buttons are indicated by use of
bold type-writer style.
Menu path:
Top Item ⇒ Sub Item
For each page the menu path to
the page is described with this
syntax. It means: First click the
Top Item menu item and in the
sub-menu revealed, click the Sub
Item menu item. See also section 4.3.
Menu path:
Top Item ⇒ Sub Item ⇒ Button Text
Top Item ⇒ Sub Item ⇒
(ctx)
© 2017 Westermo Teleindustri AB
This is an extension to the Menu
path: Top Item ⇒ Sub Item version described above. It tells you
to click a button with the text Button Text on the page navigated to
by Top Item ⇒ Sub Item.
The button may be an icon. In this
case the icon is shown. Additionally in parenthesis a sub-context
(ctx) may be described which will
identify a context on the page,
normally identified by its header.
35
Westermo OS Management Guide
Version 4.21.1-0
4.2
Logging in
To access the switch through the web interface, enter the appropriate URL (e.g.,
the factory default IP-address http://192.168.2.200) in the address field of your
web-browser. You will then be presented to the login page where you fill in the
username and password, see figure 4.1.
Figure 4.1: Web login window
Currently there is only a single user account defined, the administrator user account. Note that it is the same user account used for login in CLI. Factory default
user account and password are as follows :
ˆ Login: admin
ˆ Password: westermo
Your web session will last for ten (10) minutes after your latest ”web action”.
Clicking a link or button at least every 10 minutes will let you keep the session
36
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
forever. The same goes for pages with an automatic refresh option, given that a
refresh interval of 10 minutes or shorter is selected.
Only one user at a time can be logged into the switch Web Management Tool. If a
new user tries to log in the currently logged in user will automatically be logged
out.
© 2017 Westermo Teleindustri AB
37
Westermo OS Management Guide
Version 4.21.1-0
4.3
Navigation
After logging in you will be redirected to the start page, see fig. 4.2. In the page
header you find the menus used to navigate between different tasks. The menu
consists of two rows, the top-menu row, and the sub-menu. For some items you
will be presented to a third level sub-menu below the second level sub-menu. Its
function is analogously to the second level sub-menu .
To navigate in the menu, click on the top-menu to reveal the associated submenu. Then click on the desired sub-menu item. For example, fig. 4.2 shows the
selection of top-menu Status and sub-menu Summary (i.e., Status ⇒ Summary).
Figure 4.2: Unit Summary - the first page after logging in.
The top-level menu structure is described below:
ˆ Status - This is where you find status information of the running system (port
status, protocol status, etc.)
ˆ Configuration - This is where you configure the unit
ˆ Maintenance - This is where you do firmware upgrades, configuration file
backups, view log files, manage port monitoring, etc.
38
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
ˆ Tools - Here you find various tools for trouble-shooting and other purposes
(e.g., ”ping”).
Pages where you can change settings generally contains an Apply and a Cancel
button, as shown in fig. 4.3. The semantics of the Apply and Cancel buttons are
provided below:
Apply
Cancel
Applies the changes on the current page. Changes are applied
immediately (i.e., no reboot needed), and are also stored in
the startup configuration.
Discards changes and either returns to an overview page for
the context, or reloads current page and thus shows the current settings.
Figure 4.3: Sample web page containing Apply and Cancel buttons.
Pages with lists of ports may have additional information to display, e.g. if the
port is included in a port aggregate or bonded with PAF. This is indicated by
the background behind the port label is highlighted as shown in fig. 4.4. When
hovering a highlighted port the additional information is displayed in a pop-up.
Inside a drop-down menu, the ports are also highlighted, but no pop-ups are
presented.
© 2017 Westermo Teleindustri AB
39
Westermo OS Management Guide
Version 4.21.1-0
Figure 4.4: Sample web page with port information pop-up.
40
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
4.4
System Overview
There are two levels of system information, summary and detailed.
4.4.1
System Overview - Summary
Menu path: Status ⇒ Summary
Fig. 4.5 shows the first page you will be presented to after logging into the switch.
It provides a quick overview of the system, including a list of current alarms.
Figure 4.5: The basic system overview page.
Hostname
Location
ADSL/VDSL Status
An arbitrary name to identify this unit.
An arbitrary description to identify where the unit is
located.
Current ADSL/VDSL connection status. Displays negotiation status, IP-address, up/down speed and DSL
uptime.
Continued on next page
© 2017 Westermo Teleindustri AB
41
Westermo OS Management Guide
Version 4.21.1-0
Uptime
Date
Running Services
Alarms
Interfaces
42
Continued from previous page
The time passed since last reboot of the unit.
The current date and time. System time is configured manually or set by using a NTP-server.
A list of services currently running on the unit.
Currently active port and FRNT alarms.
Link alarms are only shown for ports where link
alarm is enabled and when the link is down. FRNT
alarms are only shown for FRNT ports with link down.
Displays the interfaces and their primary addresses.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
4.4.2
System Overview - Detailed
Menu path: Status ⇒ System
To get more information about the switch you go to the detailed page shown in
fig. 4.6. This page contains more information on hardware (e.g. versions, article
number, etc.) and system status (e.g. memory usage and CPU load).
Hostname
Location
Contact
Uptime
Base MAC Address
System Default
Gateway Address
Article Number
Main Firmware
Version
Build Details
Backup Firmware
Version
Main FPGA Version
Boot Loader Version
Serial Number
Product
Model
Type
Article No.
Batch ID
Revision
Enabled Redundancy Protocol(s)
VLANs With IGMP
An arbitrary name to identify this unit.
An arbitrary description to identify unit location.
An arbitrary description to identify a contact person who has more information about management
of the unit and the network.
The time passed since last reboot of the unit.
The base MAC address defines the starting point of
the MAC address range used within the unit. This is
a unique number assigned to each unit.
The operational default gateway for all VLANs on the
unit. Either retrieved dynamically or set statically.
The article number for the unit.
The version number of the main firmware.
The build string of the currently running firmware.
The version number of the backup firmware.
The version number of the FPGA software.
The version number of the boot loader software.
The units serial number.
The product name.
The product model.
Description for the card in the specified slot.
The article number of the card in the specified slot.
The batch identification of the card in the specified
slot.
The revision of the card in the specified slot.
A list of the redundancy protocols currently enabled
on the unit.
A list of VLANs on which IGMP is enabled.
Continued on next page
© 2017 Westermo Teleindustri AB
43
Westermo OS Management Guide
Version 4.21.1-0
SNMP
Alarms
44
Continued from previous page
Shows if SNMP support is enable or disabled.
Currently active port and FRNT alarms.
Link alarms are only shown for ports where link
alarm is enabled and link is down.FRNT alarms are
only shown for FRNT ports where link alarm is enabled and when the link is down.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Figure 4.6: Detailed system overview page.
© 2017 Westermo Teleindustri AB
45
Westermo OS Management Guide
Version 4.21.1-0
4.4.3
System Environment
Menu path: Status ⇒ Environment
To get more information about the system environment variables you go to the
environment page.
Temperature
Load
Average
Memory
Usage (%)
46
Shows system temperature i Celsius(C).
The load average is a standard Linux way of measuring system
load.
A snapshot of RAM (Random Access Memory) usage as percentage of total RAM.
Continued on next page
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
DDM/DOM
SFPs
1
Continued from previous page
Shows DDM/DOM diagnostics for each SFP.
The black bar for each graph represents the first value which
was read after boot up, and the blue bar is current value. The
DDM/DOM information will be polled for each SFP every twelfth
hour. Each graph will then be updated and can consist of up to
20 polled entries. By positioning the mouse over a graph, the
user will be presented with startup, max and min value. Please
note that each graph shows trend over time and not the absolute value, graphs for different SFP should not be compared.
1 DDM/DOM diagnostic information is only available for Westermo DDM SFPs, see the SFP
Transceiver Datasheet of your WeOS product (www.westermo.com).
© 2017 Westermo Teleindustri AB
47
Westermo OS Management Guide
Version 4.21.1-0
Chapter 5
Management via Command
Line Interface (CLI)
This chapter introduces the command line interface (CLI) tool. Switches running
WeOS include a CLI similar to what is provided by other major vendors of network
equipment. The CLI provides a more complete set of management features than
the Web interface, the WeConfig tool or SNMP. Thus, when advanced management operations are required, the CLI is the management interface of choice.
The CLI can be accessed via the console port, or remotely via secure shell (SSHv2)
and Telnet1 .
Section 5.1 introduces the CLI hierarchy and its various contexts. Section 5.2
explains how to access the CLI interface, and section 5.3 provides general information on how to use the CLI.
The last section (section 5.4) presents CLI commands available in all CLI contexts as well as their syntax. Other CLI commands are described per topic in the
chapters to follow.
5.1
Overview of the WeOS CLI hierarchy
The WeOS CLI is organised in a hierarchical structure. For management purposes,
the use of a hierarchical structure limits the available commands to those relevant for a certain topic. This in turn simplifies switch operation.
1 Telnet
48
server is by default disabled, see also section 8.3.24.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Administrator Execution Context
Global Configuration Context
Specific Execution Contexts
(RMON, Debug, ...)
Specific Configuration Contexts
Figure 5.1: CLI hierarchy
Fig. 5.1 shows an overview of the CLI hierarchy. When the user logs in as ”admin”
the user will enter the CLI with ”administrator” privileges in Admin Exec context.
(In addition to the ”admin” user, future versions of WeOS are likely to support a
”guest” account with limited privileges.)
Admin Exec context In Admin Exec context the user can execute a set of general monitoring and diagnostic functions, and also manage configuration
files and firmware versions. From Admin Exec context the user can enter a
set of specific execution contexts, e.g., to view RMON statistics.
Global Configuration context From the Admin Exec context the user can enter
the Global Configuration context. In Global Configuration the user can configure device parameters of global significance, such as hostname and location of the device. From Global Configuration the user can reach contexts
specific to certain protocols or device entities such as port, vlan, interface,
and FRNT contexts.
A simple example on CLI usage is given below. There you can see how the CLI
prompt changes to match the current context.
Example
example:/#> configure
example:/config/#> vlan 100
example:/config/vlan-100/#> untagged 1,2
example:/config/vlan-100/#> end
example:/config/#> end
example:/#>
© 2017 Westermo Teleindustri AB
49
Westermo OS Management Guide
Version 4.21.1-0
5.2
Accessing the command line interface
To login via the console port you need the username and password. Currently
there is only a single user account defined, the administrator user account. Factory default account and password:
ˆ Login: admin
ˆ Password: westermo
The same account is used for management via CLI and Web (see section 4). To
reset the administrator password to the default setting, see chapter 7.
5.2.1
Accessing CLI via console port
For WeOS switches equipped with a console port, that port can be used to access
the CLI. (For information on which WeOS devices that have a console port, see
section 1.5.1).
Console cable
See the User Guide of your specific product (section 1.5) for information on
what Diagnostic Cable to use when connecting to the console port of your
specific product.
Recommended Terminal Emulation programs:
ˆ Win32: PuTTY, http://www.chiark.greenend.org.uk/~sgtatham/putty/
ˆ UNIX: There are different terminal emulation programs for different Unix
dialects. On Linux minicom is recommended.
The following console port settings are used:
Data rate
Data bits
Stop bits
Parity
Flow control
115200 bits/s
8
1
None
None
The example in below shows how to login via the console port using the PuTTY application. Once you have installed and started PuTTY, configure the appropriate
50
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Serial settings.
Hint
In this example, the switch is accessible via the logical port ”COM3”, but
the USB/serial adapter may be mapped to a different COM port on your PC.
Please check ”Ports (COM and LPT)” in the Windows ”Device Manager” to
get information on what COM port to specify.
When the appropriate serial settings have been configured, select the ”Session”
view. Select Serial as Connection type as shown in the figure below.
To start the serial connection, press the Open button. The figure below shows
the console prompt when logging in to the CLI via the console on a unit named
example.
© 2017 Westermo Teleindustri AB
51
Westermo OS Management Guide
Version 4.21.1-0
example login: admin
Password:
.--.--.--.-----.-----.------.-----.-.--.--------.-----.
_| -__|
_| . . | _ | http://www.westermo.com
| | | | -__|__ --|_
\__/\__/|_____._____| |__| |_____|__| |__|__|__|_____|
[email protected]
Robust Industrial Data Communications -- Made Easy
\\/ Westermo WeOS v4.15.0 4.15.0 -- Jun 16 19:10 CEST 2014
Type: ’help’ for help with commands, ’exit’ to logout or leave a context.
example:/#>
5.2.2
Accessing the CLI via SSH or Telnet
To gain access to the CLI via SSH you need a SSH client, the switch IP address,
and the account information (username and password).
Recommended SSH Clients:
ˆ Win32: PuTTY, http://www.chiark.greenend.org.uk/~sgtatham/putty/
ˆ UNIX OpenSSH, http://www.openssh.com
The switch IP address can be found using the WeConfig tool, see the WeConfig
User Guide[64] (additional methods are listed in section 7.1.3).
The following example illustrates how to login to the switch using PuTTY from
a Windows based host system as user admin. In this example, the switch is a
WeOS switch with IP address 192.168.2.200 (the factory default IP address). See
section 5.2 for information about user accounts and passwords.
In the PuTTY session view, select SSH as Connection type, and enter the IP address of the switch (here 192.168.2.200).
52
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Click the Open button to start the SSH session. You will be presented to a login
prompt (see below), and enter login admin and the associated password.
example login: admin
Password:
.--.--.--.-----.-----.------.-----.-.--.--------.-----.
_| -__|
_| . . | _ | http://www.westermo.com
| | | | -__|__ --|_
\__/\__/|_____._____| |__| |_____|__| |__|__|__|_____|
[email protected]
Robust Industrial Data Communications -- Made Easy
\\/ Westermo WeOS v4.15.0 4.15.0 -- Jun 16 19:10 CEST 2014
Type: ’help’ for help with commands, ’exit’ to logout or leave a context.
example:/#>
The CLI can be accessed remotely by using a Telnet client, in the same way
as using SSH. Of security reasons, use of Telnet is discouraged and therefore
disabled by default. In order to manage the unit via Telnet, you must first:
ˆ Enable the Telnet server via the CLI, see section 8.3.24.
ˆ Enable telnet management for the desired network interface(s) via the CLI
(see section 21.6.6).
© 2017 Westermo Teleindustri AB
53
Westermo OS Management Guide
Version 4.21.1-0
5.3
5.3.1
Using the CLI
Starting out with the CLI
When first entering the CLI you end up in the Admin Exec context. In the Admin
Exec you can view system status information using various ”show” commands,
upgrade system firmware, etc., as well as other functions, which do not affect the
system configuration.
To be able to modify the switch configuration you should enter the Global Configuration context, by using the ”configure” command as shown below. From
the Global Configuration you are able to configure system parameters such as its
”hostname” or its ”date”.
Example
example:/#> configure
example:/config/#>
As described in section 5.3.2 you can reach other, specific configuration contexts
from the Global Configuration context.
Example
example:/#> configure
example:/config/#> vlan 100
example:/config/vlan-100/#> untagged 1/1,1/2
example:/config/vlan-100/#> end
example:/config/#> end
example:/#>
To get help on what commands are available in the current context, use the
”help” command (see example in fig. 5.2). First the context specific configuration commands are shown, followed by the commands to show the current
configuration settings. At the end, commands available in all contexts are shown
(see also section 5.4.).
54
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/config/vlan-100/#> help
Available Commands
==============================================================================
enable
Enable, or disable this VLAN
name <ARG>
Set name of VLAN
tagged <ARG>
Set tagged ports
untagged <ARG>
Set untagged ports
channel <ARG>
Set VLAN channel interface
priority <ARG>
Set VLAN priority, overrides port priority
igmp
Enable, or disable IGMP Snooping
show
show
show
show
show
show
show
enable
name
tagged
untagged
channel
priority
igmp
Show
Show
Show
Show
Show
Show
Show
if VLAN is active or not
name of VLAN
tagged ports
untagged ports
VLAN channel interface
VLAN priority setting
IGMP Snooping status
no <ARG>
Prefix, used to disable services or settings.
do
Shortcut to EXEC mode, e.g. do ping <IP>.
end
Save settings and return to previous mode.
leave
Save settings and return to EXEC mode.
abort
Cancel all changes and leave this mode.
show <ARG>
Show summary, or status.
repeat <ARG>
Repeat next command every second, until Ctrl-C
help <ARG>
This help text.
tutorial
Brief introduction to the CLI
==============================================================================
<ARG> - Command takes argument(s), see help <command> for further information.
Short forms of commands are possible, see the tutorial for more help.
example:/config/vlan-100/#>
Figure 5.2: Use of the ”help” command to list available commands (here in the
VLAN context).
The ”help” command can also be used to get information on a specific command
as shown below.
Example
example:/config/vlan-100/#> help igmp
Syntax:
[no] igmp
Description:
Enable, or disable IGMP Snooping
==============================================================================
The [no] keyword is when you want to disable a service or remove a property.
example:/config/vlan-100/#>
© 2017 Westermo Teleindustri AB
55
Westermo OS Management Guide
Version 4.21.1-0
The CLI supports basic TAB-completion, which can come in handy when you do
not know the exact command name, e.g., writing ”fi[TAB]” within the IP context
will expand to ”firewall”.
TAB-completion is only able to expand the full command when there is no ambiguity. Otherwise the available alternatives will be listed.
Example
example:/#> d[TAB]
do
debug
date
example:/#> d
dir
delete
Furthermore, when there is no ambiguity it is possible to use an abbreviation of
a command instead of the full command (i.e., without using TAB-completion).
Example
example:/#> con
example:/config/#>
5.3.2
Entering and leaving CLI contexts
Fig. 5.3 gives a general overview of how to enter and leave the various context
in the CLI hierarchy. The commands to move between contexts are further discussed in the text below.
To enter Global Configuration context from Admin Exec context, the ”configure”
command is used. From Global Configuration context one can reach several specific configuration contexts, and the command to enter them is context specific,
e.g.,:
vlan <VID>
port <PORT>
interface <IFNAME>
Manage VLAN settings for VLAN with given VID.
Manage port settings for port with given PORT identifier.
Manage settings for the given network interface.
By entering the Global Configuration context the user is able to interactively
change the device configuration, however, configuration changes will not take
effect until the user leaves the configuration contexts and returns to the Admin
Exec context via the ”end” or ”leave” commands.
When the user returns to Admin Exec context, the running-configuration of the
switch will be updated. To make the configuration changes permanent the running-
56
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Login prompt (console/SSH)
logout
username & password
end/logout
Administrator Execution Context
leave
configure
end
rmon
end
Port Configuration
Context
vlan <...>
end
VLAN Configuration
Context
monitor
RMON
Context
Global Configuration Context
port <...>
end
ip
end
Port Monitoring
Context
end
General IP
Config. Context
firewall
end
Firewall/NAT
Config. Context
Figure 5.3: Moving between CLI contexts. Only a subset of the available contexts
is shown. Although not shown, the leave and logout commands can be used from
all contexts.
configuration should be saved to the startup-configuration using the ”copy” command, see also chapter 7.
It is also possible to leave the configuration contexts without updating the runningconfiguration. The commands to leave a context are listed below. More information on these and other general CLI commands can be found in section 5.4.
end
leave
Ctrl-Z
Confirms configuration changes conducted in this context and
returns to the context immediately above. If issued within the
Global Configuration context, the user returns to the Admin
Exec context and the running-configuration is updated.
Confirms configuration changes made and returns to Admin
Exec context. The running-configuration is updated.
An alias for leave. Ends your configuration session and returns
to Admin Exec context.
Continued on next page
© 2017 Westermo Teleindustri AB
57
Westermo OS Management Guide
Version 4.21.1-0
abort
exit
Ctrl-D
logout
5.3.3
Continued from previous page
Discards configuration changes conducted in this context and
returns to the context immediately above. If issued within the
Global Configuration context, the user returns to the Admin
Exec context without updating the running-configuration. If
issued in Admin Exec context it works the same as logout.
An alias for abort.
An alias for abort. Blocked if any text is already input on the
command line.
Log out from the CLI. If conducted from within any of the configuration contexts, all configuration changes are discarded
(i.e., the running configuration is not updated).
CLI command conventions
This section describes the CLI command conventions used within this guide. The
syntax for a sample set of CLI commands is shown below:
ˆ [no] default-gw <ADDRESS>
ˆ igmp-interval <12|30|70|150>
ˆ show iface [IFNAMELIST]
Convention
command syntax
”command syntax”
UPPERCASE
lowercase
|
< >
[ ]
58
Description
Command syntax is generally written in typewriter
style (fixed width)
Commands described in running text use bold typewriter style enclosed by quotation marks.
A variable parameter. Enter value according to the description that follows.
A keyword parameter. Enter value according to the
given syntax.
Vertical bar. Used to separate alternative (mutually exclusive) parameters.
Angle brackets. Encloses a mandatory parameter.
Squared brackets. Encloses an optional parameter.
Continued on next page
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Convention
[< >]
Continued from previous page
Description
Angle brackets within squared brackets. Encloses a
mandatory parameter within an optional choice.
© 2017 Westermo Teleindustri AB
59
Westermo OS Management Guide
Version 4.21.1-0
5.4
General CLI commands
The majority of the CLI commands are specific to a certain context, however,
there is a set of CLI commands available in all contexts. These commands are
explained further here. The ”configure” command used to enter the Global
Configuration context from the Admin Exec context, is also covered.
Command
no <COMMAND>
do
end
leave
abort
logout
repeat <COMMAND>
help [COMMAND]
tutorial
configure [terminal]
5.4.1
Section
Section 5.4.1
Section 5.4.2
Section 5.4.3
Section 5.4.4
Section 5.4.5
Section 5.4.6
Section 5.4.7
Section 5.4.8
Section 5.4.9
Section 5.4.10
Negate/disable a setting
Syntax no <COMMAND>
Context All contexts
Usage Depending on context the ”no” command disables or resets a setting to
default.
Primarily used within configuration contexts to negate or disable a configuration setting, e.g., in port context ”no flow-control” disables flow control. For some commands, ”no” is used to reset to a default value, e.g., ”no
polling-interval” (NTP client context) sets the NTP polling-interval to its
default value (600 seconds).
The ”no” command can also be used to negate/disable certain commands
outside the configuration context, e.g., to disable debugging or port monitoring.
Default values Not applicable
60
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
5.4.2
Execute (do) command from Admin Exec context
Syntax do <COMMAND>
Context All contexts
Usage Use the ”do <COMMAND>” to execute a COMMAND available in Admin Exec
context from any context.
For example, when located in Global Configuration context, the user could
run ”do show running-config” to see the running configuration, or run
”do ping 192.168.1.1” to ”ping” IP address 192.168.1.1.
Default values Not applicable
5.4.3
End context
Syntax end
Context All contexts
Usage Leave this context and return to the context immediately above. If this
command is issued within any of the configuration contexts, the command
implies that the configuration changes conducted within that context are
confirmed. If the command is issued in the Global Configuration context,
the user returns to the Admin Exec context, and the running-configuration
is updated.
Default values Not applicable
5.4.4
Leave context
Syntax leave
Context All contexts
Usage Leave this context and return to the Admin Exec context. If this command
is issued within any of the configuration contexts, the command implies
that the configuration changes conducted are confirmed, and the runningconfiguration is updated.
Default values Not applicable
© 2017 Westermo Teleindustri AB
61
Westermo OS Management Guide
Version 4.21.1-0
5.4.5
Abort context
Syntax abort
Context All contexts
Usage Leave this context and return to the context immediately above. If this
command is issued within any of the configuration contexts, the command
implies that the configuration changes conducted within that context are
discarded. If the command is issued in the Global Configuration context,
the user returns to the Admin Exec context without updating the runningconfiguration.
Default values Not applicable
5.4.6
Logout
Syntax logout
Context All contexts
Usage Logout from system. If this command is issued within any of the configuration contexts, the command implies that the configuration changes
conducted are discarded, i.e., the running-configuration is not updated.
Default values Not applicable
5.4.7
Repeat a command
Syntax repeat <COMMAND>
Context Admin Exec context
Usage Repeat COMMAND every second until Ctrl-C is pressed.
Default values Not applicable
5.4.8
On-line help
Syntax help <COMMAND>
Context All contexts
62
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Usage Show help information specific to a certain context, or a specific command.
Default values If no COMMAND is specified, help information related to the current context is shown.
5.4.9
CLI tutorial
Syntax tutorial
Context All contexts
Usage Show CLI tutorial text.
Default values Not applicable
5.4.10
Entering Global Configuration Context
When a user logs in to the CLI the user will enter the Admin Exec context. In
Admin Exec context the user can view status information and have access to
tools such as ping and traceroute, but is not able to perform any configuration.
To configure the device, the user can use the configure command to enter the
Global Configuration context.
Syntax configure [terminal]
Context Admin Exec context
Usage Enter global Configuration Context.
The optional terminal argument is a compatibility keyword, for advanced
users. It disables all safe guards (yes-or-no questions), making it possible to
paste-in configuration files into the terminal.
Pasting in configuration files can also be done with the copy command as
copy con run to copy console to running-config.
Default values Interactive mode (i.e. the ”terminal” argument does not apply
by default)
© 2017 Westermo Teleindustri AB
63
Westermo OS Management Guide
Version 4.21.1-0
Chapter 6
WeOS SNMP Support
The Simple Network Management Protocol (SNMP) provides a standardised method
to manage and monitor IP devices remotely. The WeOS SNMP agent supports
SNMP v1, v2c and v3.
6.1
Introduction and feature overview
Table 6.1 shows WeOS SNMP control features for the Web and CLI interfaces.
Further description of the SNMP support is presented in the sections 6.1.1-6.1.6.
If you are only interested in knowing how to manage SNMP features via the Web
or CLI, please visit sections 6.2 or 6.3 directly.
6.1.1
SNMP introduction
The Simple Network Management Protocol (SNMP) provides a standardised method
to manage and monitor IP devices remotely. In SNMP a manager station can manage a set of status and configuration objects via an SNMP agent on the management unit. The WeOS SNMP agent supports SNMP v1, v2c and v3.
An SNMP manager:
ˆ can send SNMP GET messages to poll status and configuration information
from an SNMP agent.
64
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Feature
General
Enable/disable SNMP
Web
CLI
General Description
X
X
SNMPv1/v2c
Read Community
Write Community
Trap Community
Trap Host
X
X
X
X
X
X
X
X
Sec. 6.1.2
SNMPv3
Read-Only SNMPv3 User
Read/Write SNMPv3 User
X
X
X
X
Sec. 6.1.4
”
Sec. 6.1.2-6.1.3
Sec. 6.1.3
”
Table 6.1: WeOS control of SNMP features.
ˆ can send SNMP SET messages to the SNMP agent to modify the device settings (or issue commands such as ’reboot’).
ˆ can get notified by an agent when specific events occur, such as link down
event, via SNMP TRAP messages.
The objects manageable via SNMP are defined in a management information base
(MIB). The WeOS MIB support aims at providing SNMP management primarily via
standard MIBs to enable easy integration with existing SNMP management tools.
In addition, WeOS includes an enterprise MIB (private MIB) to provide access to
MIB objects not available via the standard MIBs.
6.1.2
SNMP Communities
An SNMP community is a relationship between the manager and managed station. It can be seen as a (very) basic authentication and authorisation mechanism
for SNMP v1 and v2c1 . Three types of communities are supported:
ˆ Read community: The SNMP read community is used by a manager to read
SNMP MIB objects from a managed station.
Default read community: public
1 See
section 6.1.4 for secure management using SNMPv3.
© 2017 Westermo Teleindustri AB
65
Westermo OS Management Guide
Version 4.21.1-0
SNMP
Agent
SNMP
Manager
Station
Managed
Device
MIB
Internet/Intranet
SNMP
Agent
Managed
Device
MIB
Figure 6.1: Sample SNMP setup, where one manager station controls two devices
by communicating with SNMP agents running on the managed devices.
ˆ Write community: The SNMP write community can be used to write (and
read) SNMP MIB objects to (from) a managed station. Thus, if the agent has
its write community enabled, it is possible to configure the switch via SNMP.
The write community is typically named ”private”.
Default write community: Disabled
ˆ Trap community: The SNMP trap community is used when an agent wants
to send a notification to the manager (SNMP Trap). The trap community is
typically named ”public”.
Default trap community: trap
Warning
Using the well-known community strings ”public” and ”private” could pose
a serious security problem.
6.1.3
Trap Support
SNMP traps are only generated if there is at least one Trap Host (i.e., SNMP management station) defined. Up to three Trap Hosts can be defined. If two or more
Trap Hosts are configured, traps will be sent to all of them.
66
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
The WeOS SNMP trap support is integrated with the WeOS alarm handling system (see section 24.1). This means that you as an operator have fine-grained
control of which traps to send. All traps in the list below, except Coldstart and
lldpRemTablesChange, can be controlled via the alarm handling system.
ˆ Link Alarm: A trap is generated on link up or link down, given that Link
Alarm is enabled on that specific port (see sections 24.1.3 and 10.1.5).
Link Down OID: iso(1).org(3).dod(6).internet(1).snmpV2(6).snmpModules(3).
snmpMIB(1).snmpMIBObjects(1).snmpTraps(5).linkDown(3)
Link Up OID: iso(1).org(3).dod(6).internet(1).snmpV2(6).snmpModules(3).
snmpMIB(1).snmpMIBObjects(1).snmpTraps(5).linkUp(4)
Note
When a port is being reconfigured, link down and link up events are
likely to occur. If link-alarm is enabled on that port, a couple of SNMP
traps are likely to be generated as a side-effect of the port reconfiguration.
ˆ Cold Start: A trap is generated when a system comes up.
OID: iso(1).org(3).dod(6).internet(1).snmpV2(6).snmpModules(3).
snmpMIB(1).snmpMIBObjects(1).snmpTraps(5).coldStart(1)
ˆ LLDP Remote System Update: A trap is generated when a remote system
has updated.
OID: iso(1).std(0).iso8802(8802).ieee802dot1(1).ieee802dot1mibs(1).
lldpMIB(2).lldpNotifications(0).lldpNotificationPrefix(0).lldpRemTablesChange(1)
ˆ Digital-In: A trap is generated when the voltage level on the pins of a
digital-in sensor changes from high to low, or low to high.
Digital-In High OID: iso(1).org(3).dod(6).internet(1).private(4).
enterprises(1).westermo(16177).common(2).weos(1).notifications(6).
sensorNotifications(1).sensorNotificationPrefix(0).digitalInHigh(1)
Digital-In Low OID: iso(1).org(3).dod(6).internet(1).private(4).
enterprises(1).westermo(16177).common(2).weos(1).notifications(6).
sensorNotifications(1).sensorNotificationPrefix(0).digitalInLow(2)
ˆ Power Supply: A trap is generated when the voltage level on any of the
power feeds changes from high to low, or low to high.
© 2017 Westermo Teleindustri AB
67
Westermo OS Management Guide
Version 4.21.1-0
Power Supply High OID: iso(1).org(3).dod(6).internet(1).private(4).
enterprises(1).westermo(16177).common(2).weos(1).notifications(6).
sensorNotifications(1).sensorNotificationPrefix(0).powerSupplyHigh(3)
Power Supply Low OID: iso(1).org(3).dod(6).internet(1).private(4).
enterprises(1).westermo(16177).common(2).weos(1).notifications(6).
sensorNotifications(1).sensorNotificationPrefix(0).powerSupplyLow(4)
ˆ Temperature: A trap is generated when the temperature measured by a
built-in temperature sensor reaches the configured rising or falling thresholds.
Temperature High OID: iso(1).org(3).dod(6).internet(1).private(4).
enterprises(1).westermo(16177).common(2).weos(1).notifications(6).
sensorNotifications(1).sensorNotificationPrefix(0).temperatureHigh(5)
Temperature Low OID: iso(1).org(3).dod(6).internet(1).private(4).
enterprises(1).westermo(16177).common(2).weos(1).notifications(6).
sensorNotifications(1).sensorNotificationPrefix(0).temperatureLow(6)
ˆ FRNT Ring Status: A trap is generated when a unit detects a change of
FRNT ring status, i.e., ring up (OK) or ring down (Broken).
FRNT Ring Up OID: iso(1).org(3).dod(6).internet(1).private(4).
enterprises(1).westermo(16177).common(2).weos(1).notifications(6).
frntNotifications(2).frntNotificationPrefix(0).frntRingUp(1)
FRNT Ring Down OID: iso(1).org(3).dod(6).internet(1).private(4).
enterprises(1).westermo(16177).common(2).weos(1).notifications(6).
frntNotifications(2).frntNotificationPrefix(0).frntRingDown(2)
ˆ RiCo and Dual-Homing Uplink Status: A trap is generated when a unit
detects a change of RiCo Uplink Status or Dual-Homing Uplink Status, i.e.,
uplink status up, or uplink status down.
RiCo/Dual-Homing Uplink Up OID: iso(1).org(3).dod(6).internet(1).private(4).
enterprises(1).westermo(16177).common(2).weos(1).notifications(6).
riCoNotifications(8).riCoNotificationPrefix(0).riCoUplinkUp(1)
RiCo/Dual-Homing Uplink Down OID: iso(1).org(3).dod(6).internet(1).private(4).
enterprises(1).westermo(16177).common(2).weos(1).notifications(6).
riCoNotifications(8).riCoNotificationPrefix(0).riCoUplinkDown(2)
ˆ SNR-margin: On units with a SHDSL/xDSL port traps are generated when
the SNR margin falls below (or rises above) a configurable threshold.
68
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
OID: iso(1).org(3).dod(6).internet(1).mgmt(2).mib-2(1).transmission(10).
hdsl2ShdslMIB(48).hdsl2ShdslNotifications(0).hdsl2ShdslSNRMarginCrossing(2)
ˆ LFF Status: On units with SHDSL ports, a trap is generated when a unit
detects a change in the Link Fault Forward (LFF) status on a SHDSL port, i.e.,
if the remote end reports that its Ethernet port is up or down.
LFF Remote Up OID: iso(1).org(3).dod(6).internet(1).private(4).
enterprises(1).westermo(16177).common(2).weos(1).notifications(6).
lffNotifications(3).lffNotificationPrefix(0).lffRemoteUp(1)
LFF Remote Fail OID: iso(1).org(3).dod(6).internet(1).private(4).
enterprises(1).westermo(16177).common(2).weos(1).notifications(6).
lffNotifications(3).lffNotificationPrefix(0).lffRemoteFail(2)
ˆ PoE total power consumption: On units with Ethernet ports supporting
Power over Ethernet, traps are generated with the total consumed power
rises above (or falls below) a configurable threshold.
Power consumption above threshold OID: iso(1).org(3).dod(6).internet(1).
mgmt(2).mib-2(1).powerEthernetMIB(105).pethNotifications(0).
pethMainPowerUsageOnNotification(2)
Power consumption below threshold OID: iso(1).org(3).dod(6).internet(1).
mgmt(2).mib-2(1).powerEthernetMIB(105).pethNotifications(0).
pethMainPowerUsageOffNotification(3)
ˆ Address Conflicts: On units configured with alarm trigger to detect address conflicts, traps are generated as: Detected, a new conflict is detected
(with details). Cleared, a conflict is cleared (with details). Warning, at least
one conflict. OK, no conflict is detected.
Address Conflict Detected OID: iso(1).org(3).dod(6).internet(1).private(4).
enterprises(1).westermo(16177).common(2).weos(1).notifications(6).
addressConflictNotifications(6)).addressConflictNotificationsPrefix(0).
AddressConflictDetected(1)
Address Conflict Cleared OID: iso(1).org(3).dod(6).internet(1).private(4).
enterprises(1).westermo(16177).common(2).weos(1).notifications(6).
addressConflictNotifications(6)).addressConflictNotificationsPrefix(0).
AddressConflictCleared(2)
Address Conflict OK OID: iso(1).org(3).dod(6).internet(1).private(4).
enterprises(1).westermo(16177).common(2).weos(1).notifications(6).
© 2017 Westermo Teleindustri AB
69
Westermo OS Management Guide
Version 4.21.1-0
addressConflictNotifications(6)).addressConflictNotificationsPrefix(0).
AddressConflictOK(3)
Address Conflict Warning OID: iso(1).org(3).dod(6).internet(1).private(4).
enterprises(1).westermo(16177).common(2).weos(1).notifications(6).
addressConflictNotifications(6)).addressConflictNotificationsPrefix(0).
AddressConflictWarning(4)
ˆ Summary Alarm Status: The summary alarm status (summaryAlarmStatus) follows the status of the ON LED:
– when the ON LED turns red, the summaryAlarmStatus has value Warning (1).
– when the ON LED turns green, the summaryAlarmStatus has value OK
(2).
It is possible to get SNMP traps when the summary Alarm Status changes
state (see section 24.3.16 for information of how to enable summary alarm
traps). When enabled, a summaryAlarmOK trap is sent when the ON LED
turns green, and a summaryAlarmWarning trap is sent when it turns red.
Summary Alarm OK OID: iso(1).org(3).dod(6).internet(1).private(4).
enterprises(1).westermo(16177).common(2).weos(1).notifications(6).
genericNotifications(4).genericNotificationPrefix(0).summaryAlarmOK(1)
Summary Alarm Warning OID: iso(1).org(3).dod(6).internet(1).private(4).
enterprises(1).westermo(16177).common(2).weos(1).notifications(6).
genericNotifications(4).genericNotificationPrefix(0).summaryAlarmWarning(2)
The summary alarm status can be read at the following OID:
iso(1).org(3).dod(6).internet(1).private(4).enterprises(1).westermo(16177).
common(2).weos(1).system(5).eventSystem(2).summaryAlarmStatus(1)
6.1.4
Secure management using SNMPv3
To manage a unit securely via SNMP, SNMPv3 should be used. SNMPv3 provides
privacy and integrity (per packet authentication) to the SNMP messages.
SNMPv3 introduces the notion of a SNMPv3 user, as opposed to the community
concept used in SNMPv1/v2c. The following parameters can be configured for an
SNMPv3 user.
70
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
ˆ Read-Only or Read-Write access: Defines whether the user should have read
access to the SNMP variables, or be able to read and modify them.
ˆ Security Mode: Three security modes are available:
– noAuthNoPriv: No security (i.e., neither authentication, nor encryption)
– authNoPriv: Authentication, but no privacy.
– authPriv: Authentication and Encryption
Note
As of WeOS v4.21.1, the WeOS SNMP agent accepts SNMP requests of
security level authNoPriv also for SNMPv3 users created at level authPriv. A stricter behaviour, where requests of security level authNoPriv
will be discarded for SNMPv3 users created at level authPriv, may be
implemented in future releases of WeOS.
ˆ Encryption protocol: WeOS offers SNMPv3 data encryption using DES and
AES-128.
ˆ Authentication protocol: WeOS offers SNMPv3 data integrity using using
MD5 and SHA1.
ˆ Scope: A user can be restrained to only access a part of the MIB tree supported by the unit.
The encryption and authentication passwords are strings of 8-16 characters.
ASCII characters 33-126 except ’#’ (ASCII 35) are allowed.
A maximum of 8 SNMPv3 users can be defined, each with their own parameter
set.
6.1.4.1
SNMPv3 example
This example illustrates the configuration of an SNMPv3 user on the a WeOS
switch. The user alice is grated read-only access to the full MIB tree. Security
level authNoPriv is used where SHA1 is used as authentication protocol.
© 2017 Westermo Teleindustri AB
71
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/#> configure
example:/config/#> snmp-server
example:/config/snmp/#> rouser alice auth sha1 alicepwd
example:/config/snmp/#> leave
example:/#> cp running start
Section 6.1.6 lists recommended SNMP management software. Those tools have
graphical user interfaces and should be straight forward to use. For a simple
test you could also use the (Unix) Net-SNMP ”snmpwalk” command. (Here it
is assumed that the switch is accessible on IP address 192.168.2.200 and the
”walk” is limited to the mib-2 system’s group).
Example
mypc:~$ snmpwalk -v3 -u alice -l authNoPriv -a SHA -A alicepwd 192.168.2.200 system
SNMPv2-MIB::sysDescr.0 = STRING: Westermo RedFox Industrial, primary: v4.4.0, backup: v4.
bootloader: v2.01, fpga: v20080626
SNMPv2-MIB::sysObjectID.0 = OID: SNMPv2-SMI::enterprises.16177
DISMAN-EVENT-MIB::sysUpTimeInstance = Timeticks: (94018) 0:15:40.18
SNMPv2-MIB::sysContact.0 = STRING:
SNMPv2-MIB::sysName.0 = STRING: redfox
SNMPv2-MIB::sysLocation.0 = STRING:
SNMPv2-MIB::sysServices.0 = INTEGER: 79
SNMPv2-MIB::sysORLastChange.0 = Timeticks: (0) 0:00:00.00
mypc:~$
6.1.5
6.1.5.1
Supported MIBs
Standard MIBs
As of WeOS v4.21.1 the following standard MIBs are supported2 :
ˆ RFC1213 MIB-2: The original MIB-2 standard MIB.
ˆ RFC2863 Interface MIB: The ifXTable of the IF-MIB is supported.
ˆ RFC2819 RMON MIB: RMON Ethernet statistics (etherStatsTable) is supported.
ˆ RFC4188 Bridge MIB
ˆ RFC4318 RSTP MIB
2 Indications of level of support for each MIB is shown in the list of supported MIBs. For some
MIBs, you find more detailed MIB conformance information in the WeOS release zip-archive.
72
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
ˆ RFC4363 Q-BRIDGE MIB: The dot1qVlan group and dot1qVlanStaticTable are
supported, enabling support for static VLAN configuration.
ˆ RFC4836 MAU MIB: The dot3IfMauBasicGroup and dot3IfMauAutoNegGroup
of the MAU MIB are supported.
ˆ RFC3635 Ether-like Interface MIB: The dot3StatsTable is supported, enabling
monitoring of various error counters for Ethernet ports.
ˆ RFC4133 Entity MIB: The entityPhysical group of the Entity MIB is supported.
It can be used to read unit serial number, firmware version, etc.
ˆ RFC3433 Entity Sensor MIB: The Entity Sensor MIB can be used to monitor
the status of unit sensors for temperature, power supply, and ”digital-in”,
etc.
ˆ RFC 4319 HDSL2/SHDSL MIB: On products with SHDSL ports, the
hdsl2ShdslSpanConfTable, hdsl2ShdslSpanStatusTable,
hdsl2ShdslInventoryTable and hdsl2ShdslSpanConfProfileTable are supported
(read-only).
ˆ RFC 3621 Power Ethernet MIB: The PoE MIB is supported on products with
PoE ports.
ˆ IEEE 802.1AB LLDP MIB: Partial support for lldpConfigGroup, lldpConfigTxGroup, lldpStatsRxGroup, lldpStatsTxGroup, lldpLocSysGroup, and lldpRemSysGroup of IEEE 802.1AB-2005 LLDP MIB.
ˆ IEEE 802.1AX LAG MIB: Support for dot3adAggPortListGroup, and partial
support for dot3adAggGroup and dot3adAggPortGroup of IEEE 802.1AX-2014
Link Aggregation MIB (LAG MIB).
ˆ RFC2787 VRRPv2 MIB: The vrrpOperations group is supported (read-only).
ˆ RFC6527 VRRPv3 MIB: The vrrpv3Operations group is supported (read-only).
ˆ IEC 61375-2-5 TTDP MIB: The TTDP MIB is supported on RFR-212-FB (”Corazon” platform), with some adjustments for errors in the MIB file.
ˆ UCD SNMP MIB: Applicable objects in the memory branch of the ucdavis
module are supported. This MIB can be used to monitor system memory
usage, and is maintained by the Net-SNMP Project3 .
3 http://www.net-snmp.org
© 2017 Westermo Teleindustri AB
73
Westermo OS Management Guide
Version 4.21.1-0
6.1.5.2
Private MIB
To use the WeOS private MIB, two Westermo specific MIB files should be loaded
into your SNMP management software (see section 6.1.6 for information on recommended management software):
ˆ WESTERMO-MIB: Defines the top level objects of the Westermo Private MIB
name space.
ˆ WESTERMO-WEOS-MIB: Defines the WeOS branch of the Westermo Private
MIB.
6.1.6
Recommended Management Software
The following SNMP managers are recommended:
ˆ OidView from ByteSphere4 .
ˆ MG-SOFT MIB Browser Pro. from MG-SOFT5 .
ˆ SNMPc from Castlerock Computing6 .
6.1.7
WeOS ifIndex allocation
SNMP ifIndex is used in various SNMP MIBs as reference to specific interfaces, be
it Ethernet ports, VLAN interfaces or higher layer network interfaces. In particular, the ifTable and ifXTable of MIB-2 and IF-MIB make use of the ifIndex when
listing the available interfaces on the WeOS unit.
Note
As a general recommendation, it is strongly discouraged to put any semantic
interpretation on the ifIndex allocated for interfaces on a WeOS unit. Management systems should not be hard-coded expecting that a specific interface is always assigned a certain ifIndex.
This said, there may be situations where knowledge about WeOS algorithm to
allocate ifIndexes can be valuable. Sections 6.1.7.1 and 6.1.7.2 gives a brief
4 http://www.oidview.com/oidview.html. OidView is a trademark of BYTESPHERE TECHNOLOGIES LLC.
5 http://www.mg-soft.com/mgMibBrowserPE.html.
6 http://www.castlerock.com/. SNMPc is a trademark of Castlerock Computing.
74
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
description of default ifIndex allocation and mechanisms to control the ifIndex
allocation in WeOS v4.21.1.
Note
The ifIndex allocation behaviour may change in future releases of WeOS.
6.1.7.1
Default behaviour
In WeOS interfaces are classified as physical or virtual interfaces:
ˆ Physical interfaces: Physical interfaces are Ethernet ports, DSL ports, and
link aggregates. As of WeOS v4.21.1, the default behaviour of the physical
interfaces are as follows:
– Ethernet and DSL ports: Ethernet/DSL ports have fixed allocation ifIndexes
starting from 4096. Indexes are allocated in ascending order (4096,
4097, . . . , ”4096+NB_OF_PORTS-1”).
– Link Aggregates: Link aggregates are assigned ifIndexes dynamically,
starting from ”4096+NB_OF_PORTS”.
ˆ Virtual interfaces: Examples of virtual interfaces are VLAN (layer-2) interfaces, and various types of network and tunnel interfaces (vlan, ppp, gre,
ssl, etc.), as well as the loop-back interface (lo). As of WeOS v4.21.1, the
default behaviour of the virtual interfaces are as follows:
– Loopback interface: The loop-back interface is assigned ifindex ”1”
(fixed).
– Other virtual interfaces: Other virtual interfaces are assigned ifIndexes
dynamically within the range 2-4095 (random order). If an interface,
e.g., ”VLAN 2”, is removed and later re-created, that interface will typically get a different ifIndex the second time.
6.1.7.2
Controlling ifIndex allocation
It is possible to modify the ifIndex allocation by changing the start ifIndex for
physical and virtual interfaces. Below is an example where the ifIndex start for
physical interfaces is set to ”1” on a 10-port WeOS product.
© 2017 Westermo Teleindustri AB
75
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/config/snmp/#> show ifindex-start
Physical 4096, Virtual 1
example:/config/snmp/#> ifindex-start physical 1
Virtual ifIndex was automatically adjusted to 19
example:/config/snmp/#> show ifindex-start
Physical 1, Virtual 19
example:/config/snmp/#>
In this example, the ifIndex start for virtual interfaces automatically adjusted to
avoid overlap with the physical interfaces.
76
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
6.2
Managing SNMP via the web interface
Menu path: Configuration ⇒ SNMP
On the SNMP configuration page you will be presented to the current settings for
SNMP on your switch, see below. You may change the settings by editing the
page.
On the lower part of the page there is a list of SNMP v3 Users.
Enabled
Read Community
Write Community
Trap Community
Check the box to enable SNMP. If you have a
JavaScript enabled browser the other settings
will not be displayed unless you check this box.
A community identifier for read access. Leave
blank to disable read community.
A community identifier for read/write access.
Leave blank to disable write community.
A community identifier for traps. Defaults to
community identifier trap.
Continued on next page
© 2017 Westermo Teleindustri AB
77
Westermo OS Management Guide
Version 4.21.1-0
Trap Host Address 1/2/3
ifIndex Start
78
Continued from previous page
IP address of SNMP trap management station.
None, one , two or three addresses may be
filled in. Leave all blank to disable SNMP traps.
The start values for ifIndex (Physical and virtual). Select override to change default values
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
6.2.1
Manage SNMP v3 Users
On the lower part of the SNMP configuration page you will be presented to the
list of currently configured SNMP v3 users.
Figure 6.2: Listing of SNMP v3 users.
Type
Access rights for the user.
rwuser User has read and write access.
rouser User has read access only.
Name
A text string defining the user. Max 32 characters. Valid characters are ASCII 33-126 except ’#’ (ASCII 35).
Achieve message integrity protection by specifying MD5 or
SHA1 message authentication.
The authentication password is a string of 8-16 characters.
ASCII characters 33-126 except ’#’ (ASCII 35) are allowed.
Achieve message privacy by specifying DES or AES128 message encryption.
The encryption password is a string of 8-16 characters. ASCII
characters 33-126 except ’#’ (ASCII 35) are allowed.
Limit access to a certain branch of the supported MIB. Defaults
to the whole tree (’1.’)
Auth
Auth.
Passphrase
Crypto
Crypto
Passphrase
OID Tree
Edit
Delete
New User
Click this icon to edit the SNMP v3 user in that table row.
Click this icon to remove a the SNMP v3 user in that table row.
Click on this button to create a new SNMP v3 user.
When clicking the New User button, the SNMP v3 user edit page will be displayed.
© 2017 Westermo Teleindustri AB
79
Westermo OS Management Guide
Version 4.21.1-0
Figure 6.3: New SNMP v3 user.
See table above for description of fields.
80
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
6.3
Manage SNMP Settings via the CLI
Command
SNMP Server Configuration
[no] snmp-server
[no] rocommunity <COMMUNITY>
[no] rwcommunity <COMMUNITY>
[no] trapcommunity <COMMUNITY>
[no] host <IPADDR>
[no] ifindex-start [physical <VALUE>]
[virtual <VALUE>]
[no] rouser <USERNAME>
[auth <md5|sha1> <PASSPHRASE>
[crypto <des|aes128> <PASSPHRASE>]]
[OIDTREE]
[no] rwuser <USERNAME>
[auth <md5|sha1> <PASSPHRASE>
[crypto <des|aes128> <PASSPHRASE>]]
[OIDTREE]
SNMP Server Status
show snmp-server
6.3.1
Default
Section
Enabled
public
Disabled
trap
Disabled
Section
Section
Section
Section
Section
Section
Disabled
Section 6.3.7
Disabled
Section 6.3.8
6.3.1
6.3.2
6.3.3
6.3.4
6.3.5
6.3.6
Section 6.3.9
Manage SNMP Server
Syntax [no] snmp-server
Context Global Configuration context.
Usage Enter SNMP Server Configuration context. If the SNMP server is disabled,
it will be enabled when issuing the ”snmp-server” command. Use ”no
snmp-server” to disable the SNMP server.
Use ”show snmp-server” to show all SNMP server settings. (Also available
as ”show” command within the snmp-server context.)
Default values Enabled.
© 2017 Westermo Teleindustri AB
81
Westermo OS Management Guide
Version 4.21.1-0
6.3.2
Manage SNMP Read Community
Syntax [no] rocommunity <COMMUNITY_STRING>
Context SNMP Server Configuration context.
Usage Configure the SNMP Read Community string. Use ”no rocommunity” to
disable the SNMP Read Community.
Use ”show rocommunity” to show the SNMP Read Community setting.
Default values rocommunity public
6.3.3
Manage SNMP Write Community
Syntax [no] rwcommunity <COMMUNITY_STRING>
Context SNMP Server Configuration context.
Usage Configure the SNMP Write Community string. Use ”no rwcommunity” to
disable the SNMP Read Community.
Use ”show rwcommunity” to show the SNMP Write Community setting.
Default values Disabled.
6.3.4
Manage SNMP Trap Community
Syntax [no] trapcommunity <COMMUNITY_STRING>
Context SNMP Server Configuration context.
Usage Configure the SNMP Trap Community string. ”no trapcommunity” will
reset the trap community to the default string (”trapcommunity trap”).
Use ”show trapcommunity” to show the SNMP Trap Community setting.
Default values trap
6.3.5
Manage SNMP Trap Hosts
Syntax [no] host <IPV4ADDRESS>
82
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Context SNMP Server Configuration context.
Usage Configure a SNMP Trap Host. Up to three trap hosts can be configured (issue the ”trap-host” command multiple times with different IP addresses).
Use ”no host <IPV4ADDRESS>” to remove a trap-host and ”no host” to
remove all trap hosts.
Without any defined trap host, SNMP traps will not be sent.
Use ”show host” to show the configured SNMP Trap Hosts.
Default values Disabled.
6.3.6
Manage SNMP ifIndex-start
Syntax [no] ifindex-start [physical <VALUE>] [virtual <VALUE>]
Context SNMP Server Configuration context.
Usage Configures the ifIndex-start value used by the WeOS SNMP server. Use
”show ifindex-start” to show current settings.
Default values Physical = 4096, Virtual = 1.
6.3.7
Manage SNMPv3 Read-Only User
Syntax [no] rouser <USERNAME> [auth <md5|sha1> <PASSPHRASE> [crypto
<des|aes128> <PASSPHRASE>]] [OIDTREE]
Context SNMP Server Configuration context.
Usage Configure a SNMP read-only user.
ˆ USERNAME: A text string defining the user. Max 32 characters. Valid
characters are ASCII 33-126 except ’#’ (ASCII 35).
ˆ Authentication: Achieve message integrity protection by specifying MD5
or SHA1 message authentication. The authentication password is a string
of 8-16 characters. ASCII characters 33-126 except ’#’ (ASCII 35) are
allowed.
ˆ Encryption: Achieve message privacy by specifying DES or AES128
message encryption. The encryption password is a string of 8-16 characters. ASCII characters 33-126 except ’#’ (ASCII 35) are allowed.
© 2017 Westermo Teleindustri AB
83
Westermo OS Management Guide
Version 4.21.1-0
ˆ OIDTREE: Limit access to a certain branch of the supported MIB. Defaults to the whole tree (’1.’)
Use ”no rouser <USERNAME>” to remove a specific read-only user, or ”no
rouser” to remove all read-only users.
Use ”show rouser” show settings for configured SNMPv3 read-only users.
Default values Disabled.
Examples
ˆ Authentication and encryption:
”rouser alice auth sha1 alicepwd1 crypto aes128 alicepwd2”
ˆ Authentication with access to dot1dBridge subtree:
”rouser bob auth md5 bobspwd1 1.3.6.1.2.1.17”
6.3.8
Manage SNMPv3 Read-Write User
Syntax [no] rwuser <USERNAME> [auth <md5|sha1> <PASSPHRASE> [crypto
<des|aes128> <PASSPHRASE>]] [OIDTREE]
Context SNMP Server Configuration context.
Usage Configure a SNMP read-write user. For more information, see section 6.3.7.
Use ”show rwuser” show settings for configured SNMPv3 read-write users.
Default values Disabled.
Examples See section 6.3.7.
6.3.9
Show SNMP server status
Syntax show snmp-server
Context Admin Exec context.
Usage Show whether SNMP server is running or not.
Examples
SNMP server enabled
84
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/#> show snmp-server
SNMP server running as PID: 540
example:/#>
SNMP server disabled (see ”no snmp-server” in section 6.3.1).
Example
example:/#> show snmp-server
No SNMP server currently running
example:/#>
© 2017 Westermo Teleindustri AB
85
Westermo OS Management Guide
Version 4.21.1-0
Part II
Common Switch Services
86
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Chapter 7
General Switch Maintenance
7.1
Overview
The table below summarises maintenance features available for the different
management tools. General descriptions of these features are presented in sections 7.1.1-7.1.9. If you are only interested in knowing how to manage maintenance features via the Web or CLI, please visit sections 7.2 or 7.3 directly.
Feature
Firmware Upgrade
Upgrade primary firmware
Upgrade backup firmware
Upgrade bootloader
View firmware versions
Web
CLI
X
X
X
X
X
X
Section 7.1.1
-”-”-”-
X
X
X
X
Section 7.1.2.2
-”-”-”-
X
Section 9.1.1.1
Section 7.1.3
Continued on next page
X
Bootstrap Options
Manage Cable Factory Reset
Configuration File Media
BOOTP Bootstrap Settings
USB Bootstrap Settings
Login Account management
Set Admin Password
Recover from lost Admin Password
© 2017 Westermo Teleindustri AB
X
General Description
87
Westermo OS Management Guide
Version 4.21.1-0
Feature
Configuration Files and Reboot
Reset to Factory Default
Reboot
View Configuration Files
Alternate Configuration Files
Configuration Backup
Configuration Upload
Auto-Backup and Restore (USB)
Configuration Deployment (USB)
Virtual File System
Maintenance of Configuration
Log and USB files
Certificate and Key Management
Upload PKCS#12 Bundle
Upload PEM file
Public Certificate
Private Key
CA Certificate
Upload OpenVPN static key file
Set (non-default) Label
Maintenance and diagnostic tools
Ping
Traceroute
IPConfig Client
Port Monitoring
Wake-On-LAN
SSH Client
Telnet Client
Tech Support
Web
X
X
(X)
Continued from previous page
CLI General Description
X
X
X
X
X
X
X
Section 7.1.3
Section 7.1.4
-”Sections 7.1.4 and 7.1.5
Sections 7.1.4 and 7.1.5
Sections 7.1.4 and 7.1.5
Section 7.1.6
Section 7.1.7
X
X
Section 7.1.5
-”-
X
X
X
X
X
X
X
X
X
X
X
X
X
Section 7.1.8
-”-”-”-”-”-”-
X
X
X
X
X
X
X
X
X
X
X
X
Section 7.1.9
-”-”-”-”-
X
X
(X)
X
Continued on next page
88
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Feature
Other maintenance features
Show System Environment Sensors
Show System Uptime
Show Memory Usage
Show Running Processes
Show Flash Table
Update Flash Table1
7.1.1
Web
X
X
X
Continued from previous page
CLI General Description
X
X
X
X
X
X
-”-
WeOS Firmware
A WeOS unit holds two types of firmware:
ˆ System firmware: The system firmware holds the operating system, which
is what we usually refers to when we say WeOS. For robustness purposes, a
WeOS unit typically holds two separate system firmware images.
– Primary firmware image: The primary firmware image (or primary image) contains the system firmware loaded by default by the bootloader.
– Backup firmware image: The backup firmware image (also known as
backup image or secondary image) contains the system firmware image loaded in case an error is encountered while loading the primary
image.
Hint
It is strongly recommended to use the same system firmware version for the primary and backup image to ensure that the backup
and primary firmwares interpret the configuration the same way.
For information on how to keep the primary and backup firmware synchronised, see section 7.1.1.2.
1 The flash partition table can only be updated on early RedFox units (RFI and RFR), in order to
upgrade to WeOS 4.3.0 or later. See section 7.1.10 for details.
© 2017 Westermo Teleindustri AB
89
Westermo OS Management Guide
Version 4.21.1-0
ˆ Bootloader: The bootloader firmware (or simply ”bootloader”) is the basic
firmware run to bootstrap the system. The bootloader will in turn load the
system firmware (trying the primary image first).
It is possible to upgrade both the system firmware (primary and secondary image) and the bootloader firmware. As of WeOS v4.21.1, the system firmware can
be upgraded via the Web or via the CLI, while the bootloader is only possible to
upgrade via the CLI.
Warning
There is no general guarantee that an older system firmware can be loaded
into the switch, i.e., downgrade is not generally guaranteed to work. However, if the firmware is downgraded for example from version 4.16.0 to
4.15.1, it is recommended to reboot the switch once the old firmware has
been installed. When the switch comes up with the old firmware (here
4.15.1), copy the factory default configuration to the running configuration.
See section 7.1.4 for more information on configuration files.
7.1.1.1
Upgrading firmware and bootloader
WeOS firmware and bootloader can be downloaded from www.westermo.com.
The method to upgrade firmware and bootloader differs somewhat if the unit to
upgrade is running WeOS 4.13.1 (or later), as compared to units running releases
before 4.13.1.
ˆ Units running WeOS 4.13.1 or later: The WeOS firmware and bootloader can
be upgraded using a common ”pkg” file in WeOS 4.13.1 and later. This is
explained further in section 7.1.1.1.2.
Note
WeOS releases older than 4.13.1 (e.g., WeOS 4.13.0 or 4.11.2) are unable to handle ”pkg” files.
ˆ Units running releases earlier than WeOS 4.13.1: When upgrading WeOS
units running older versions than WeOS 4.13.1 (e.g., WeOS 4.13.0 or 4.11.2),
there are individual firmware and bootloader files per WeOS product. This is
described in section 7.1.1.1.1.
90
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Hint
If your unit is running, e.g., WeOS4.12.0, and you wish to upgrade using a ”pkg” installation file (e.g., ”WeOS-4.14.0.pkg”) you first need to
upgrade to WeOS 4.13.1 using the old method in section 7.1.1.1.1.
Hint
If the switch reports lack of free memory when trying to upgrade the
firmware, try to disable non-essential services on the switch.
7.1.1.1.1 Upgrading when running older firmware than WeOS 4.13.1
Before WeOS 4.13.1 the firmware installation file to use differed per product family. Similarly, there were different bootloader installation files per product. A
summary of name conventions is given in the table below:
Product
Primary and
secondary FW
Bootloader FW
RedFox
rwXXXX.img
(e.g., rw4112.img)
lwXXXX.img
(e.g., lw4112.img)
wwXXXX.img
(e.g., ww4112.img)
fwXXXX.img
(e.g., fw4112.img)
xscale-redboot-YYY.bin
(e.g., xscale-2.03.bin)
imx27-redboot-ZZZ.bin
(e.g., imx27-redboot-4.11.bin)
”
”
”
”
Lynx and
Viper
Wolverine
Falcon
If you run a release older than 4.13.1, and wish to upgrade to 4.14.0 or later,
where only ”pkg” files are supported, you must first upgrade to 4.13.1 (or some
later 4.13.x release) using ”img” files1 .
Hint
Although any 4.13.x release from 4.13.1 and later can be used as intermediate release when upgrading to pkg files, it is recommended that you use the
most recent 4.13.x release. See www.westermo.com for download of WeOS
4.13 releases.
1 WeOS 4.13.1 and later 4.13.x releases are available both as ”img” and ”pkg” files, while only
”pkg” files are available from WeOS4.14.0 and onward.
© 2017 Westermo Teleindustri AB
91
Westermo OS Management Guide
Version 4.21.1-0
Below there are examples showing how to upgrade the primary firmware to a
WeOS 4.13 release with support for ”pkg” files (here ”4.13.4” is used) and bootloader via a FTP server (or TFTP server) at 192.168.3.10 on a WeOS Lynx unit.
ˆ Upgrading primary firmware via CLI on a Lynx (before WeOS 4.13.1). Here
we upgrade to WeOS 4.13.4 from a FTP server at 192.168.3.10.
Example
example:/#> upgrade primary 192.168.3.10 lw4134.img ...
ˆ Upgrading bootloader via CLI on a Lynx (before WeOS 4.13.1). Here we
upgrade the bootloader to ”imx27-redboot-4.11.bin” from a FTP server at
192.168.3.10.
Example
example:/#> upgrade boot 192.168.3.10 imx27-redboot-4.11.bin ...
7.1.1.1.2 Upgrading when running WeOS 4.13.1 (or later) If you have
WeOS 4.13.1 or later installed, upgrading firmware or bootloader is simplified
in the sense that the same installation file (a ”pkg” file) is used for all types of
upgrades (bootfile or firmware) on any type of WeOS product.The table below
lists the firmware used upgrade system firmware and bootloader.
Product Family
System Firmware
(Primary/Secondary Image)
Bootloader Firmware
All WeOS products
WeOS-X.X.X.pkg
(e.g., WeOS-4.21.1.pkg)
WeOS-X.X.X.pkg
(e.g., WeOS-4.21.1.pkg)
Thus, upgrading the primary (or secondary) system firmware image, or the bootloader will be done using the same (pkg) installation file.
Note
If you use TFTP for upgrading with ”pkg” files, make sure your TFTP server
supports large files as defined in RFC2347[27].
Be aware that upgrade using TFTP may be much slower compared to the FTP or
HTTP methods. This is of particular concern if the link you are transferring data
through has high latency. Some examples are: ADSL/VDSL/SHDSL links, 3G/4G
92
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
links or accessing via VPN tunnel.
This is an effect of how the TFTP protocol works. Every data block that is sent
is ACKed by the other end, and the sender will wait for this ACK before sending
the next piece of data. FTP and HTTP use TCP for transfer, and TCP has its sliding
window algorithm that is much better suited for high latency scenarios.
Example
An example calculation of approximate transfer time for a high latency link:
Let’s say the data is 50 Mbyte (PKG files are often larger than this) and the
latency, or round-trip-delay, is: 50 ms.
The standard TFTP block size is 512 bytes.
50 Mbyte divided in 512 byte sized blocks means 102400 blocks.
This translates to 5120 seconds at 50 ms per block (1 hour and 25 minutes).
Below you find examples of upgrading firmware and bootloader using ”pkg” files:
ˆ Upgrading firmware via CLI: Here we upgrade the primary firmware to ’WeOS
4.21.1 from a FTP server (or TFTP server) at 192.168.3.10.:
Example
example:/#> upgrade primary 192.168.3.10 WeOS-4.21.1.pkg
...
ˆ Upgrading bootloader via CLI: Here we upgrade to the bootloader from a FTP
server (or TFTP server) at 192.168.3.10.):
Example
example:/#> upgrade boot 192.168.3.10 WeOS-4.21.1.pkg
...
Note
If your unit has an older version than WeOS 4.13.1 (e.g., WeOS 4.12.1), you
are not able to upgrade using WeOS ”pkg” installation files directly. You
first need to upgrade to WeOS 4.13.1 (or a later 4.13.x release) using the
methods described in section 7.1.1.1.1.
© 2017 Westermo Teleindustri AB
93
Westermo OS Management Guide
Version 4.21.1-0
7.1.1.2
Keeping Primary and Backup Firmware Synchronised
It is recommended to use the same version for primary and backup firmware.
This ensures that your unit will have same functionality if it boots on the backup
firmware as on the primary firmware.
Therefore, when upgrading the primary firmware, you are recommended to upgrade the backup firmware too. This section includes a 4-step example, where it
is assumed you wish to upgrade the primary firmware on a WeOS unit from WeOS
4.13.4 to WeOS 4.14.1, i.e., from image ”WeOS-4.13.4.pkg”2 to ”WeOS-4.14.1.pkg”.
1. Prepare: (This step is not necessary if you did steps 3 and 4 during an earlier
upgrade, or if you have never upgraded your unit.)
Before upgrading the primary firmware, check that the backup firmware is
of the same version as the primary (here WeOS 4.13.4), and that the startup
configuration file is matching the firmware version.
(a) Startup Configuration file matching current firmware version (here WeOS
4.13.4): The simplest way to ensure that your startup configuration file
is in-line with the current firmware version is to click an Apply ”button” in the Web (e.g., Apply in the IGMP configuration page, see section 20.2), or to run ”copy running-config startup-config” in the
CLI (see section 7.3.23).
Note
From WeOS 4.15.0 and onward, this step is no longer necessary, as
the startup configuration will then automatically be updated in-line
with the current firmware version. See also section 7.1.4.
(b) Verify that version of backup image is the same as the primary firmware:
To find out what firmware version you are using, see Detailed System Overview page in the Web (see section 4.4.2) or use the ”show
system-information” in the CLI (see section 7.3.2). In the example
below the primary firmware version is 4.13.4 and the backup is 4.9.2.
2 WeOS
94
4.13.1 and later 4.13.x releases are available both in ”pkg” and ”img” format.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/#> show system-information
System Information
=============================================================================
System
System
System
System
Name
Contact
Location
Timezone
: example
:
:
: Etc/UTC
Product Family
: Lynx
Architecture
: mxc
Article number
: 3643-0105-007
Boot loader ver.
: 4.11
Main firmware ver. : 4.13.4
... (More info follows)
Model
:
Base MAC Address
:
Serial Number
:
Active firmware
:
Backup firmware ver:
L210
00:07:7c:10:de:80
16975
Main
4.9.2
If the backup image is of a different version (as in the example above),
you should upgrade the backup firmware (to WeOS 4.13.4) before moving to step 2. To upgrade the backup firmware (to WeOS 4.13.4), either use the Web upgrade facility, see section 7.2.1, or use the CLI
”upgrade” command, see section 7.3.1. The example below shows an
upgrade of the backup firmware from a FTP/TFTP server at 192.168.3.10.
Example
example:/#> upgrade secondary 192.168.3.10 WeOS-4.13.4.pkg
==> Upgrade in progress, console disabled.
Please stand by ... <==
Connecting to 192.168.3.10:21 (192.168.3.10:21)
WeOS-4.13.4.pkg
100% |*****************************| 57747k
0:00:00 ETA
Checking download ...
Unpacking weos (from /upgrade/download)...
Setting up weos (4.13.4-1)...
Checking
Type:
ID:
Size:
CRC:
lw4134.img ...
CramFS
OK (Lnx2)
OK
OK 0xDC73D8CD
Flashing /dev/mtd2 ...
100% - [====================================================================]
Updating RedBoot directory with new CRC ...
100% [====================================================================]
Done.
example:/#>
© 2017 Westermo Teleindustri AB
95
Westermo OS Management Guide
Version 4.21.1-0
2. Upgrade primary: To upgrade the primary firmware to WeOS 4.14.1, either
use the Web upgrade facility (see section 7.2.1), or use the CLI ”upgrade”
command from the CLI (see section 7.3.1). E.g., use ”upgrade primary
192.168.3.10 WeOS-4.14.1.pkg” to upgrade the primary firmware from a
FTP/TFTP server at 192.168.3.10. Compare with the example in step 1b.
Note
As you are running your unit on a primary firmware, upgrading the
primary firmware implies that the unit will automatically be rebooted
when the upgrade finishes.
3. Login and confirm configuration: At the end of the upgrade process, the
unit will reboot, using the new primary image if the upgrade procedure succeeded. After logging in again, do the following steps:
(a) Verify configuration: Verify that the unit works as expected, doing whatever tests you find necessary for your use case. If the unit does not
work as excepted, you should either consider downgrading to the previous version (here WeOS 4.13.4) or to inspect the running configuration
to find and correct the cause of your problems.
Note
If you decide to downgrade, it is recommended to do that before changing or saving startup configuration for the new version
(WeOS 4.14.1), as there are no general guarantees that the older
WeOS version can interpret a later configuration file in exact the
same way.
(b) Make Startup Configuration file match the new firmware version (here
WeOS 4.14.1): (This is similar to step 1a, but now for the new firmware.)
If the unit works as expected, store the configuration in-line with the
new firmware (WeOS 4.14.1). The simplest way is to click an Apply
”button” in the Web (e.g., Apply in the IGMP configuration page, see
section 20.2), or to run ”copy running-config startup-config” in
the CLI (see section 7.3.23).
Note
From WeOS 4.15.0 and onward, this step is no longer necessary, as
the startup configuration will then automatically be updated in-line
with the current firmware version. See also section 7.1.4.
96
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
4. Upgrade backup firmware: The last step is to upgrade the backup firmware
to the new WeOS version (here 4.14.1). For this you can use the Web
upgrade facility, see section 7.2.1, or the CLI ”upgrade” command, e.g.,
”upgrade secondary 192.168.3.10 WeOS-4.14.1.pkg” to upgrade the secondary firmware from a FTP/TFTP server at 192.168.3.10. Compare with the
example in step 1b.
© 2017 Westermo Teleindustri AB
97
Westermo OS Management Guide
Version 4.21.1-0
7.1.2
System bootstrap
During system bootstrap, the bootloader firmware is responsible for loading the
system firmware. This is described further in section 7.1.2.1.
As part of the bootstrap, the WeOS unit is also capable of conducting a cable
factory reset (section 7.1.3.3). The configuration is typically read from flash
(startup-configuration file), but it is possible to retrieve the configuration from
USB (section 7.1.6-7.1.7), or via BOOTP. Options for controlling these and other
bootstrap related settings is covered in section 7.1.2.2.
7.1.2.1
Loading System Firmware (WeOS)
The bootloader attempts to load the primary system firmware image, with fallback to loading the secondary system firmware if fails to load the primary firmware.
As described further below, different WeOS products use different bootloaders
(Barebox, U-boot or RedBoot).
The Barebox bootloader enables you to stop the bootstrap process (from console
port, press Ctrl-C at system startup), and enter an interactive boot-menu.
Example
Barebox Boot Menu
1: Primary Partition
2: Secondary Partition
3: Network (BOOTP)
4: System Recovery
5: Shell
Access to the Barebox boot-menu can be password protected (section 7.1.2.2).
From the boot-menu you can select which system firmware image (WeOS) to
load (primary or secondary image on flash), but you can also choose to download
a firmware remotely via TFTP into RAM, by entering the rescue-mode (System
Recovery).
Note
As of WeOS v4.21.1, use of BOOTP in the Barebox boot-menu (alternative
”3.”) is a technology preview. Use of TFTP (rescue mode) or BOOTP is
limited to Ethernet ports with ”internal PHY”; SFP ports can for example not
be used.
98
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Warning
Do not enter the bootloader shell (option ”5.”) unless you know what you
are doing. Use of the bootloader shell is unsupported and can result in a
broken unit.
If Barebox fails to load both the primary and secondary firmware, it will enter
the rescue-mode, which you can access via the console port. As when entering
rescue-mode from the regular boot-menu, you can download a new firmware into
RAM via TFTP. Once the unit has booted, you can login and conduct a regular
firmware upgrade (storing the firmware to flash).
In rescue-mode, Barebox also provides a rescue console service (UDP network
console), which is useful if you do not have access to a console cable, or if your
WeOS product lacks a console port. The rescue console can be accessed using
any tool that can open a UDP socket, e.g., netcat on a Unix system ”nc -u -p
6000 192.168.2.200 6000” if the default IP and UDP port numbers are used;
this assumes your PC has IP address 192.168.2.1. Section 7.1.2.2 gives more
information on configuration options related to the rescue console.
WeOS units run different types of bootloaders (Barebox, U-boot or RedBoot), and
the boot-menu and rescue-mode features described above only apply to Barebox. The following bootloaders are used by different the different WeOS product
platforms.
ˆ Atlas: Products based on the Atlas platform use the RedBoot bootloader
ˆ Basis: Products based on the Basis platform also use the RedBoot bootloader
ˆ Corazon: Products based on the Corazon platform use the U-boot or Barebox bootloader. Barebox is supported from WeOS 4.15.2, and is now the
preferred bootloader for Corazon products.
ˆ Coronet: Products based on the Coronet platform use the Barebox bootloader.
For information about what platform your product has, see section 4.4.2 (Web),
or section 7.3.2 (CLI), or see the product list in section 1.5.
If you wish to check what type of bootloader (Barebox, U-boot or RedBoot) your
unit runs, use the ”show partitions” command as described in section 7.3.46.
See section 7.1.1.1 for information on how to upgrade your bootloader.
© 2017 Westermo Teleindustri AB
99
Westermo OS Management Guide
Version 4.21.1-0
7.1.2.2
Bootstrap options
ˆ Enable/disable cable factory reset: As part of the bootstrap, the WeOS unit is
capable of conducting a cable factory reset as described in section 7.1.3.3.
Cable factory reset is by default enabled.
Disabling cable factory reset speeds up the boot and, for some installations,
reduces the risk of inadvertently triggering a factory reset on devices where
only one looped cable is required to perform factory reset.
Example
example:/#> boot
example:/boot/#> no cable-reset
example:/boot/#> show cable-reset
Disabled
example:/boot/#> end
example:/#>
Warning
When cable factory reset is disabled, the console port remains the last
means to perform a factory reset.
ˆ Configuration Boot Media: WeOS supports two methods to retrieve configuration file(s): from the on-board flash (default), from TFTP server (by use of
BOOTP), and there are also options to deploy or restore configuration from
a USB stick.
– Flash: By default the WeOS unit boots using configuration files (startupconfiguration, VPN certificates, etc.) from the (on-board) flash. The
configuration on flash is also used as fall-back when other methods fail.
– BOOTP: It is possible to bootstrap the configuration using BOOTP. For
this you need a DHCP/BOOTP Server (section 22), and a TFTP Server,
holding the unit’s configuration file. As of WeOS v4.21.1, it is only possible to use BOOTP/TFTP to download the WeOS configuration file (certificates for IPsec, etc., can not be downloaded).
Note
Bootstrapping the configuration file using BOOTP is only possible
over the WeOS unit’s Ethernet ports. DSL ports (SHDSL, ADSL,
VDSL) can not be used.
100
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
– USB: It is possible to retrieve the configuration from a USB stick3 by
utilising WeOS USB Auto-Backup & Restore (section 7.1.6) or WeOS USB
Deployment (section 7.1.6) functions4 . These services have precedence
over bootstrapping from Flash and BOOTP, but can be disabled (see USB
Bootstrap Settings below).
ˆ BOOTP Bootstrap Settings: When using BOOTP as configuration boot media,
you can specify the BOOTP timeout (default 5 minutes), i.e., the maximum
time to wait for the BOOTP/TFTP configuration file download to succeed.
Fall-back is to use configuration on on-board flash.
By default, the downloaded configuration file is only stored in RAM. You can
manually store it to flash (e.g., by ”cp running-config startup-config”),
but you can also configure the WeOS to store the file to startup-config on
flash automatically after download.
ˆ USB Bootstrap Settings: During bootstrap, a WeOS unit checks if there is a
USB stick attached in order to restore section 7.1.6) or deploy (section 7.1.6)
a configuration from the USB stick.
– Timings: There are two timings related to Bootstrap and USB services:
* Delayed USB backup/restore and USB deploy: (Non-configurable) A
USB media not plugged in (or detected) when the device boots up
can still be used to backup/restore or deploy the device configuration up to 30 seconds after power on.
* USB bootstrap timeout: (Configurable) The USB bootstrap timeout
halts boot for specified number of seconds, waiting for USB media
to settle and be detected by the device. Before the timeout has
elapsed and no media has been detected the device is unreachable
with all ports remaining in blocking. Default: Disabled (i.e., zero
delay)
3 See
section 1.5.1 for WeOS products with USB interfaces, and section 7.1.5.1 for list of USB
sticks verified for use with WeOS.
4 As a technology preview feature, there is also a boot media option referred to as ”boot from
USB”. See WeOS release notes for more information on WeOS technology previews in general and
for specific information on the ”boot from USB” function.
© 2017 Westermo Teleindustri AB
101
Westermo OS Management Guide
Version 4.21.1-0
Hint
Setting a ”USB bootstrap timeout” is useful to avoid a situation
where the unit first applies the configuration from on-board flash,
and afterwards detects the USB stick and applies USB restore or
deploy (”Delayed USB backup/restore and USB deploy”).
– Enable/Disable: USB bootstrap services can be disabled. Disabling USB
bootstrap services implies disabling USB Deployment and automatic
USB Backup & Restore features. Manual backup and restore to/from
a USB stick is still possible. Default: Enabled
Warning
USB bootstrap services are enabled by default for ease of use and
robustness. However, it gives users with physical access to the
switch the opportunity to modify or retrieve the configuration without logging in. If unauthorised personnel have physical access to
the unit it is recommended to disable USB bootstrap services for
security purposes.
Below is an example of how to disable USB Bootstrap services.
Example
example:/#> boot
example:/boot/#> usb
example:/boot/usb/#> no enable
example:/boot/usb/#> show
Status
: Disabled
Timeout
: Disabled
example:/boot/usb/#> leave
example:/#>
ˆ Barebox boot-menu options: Boot options related to the Barebox boot-menu
(boot-menu password, rescue console settings, etc.) are described in sections 7.3.16-7.3.21.
102
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
7.1.3
What to do if you cannot access your switch
Occasionally you may end up in a situation where you cannot access your switch:
ˆ Forgetting IP address: If you have forgotten what IP address you assigned to
your switch, you will no longer be able to access it remotely (Web, SSH, Telnet, SNMP). Section 7.1.3.1 presents different methods to find the IP address
of your switch.
ˆ Forgetting password: If you have forgotten the admin password you assigned to your switch, you should conduct either a factory reset or a password reset. Both alternatives require that you have physical access to the
switch.
– Factory Reset: By resetting to factory defaults, the switch configuration (including the ”admin” password) will be reset5 . I.e., the ”admin”
password is restored to ”westermo”, enabling you to login again.
The way to accomplish a factory reset may differ if the switch has a console port (section 7.1.3.2) or if it lacks a console port (section 7.1.3.3).
– Password Reset: On switches with a console port there is a possibility to
reset the ”admin” password to its default value (”westermo”) without
affecting the rest of the configuration, see section 7.1.3.2.
ˆ Misconfiguration: You may also lose the ability to access your switch remotely (Web, SSH, Telnet, SNMP, WeConfig) due to misconfiguration, e.g.,
by disabling all Ethernet ports, or moving them to a VLAN where the switch
has no IP address assigned. This case can be resolved by logging into the
switch via the console port, and change the configuration appropriately via
the CLI (see chapter 5 on information of how to access the CLI via the console port).
However, if the switch does not have a console port, you may need to conduct a factory reset as described in section 7.1.3.3.
5 Only
files on built-in flash are affected. Files on any attached USB stick are not affected.
© 2017 Westermo Teleindustri AB
103
Westermo OS Management Guide
Version 4.21.1-0
7.1.3.1
Discovering the IP address of your switch
The factory default IP setting enables access to the switch via IP address 192.168.2.200,
as well as via an address assigned via a DHCP server6 (see table 7.4).
Primary IP address
Secondary IP address
Address
Netmask
Gateway
Dynamic (DHCP)
192.168.2.200
(Dynamic)
255.255.255.0
(Dynamic)
Disabled
Table 7.4: Factory Default IP settings.
There are several ways to discover the IP address of a product:
1. WeConfig (from PC): The WeConfig tool is designed to scan for (Westermo)
switches on the local network. See the WeConfig User Guide[64] for details on how to use the WeConfig tool. This option is probably the simplest
method to find the IP address of a switch, but will not work if the IPConfig
service has been disabled on your switch (see section 8.3.21 for information
on how to enable/disable IPConfig on your switch).
2. IPConfig client (from switch): The WeOS CLI and the Web contain an IPConfig client scanning facility, thus if you are logged into a switch you are to
scan for neighbour switches. As in the previous step, switches can only be
discovered this way if they have the IPConfig service enabled.
3. Via console port: On switches equipped with a console port, the IP address of
the switch can be found using the switch Command Line Interface (CLI). See
chapter 5 for more information of how to use the CLI. (If you have forgotten
the admin password, please see section 7.1.3.2).
4. LLDP: When LLDP is enabled (section 8.1.1), WeOS announces itself (IP address) in LLDP messages. Thus, an LLDP client (or a network sniffer such as
Wireshark7 ) can be used to discover the IP address of the switch.
In case you are not able to discover the IP address by any of these methods,
conducting a factory reset will take the switch back to its original IP configuration
(as shown in table 7.4). See sections 7.1.3.2 and 7.1.3.3 for information on how
to conduct a factory reset.
6 In addition, the unit will autoconfigure itself with a link-local address in the 169.254.x.x range,
where ’x’ is in interval 0-255. See section 21.2.6 for more information.
7 Wireshark network protocol analyser, http://www.wireshark.org.
104
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
7.1.3.2
Password or Factory Reset via Console Port
For WeOS switches equipped with a console port, it is possible to conduct a factory reset or just a password reset using the special accounts (factory or password). For security reasons, these special accounts can only be used via the
console port. For security hardening purposes, these two special accounts can be
disabled in the device’s boot context, in the CLI (see sections 7.3.11 and 7.3.12).
ˆ Admin password reset: It is possible to recover from a lost admin password
by using the following login and password from the console port. The admin
password will be reset to its default value (westermo), and thereby enable
you to login to the switch again.
– Login: password
– Password: reset
ˆ Factory reset: It is possible to reset the switch to factory defaults with the
following login and password from the console port. The switch configuration
(including the ”admin” password) will be reset8 .
– Login: factory
– Password: reset
7.1.3.3
Factory Reset without using Console Port
There is a mechanism to conduct a factory reset without using the console port
or being logged into the unit – this method is referred to as ”cable factory reset”.
The cable factory reset mechanism can be disabled as a bootstrap option, see
section 7.1.2.2.
Note
Depending on the type of product, cable factory reset is conducted by connecting one pair of Ethernet ports (single cable) or two pairs of Ethernet
ports (two cables) as shown in the table below.
On products where a single cable is used to perform the reset, with ports
that are otherwise used for FRNT or RSTP, it is recommended to disable
cable factory reset in the bootstrap options.
8 Only
files on built-in flash are affected. Files on any attached USB stick are not affected.
© 2017 Westermo Teleindustri AB
105
Westermo OS Management Guide
Version 4.21.1-0
1. Power off the switch and disconnect all Ethernet cables (including copper
and fiber cables) and DSL cables.
2. Connect one pair (or two pairs) of Ethernet ports as described in the table
below. The ports need to be connected directly, i.e., not via a hub or switch.
Use a straight cable - not cross-over cable - when connecting a port pair.
Product/Model
Falcon
FDV-206-1D1S
Lynx
L106/206-F2G
L110/210
Lynx-DSS
L105/205-S1
L106/206-S2
L108/208-F2G-S2
RedFox Industrial
All RFI models
RedFox Industrial Rack
All RFIR models
RedFox Rail
RFR-12-FB
Viper
All Viper-20A models
All Viper-12A models
All Viper-12 models
Wolverine
DDW-x42
DDW-x42-485
DDW-225/226
Ethernet Port Pair 1
Ethernet Port Pair 2
port 1 ⇔ port 4
port 2 ⇔ port 3
port 3 ⇔ port 6
port 3 ⇔ port 10
port 4 ⇔ port 5
port 6 ⇔ port 7
port 1 ⇔ port 4
port 1 ⇔ port 4
port 3 ⇔ port 6
port 2 ⇔ port 3
port 2 ⇔ port 3
port 4 ⇔ port 5
port 1/1 ⇔ port 1/2
Not applicable
port 1 ⇔ port 2
Not applicable
port X1 ⇔ port X6
port X2 ⇔ port X5
port X9 ⇔ port X16
port X5 ⇔ port X10
port X1 ⇔ port X6
port X10 ⇔ port X15
port X6 ⇔ port X9
port X2 ⇔ port X5
port 1 ⇔ port 2
port 1 ⇔ port 2
port 2/1 ⇔ port 2/4
Not applicable
Not applicable
port 2/2 ⇔ port 2/3
3. Power on the unit.
4. Wait for the unit to start up. Verify that the ON LED is flashing red. This
flashing indicates ready to reset. You can now go ahead with the factory
reset, or abort the procedure.
ˆ Go ahead with factory reset: Acknowledge the factory reset by unplugging (one of) the cable(s). The ON LED stops flashing.
106
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
This initiates the factory reset process, and after a short while, the unit
will restart with factory default settings.
ˆ Abort the factory reset: Simply wait for the procedure to time out, do
not remove any of the cables. This takes approximately 30 seconds
after the ON LED started flashing RED. The switch will conduct a normal
boot with the current startup config.
© 2017 Westermo Teleindustri AB
107
Westermo OS Management Guide
Version 4.21.1-0
7.1.4
Configuration Files and Reboot
The system keeps three special configuration files:
ˆ Startup Configuration: The configuration file used by the switch after system
boot or reboot. The startup configuration is stored in non-volatile memory
(flash)9 .
Note
From WeOS 4.15.0 and onward, the startup configuration is verified to
be in-line with the syntax of the current firmware version upon system
boot. If there are deviations (which may be the case after a firmware
upgrade), the startup configuration is automatically updated.
ˆ Running Configuration: The configuration currently used by the switch. The
running configuration is kept in volatile memory (RAM).
The running configuration is identical to the startup configuration when using the Web interface, the WeConfig tool or SNMP. That is, when using these
methods to manage the switch, a change in the running configuration is
immediately stored in the startup configuration.
In contrast, when managing the switch via the CLI, configuration changes
only affect the running configuration. Thus, to survive a reboot, the running
configuration must be copied to the startup configuration.
ˆ Factory Default Configuration: The system always has a factory default configuration. It is kept in non-volatile memory (flash) and cannot be overwritten. At first boot, as well as after a factory reset, the startup configuration
file is identical to the factory default configuration file.
In addition to these configuration files, it is possible (via CLI) to keep a set of additional configuration files on the switch, which enables easy swapping between
alternate configurations.
9 As described in section 7.1.5, it is possible to keep several configuration files on flash. The
startup configuration file is actually a symbolic name for one of the stored configuration files.
108
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Warning
Configuring the switch via multiple management interfaces in parallel is discouraged, since it may lead to unexpected behaviour.
For example, consider the case when two users are accessing the switch at
the same time, one user via the CLI and another user via the Web interface:
Assume the ”CLI user” makes changes to the running configuration, but of
some reason do not wish to copy these changes to the startup configuration
(yet).
If the another user, the ”Web user”, applies a single change using the web
management tool, all the changes done to the running configuration (by the
”CLI user”) will be saved to the startup configuration. (Actually clicking the
Apply button, even without changing any values has the same affect.)
7.1.4.1
Account password when loading a configuration file
Configuration files contain information on user account and (hashed) passwords,
e.g., for the ”admin” account. Thus, when loading a configuration file to the
switch (i.e., overwriting the startup-configuration or running-configuration), the
account passwords will also be replaced according to the setting in the new configuration file.
Warning
To copy a new configuration file to the running-config or startup-config while
keeping the existing user names and passwords, the lines in the new configuration file containing the ”username” command should be removed before
installing the new configuration file.
If you unintentionally happen to loose the admin password because you copied
a configuration file including an unknown admin password, see section 7.1.3 for
information on how to regain access to the switch.
7.1.5
Virtual File System
WeOS keeps various files of interest for the operator:
ˆ Configuration files: By default there is only one configuration file (named
config0.cfg stored on the switch. However, it is possible to create and keep
© 2017 Westermo Teleindustri AB
109
Westermo OS Management Guide
Version 4.21.1-0
multiple configuration files on the switch, both for backup purposes of for
easy shifting between configuration setups. Configuration files are commonly named with the prefix config and will always have ’.cfg’ as extension.
As mentioned in section 7.1.4 there are also three special configuration files:
– Running Configuration: The running configuration is only stored in RAM,
thus, it is not kept over a reboot.
– Startup Configuration: The startup config is mapped to one of the stored
configurations. By default it points to config0.cfg, but the mapping
can be changed (using the CLI ”copy” command as described in section 7.3.23).
– Factory Default Configuration: The factory default configuration file
cannot be modified (except through a firmware upgrade). Its available
for the purpose of conducting a factory reset.
ˆ Log files: Events are logged in various log files, e.g.:
– auth.log
– kern.log
– messages
– mgmt.log
– snmpd
– ppp.log
For units equipped with a USB port, the operator is also able to access files on a
mounted USB stick.
The files are organised in a virtual file system, and are made available both for
local and remote access.
Configuration files
Log files
USB files
Local File Path
Remote File Path
cfg://
log://
usb://
/cfg/
/log/
/usb/
Section 7.1.5.1 gives general information on the use of USB memory sticks in
WeOS products. Section 7.1.5.2 describes available methods for file maintenance
when logged into the switch, while section 7.1.5.3 covers methods available for
maintaining files remotely.
110
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
7.1.5.1
General information on using USB memory sticks
In order to copy files to/from a USB memory stick attached to USB port of the
WeOS product10 , the USB memory stick must:
ˆ be partitioned
ˆ be formatted as VFAT or FAT32 on the first partition
As of WeOS v4.21.1 the following USB stick(s) are verified for use with WeOS
products:
Westermo USB stick 3641-0190 (Serial number 1195 or higher) for RedFox Rail
and Viper, see user guides in section 1.5.
If a factory reset is conducted on the WeOS unit, only files on unit flash (configuration, IPsec certificates, etc.) will be affected by the factory reset. Files on an
attached USB stick (if present) will not be affected.
7.1.5.2
File access when logged into the switch
An operator logged in to a switch can copy, download or upload files using the
CLI ”copy” command. Services available when logged into the system include:
ˆ Making local backup copies of files, e.g.,
”copy log://messages log://messages.5”
ˆ Upload or download to/from a remote server via TFTP, FTP, and SCP. (Downloading is also available via HTTP.)
Upload example using TFTP:
”copy cfg://config0.cfg
tftp://server.example.com/myswitchconfig.txt”
ˆ Copying between systems: The CLI copy command can be used to copy files
between remote systems via TFTP, FTP, SCP, and HTTP (HTTP can only be
used as source, not destination).
Example copying from HTTP server to TFTP server:
”copy http://server1.example.com/original.txt
tftp://server2.example.com/backup.txt”
10 For information on WeOS products equipped with a USB port, see section 1.5.1, or the User
Guide of your WeOS product (see section 1.5).
© 2017 Westermo Teleindustri AB
111
Westermo OS Management Guide
Version 4.21.1-0
7.1.5.3
Remote file access
An operator is able to upload and download files to/from the switch remotely via
SCP. This feature is convenient and saves time, since files can be maintained
without the need to log into each switch.
Example with remote file upload:
Example
unix> scp config1.cfg [email protected]:/cfg/
Password for [email protected]:
unix>
Example with remote file download:
Example
unix> scp [email protected]:/log/messages .
Password for [email protected]:
unix>
112
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
7.1.6
Automatic Backup and Restore to/from USB
On WeOS units equipped with a USB port, a USB memory stick can be used for
automatic backup and restore. The intended application for the auto-backup
function is to simplify unit replacement in case of unit failure.
Once activated, it works seamlessly. If a stick is already prepared nothing else
is needed. If a unit fails you simply replace it, moving the USB stick to the replacement unit. Which must be of same mark and model. At first boot, the
replacement unit automatically restores all necessary files from the faulty unit.
Note
The auto-backup and restore function only handles configuration. It does
not handle backup/restore of WeOS firmware images. You must not only
ensure that your replacement unit is of the same model as the original unit.
It should also have same WeOS firmware version loaded as the original unit.
Details of how to activate auto-backup, and how to perform restore are provided
in sections 7.1.6.1-7.1.6.2. Section 7.1.6.3 contains information on USB directories for auto-backup and restore.
7.1.6.1
Procedure for activating auto-backup
ˆ Basic preparations the USB stick: See section 7.1.5.1 for formatting and
partitioning requirement for USB memory sticks used with WeOS units.
ˆ Insert USB stick: Insert the USB stick into WeOS unit and power it up.
ˆ Log in to CLI: Log into the unit (CLI), either via console port or remotely via
SSH (see section 5.2).
ˆ Activate auto-backup: Run the CLI ”backup” command.
© 2017 Westermo Teleindustri AB
113
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/#> backup
WeOS Auto Backup & Restore for USB Media
===============================================================================
This command initializes a USB media, usually a memory stick, to be used for
automatic backup and restore of configuration files (including certificates).
Intended use-case is to have one memory stick for each device in the network
to ease replacement of faulty units.
The replacement WeOS unit will at boot automatically restore the backup and
seamlessly pick up where the faulty unit left off.
Configuration and certificate files, including private keys (!) are backed up
to /usb/westermo/backup/
Activate WeOS auto-backup & restore on this USB stick, are you sure (y/N)? y
Performing initial backup...
Backup done.
example/#>
The configuration files (including certificates and private keys) are now backed
up to sub-directories under ”/usb/westermo/backup/” (see section 7.1.6.3).
ˆ Keep USB inserted: The USB memory stick should stay attached to the WeOS
unit. Any changes to the configuration files on unit flash will be continuously
backed-up to USB.
An alternative method to initialise auto-backup is to create the (empty) directory
on the USB stick /westermo/backup/ (see section 7.1.6.3) before inserting it to
the WeOS unit. Power off the unit and insert the USB stick. When the unit is then
powered up, all configuration files (including certificates and private keys) will be
backed up on the USB stick automatically. If you instead insert the prepared USB
stick into a running unit, files start being backed up at your first (manual) save of
the current configuration, e.g., ”cp running startup” from CLI, or upon reboot
of the unit.
7.1.6.2
Restoring configuration from USB to replacement unit
When booting a WeOS unit checks if a USB stick is attached. If a USB stick is
found with auto-backup activated, the WeOS unit checks if a restore operation
should take place or not. This automatic restore operation only takes place at
boot-up (configuration file is copied from USB to on-board flash, and used as
startup configuration), or within an interval of 30 seconds after boot-up. In the
latter case, which can occur if the USB stick is not ready at system boot time, the
114
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
WeOS unit starts with and runs the configuration on on-board flash for a short
while; restore operation then updates both the startup-configuration and running
configuration.
Note
While replacing a WeOS unit using the USB auto-backup and restore support,
it is recommended that the unit is disconnected from the network (see step 5
in the procedure below), and therefore there should be no problem if the
replacement unit runs with the configuration on the on-board flash for a
short while. Still, if it is important that the restore operation takes place
before the WeOS reads its startup configuration, an additional boot delay
can be added (see section 7.1.2.2 as well as step 1 in the procedure below).
1. Prepare replacement unit: The replacement should be of the same model as
the original unit (e.g., a Lynx L210-F2G should be replaced by another Lynx
L210-F2G), and ensure that it has the same WeOS firmware version loaded
as the original unit.
Hint
If you are unsure of what firmware version your original unit was running, you can inspect the configuration file on your USB stick – at the
top of the configuration file used as ”startup-configuration” you
should see the WeOS version, e.g., WeOS 4.15.2.
It is recommended that the replacement unit has not had the auto-backup
feature activated already. If unsure, please do a factory reset11 of the replacement unit before proceeding. Use either of the methods described in
section 7.1.3.2 (factory reset via console port), section 7.1.3.3 (cable factory
reset), or section 7.2.4 (factory reset via web interface).
Optionally, you can then login to the replacement unit and set a USB delay
in the boot context. For example, to extend the time to discover a USB stick
at boot with up to 10 seconds, use the following commands:
Example
example:/#> boot
example:/boot/#> usb
example:/boot/usb/#> timeout 10
11 Only files on unit flash (configuration file(s), IPsec certificates, etc.) will be affected by the
factory reset. Files on an attached USB stick (if present) will not be affected.
© 2017 Westermo Teleindustri AB
115
Westermo OS Management Guide
Version 4.21.1-0
This gives the USB stick more time to settle at boot, and be ready for use
when configuration is activated (see remark at the start of this section).
Suitable USB delay differs depending on what WeOS product you are using
(boot time differs) and what USB stick you are using (see section 7.1.5.1 for
information on USB sticks verified for WeOS).
2. Unplug power of replacement unit: Before inserting the USB memory stick
holding the backup configuration you should unplug the power of the replacement unit.
3. Insert USB stick in replacement unit
4. Power up the replacement unit: When the replacement unit boots, the configuration files on USB will automatically be restored to unit flash.
5. Connect network cables: It is recommended to connect the network cables
after powering up the replacement unit. You may also connect them before
powering up the unit (see comments on timings for detecting USB stick at
the start of this section).
6. Keep USB attached: The USB memory stick should be stay attached to the
WeOS unit. Any changes to the configuration files on unit flash will be continuously backed up to USB.
The automatic restore operation is only done when booting the WeOS unit, or
within 30 seconds after boot-up12 . If the USB stick (holding backup information) is
inserted into a running unit need to reboot the unit for the auto-restore operation
to occur. Alternatively, you can run the CLI ”restore” command to manually
trigger it.
Example
example:/#> restore
Restore backup from USB stick and activate to running-config, are you sure (y/N)? y
Stopping DHCP/DNS Server ................................... [ OK ]
Starting DHCP/DNS Server ................................... [ OK ]
example:/#>
7.1.6.3
Backup files in USB directory tree
Backup files will be stored on the USB in the following directory tree.
12 The restore operation is not conducted if ”auto-backup” is already activated on the WeOS unit
and the ”gen.id” counter on the USB and unit flash have the same value, see also section 7.1.6.3.
116
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
/usb/
+-- westermo/
+-- backup/
+-- cfg/
+-- crt/
<-- Automatic Backup & Restore directory
<-- Configuration files
<-- Certificates and keys
Additional details: The ”/usb/westermo/backup/cfg/” directory will contain some
additional files: ”startup-config.lnk” specifies which config file is used as
”startup-configuration”, and ”gen.id” contains a counter. The corresponding ”gen.id” file on unit flash is incremented every time a change on unit flash
is detected. For every change the unit flash is synchronised to USB.
During the boot procedure, the ”gen.id” values on USB and unit flash are compared. If equal, it is assumed that the configuration files are synchronised (no
restore conducted). This is the case when rebooting a unit with auto-backup
activated.
7.1.7
Configuration Deployment via USB
The USB configuration deployment function can be used for several purposes:
ˆ Easy configuration deployment of one or more WeOS units: The USB stick is
only attached during unit configuration, and can then be moved to the next
unit to be configured.
ˆ To ensure a WeOS unit always boots up with a pre-defined configuration:
In this case, the USB stick will always be attached to the WeOS unit. The
configuration on USB is copied to unit flash on every boot.
Note
For this use case, you may consider setting a boot delay (section 7.1.2.2) to avoid the risk that your unit starts with and temporarily
uses the configuration on the on-board flash, see below for more explanations.
This ”USB configuration deployment” function differs from ”USB auto-backup and
restore” described in section 7.1.6 in that configuration changes applied after
boot only apply to the WeOS unit’s on-board flash – the configuration files on the
USB memory stick are not affected.
ˆ The model and WeOS version of the unit to be configured should match the
intended configuration file(s) on the USB memory stick.
© 2017 Westermo Teleindustri AB
117
Westermo OS Management Guide
Version 4.21.1-0
ˆ The USB memory stick (prepared for deployment) is inserted before the unit
is powered up. When the unit boots up configuration files will be copied from
USB to unit flash, and used during startup configuration.
ˆ The deployment function is also automatically activated if a USB stick (prepared for deployment) is detected up to 30 seconds after boot-up. In the
latter case, which can occur if the USB stick is not ready at system boot
time, the WeOS unit starts with and runs the configuration on on-board
flash for a short while; deployment operation then updates both the startupconfiguration and running configuration.
Note
To prohibit that the unit first boots using configuration stored on the
unit’s on-board flash, you can setting a boot-delay (e.g., ”boot wait
10” to extend the boot time with 10 seconds). By setting the delay
large enough, the USB stick gets enough time to be ready when startup
configuration is applied. Suitable boot delay differs depending on what
WeOS product you are using (boot time differs) and what USB stick you
are using (see section 7.1.5.1 for information on USB sticks verified for
WeOS)
ˆ The USB configuration deployment function is activated if the directory ”westermo/deploy/” is detected on an attached USB during boot-up. USB configuration deployment has precedence over USB auto-backup and restore.
That is, if the USB memory stick contains both a ”westermo/deploy/” and
a ”westermo/backup/” directory, the configuration deployment function will
be activated.
Section 7.1.7.1 provides information on the file structure and format of the
files in the ”westermo/deploy/” directory.
7.1.7.1
Deployment files in USB directory tree
Deployment configuration files should reside on the USB in the following directory
tree.
/usb/
+-- westermo/
+-- deploy/
+-- cfg/
|
+-- <FILE>.cfg
|
+-- startup-config.lnk
118
<-- USB Deploy
<-- Actual configuration file, e.g., config0.cfg
<-- Windows style .lnk file
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
+-- crt/
+-- ...
<-- Certificates and keys
The startup-config.lnk file holds the file name of the startup configuration file.
The format of this file is:
ˆ No leading directories, to avoid any / or \ confusion
ˆ No end-of-line after file name, to avoid any DOS/UNIX/Mac confusion
ˆ File name stored at first position in file, e.g., config0.cfg
As of WeOS v4.21.1 there is no CLI or Web function for setting up a USB configuration deployment memory stick for use with WeOS. Meanwhile the easiest way
might be to
1. perform a USB auto-backup (see section 7.1.6.1), and
2. plug the USB stick into a PC and rename the backup directory to deploy.
7.1.8
Certificate and Key Management
WeOS supports upload and management of certificate and key files. As of WeOS
v4.21.1, use of certificates is limited to the Web server (section 8.1.2), IPsec VPNs
(chapter 35) and SSL VPNs (chapter 36).
It is possible to upload/import PKCS#12 bundles containing public certificate, private key and the certificate of the issuing certificate authority (CA certificate).
The PKCS bundle can be password protected (recommended).
It is also possible to upload individual certificate files in PEM format or OpenVPN static key files. For further information on certificate management, see sections 7.2.6 (Web) and 7.3.29 (CLI).
A unique self-signed certificate will be generated for the Web server if no custom
certificate has been chosen. The generated certificate will use the reserved label
web-default and it will stay in the certificate repository until it is either deleted
or a factory reset is performed. It is still recommended to use a custom certificate with the Web server, as the generated certificate is self-signed. See section 7.2.6.1 (Web) and 7.3.29 (CLI) for uploading certificates, and section 8.1.2
for applying a custom certificate to the Web server.
© 2017 Westermo Teleindustri AB
119
Westermo OS Management Guide
Version 4.21.1-0
7.1.9
Maintenance and diagnostic tools
The switch supports a set of maintenance and diagnostic tools:
Ping and Traceroute The standard Ping and Traceroute commands are available via the CLI and the Web, and are useful as basic troubleshooting tools.
Port monitoring The switch supports port monitoring, thus the user can monitor the traffic exchanged on one or more Ethernet ports on a dedicated
monitor port. Only correct Ethernet packets will be forward onto the monitor destination port. To monitor occurrence of packet drops due to bad CRC,
etc., we refer to the RMON statistics counters, see chapter 11.
Note
To observe all traffic on the monitor source ports, the total amount of
traffic on the monitor source ports should not exceed the capacity of
the monitor destination port.
WeOS IPConfig Client As mentioned in chapter 3 WeOS provides the WeConfig
PC tool for discovery and rudimentary management of Westermo switches.
The CLI and the Web provides a similar mechanism (IPConfig client), i.e.,
once logged into the switch, it is possible to scan for other Westermo units
on the same LAN.
Wake-On-LAN A Wake-On-LAN (WOL) client is available via the CLI and the Web.
This allows a computer to be turned on or woken up by a network message
(magic packet).
Additional features relevant for maintenance and diagnostics are described in
chapter 11 (RMON Statistics), chapter 25 (Event and Alarm Logging), chapter 6
(SNMP), and chapter 24 (Alarm handling, Digital I/O and Front-panel LEDs).
7.1.10
Upgrading early RedFox Units to 4.3.0 or later
Early RedFox units (Industrial and Rail) delivered with WeOS 4.0.0, comes with
a flash memory partition unsuitable for the larger firmware image size of WeOS
4.3.x13 and later.
ˆ How to determine if your RedFox has an old partition table:
For RedFox Industrial, only products shipped with WeOS 4.0.0 came with the
13 WeOS
120
4.3.x refers to all patch releases (4.3.0, 4.3.1, . . . ) of the WeOS 4.3 feature branch.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
old partition table. You can determine if your product has the old partition
table by inspecting the product’s model (or the article number) and serial
number – if the serial number is lower than the ones listed below, your
product was shipped with the old partition table.
You find information on your product’s type of model, article number, and
serial number via the Web interface (Menu path: Status ⇒ System, see
section 4.4.2), or via the CLI ”show system-information” command, see
section 7.3.2).
Model
(Article number)
Serial number
RFI-18-F4G-T4G
RFI-14P-F4G
RFI-10P
RFI-18P
3641-3300
3641-3200
3641-3110
3641-3100
<
<
<
<
1190
1180
1220
1111
If you are unsure whether your flash table is already updated, you can use
the CLI ”show flash-table” command available on WeOS 4.2.0 and later
(see section 7.3.45) to list information on the flash partition table:
– Main and backup partitions of size 12.5 MB (hex 0x00c80000) means
the new partition table is used.
– Main partition of size 8.5 MB (hex 0x00880000) and backup of size 7 MB
(hex 0x00700000) means the old partition table is used.
ˆ Do you need to update your partition table?
It is possible to upgrade the primary firmware to WeOS 4.3.x even if your
RedFox has an old partition table. If your RedFox has an old partition table,
you must update it in order to:
1. Upgrade your backup firmware (i.e., the firmware on the backup partition) to WeOS 4.3.x or later.
2. Upgrade your primary firmware to a WeOS image larger than 8.5 MByte.
The WeOS 4.3.x image for RedFox is below this limit, but later firmware
versions (4.4.x and later) may be larger than 8.5 MB, and then the flash
table needs to be updated.
ˆ How to update your flash table:
© 2017 Westermo Teleindustri AB
121
Westermo OS Management Guide
Version 4.21.1-0
Warning
Updating the flash partition table will corrupt your system! Configuration files, certificates and backup image will be destroyed.
Although this update facility has been tested extensively by Westermo there
are no guarantees it will work flawlessly for all use cases.
Therefore, we do not recommend this action on active running units in the
field. Instead, replace the unit with a spare one first.
1. Backup your .cfg files, startup-config and any certificate files to a USB
stick or remote (T)FTP/SCP server (see section 7.3.23).
2. If you are running WeOS version 4.2.0 or later proceed directly to the
next step. If you are running WeOS 4.0.0, you must first upgrade your
primary firmware to a later release, e.g., 4.2.0 or 4.3.0 (see section 7.2.1
or 7.3.1).
3. Access the CLI via the console port, run the ”flash-table-update”
command (see section 7.3.47), and wait for it to finish. The unit reboots
when it has completed the update.
Note
The ”flash-table-update” command is only available on WeOS
4.2.0 and later, and is only visible if you have a RedFox with the
old partition table.
4. Restore configuration files, any necessary certificates and the system
backup image.
122
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
7.2
Maintenance via the Web Interface
7.2.1
Managing switch firmware via the Web Interface
Menu path: Maintenance ⇒ F/W Upgrade
On the firmware upgrade page you are able to upgrade firmware by downloading
an image using FTP/TFTP or by direct upload via the Web browser.
7.2.1.1
Firmware Upgrade Using File Upload
Image File
Upgrade
7.2.1.2
Select the file to upload (browser dependent).
Click the Upgrade button to initiate firmware upgrade.
Firmware Upgrade Using TFTP/FTP Server
Image name
Server address
Upgrade
The file name of the image file on the FTP/TFTP server.
The IP address of the FTP/TFTP server.
Click the Upgrade button to initiate firmware upgrade.
Note
If you use TFTP for upgrading with ”pkg” files, make sure your TFTP server
supports large files as defined in RFC2347[27].
© 2017 Westermo Teleindustri AB
123
Westermo OS Management Guide
Version 4.21.1-0
7.2.2
Port Monitoring
Menu path: Tools ⇒ Port Monitoring
Enabled
Destination Port (Mirror)
Source Ports (Sniff Ports)
124
Check the box to enable port monitoring. If
you have a JavaScript enabled browser the
other settings will not be displayed unless
you check this box.
Select one port to which data from source
ports will be copied (mirrored).
Select one or more ports to monitor by selecting the ports desired sniff mode. Available modes are:
In
Inbound (ingress) traffic.
Out
Outbound (egress) traffic.
Both Both inbound and outbound traffic.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
7.2.3
Backup and Restore
Menu path: Maintenance ⇒ Backup&Restore
To create a backup of your switch configuration on your host, visit the backup
and restore page.
Backup
File Path
Restore
Click this button to download a copy of the running configuration on your switch. You will be asked to open or save the
file. Normally chose save to save the file to your host. The behaviour is web browser specific and may also depend on your
current browser settings. See Fig. 7.1 for an example.
Click the Browse button to browse for the file. The behaviour
of the file selection is browser specific.
Click this button to restore the configuration the configuration
described in the file you selected in File Path.
© 2017 Westermo Teleindustri AB
125
Westermo OS Management Guide
Version 4.21.1-0
Figure 7.1: Example save dialogue (this example is from a Firefox browser)
7.2.4
Factory Reset
Menu path: Maintenance ⇒ Factory reset
To conduct a factory reset, press the Reset button.
Only configuration files on unit flash will be affected by a factory reset. Files on
an attached USB stick (if present) will not be affected.
126
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
7.2.5
Restart
Menu path: Maintenance ⇒ Restart
To restart the switch press the Restart button.
7.2.6
Managing certificates and keys
Menu path: Maintenance⇒Certificates
When entering the certificates page you will be presented to a list of all certificates and keys available on your switch. Here you can import or delete certificates/keys.
Type
Label
The type of certificate/key: Public (regular certificate),
Private (a private key belonging to a regular certificate), CA
(a CA certificate), or OpenVPN (an OpenVPN static key).
A label identifying the certificate/key. Unique per certificate
file type (Public, Private, CA and OpenVPN).
Continued on next page
© 2017 Westermo Teleindustri AB
127
Westermo OS Management Guide
Version 4.21.1-0
Common
Name (CN)
Expires
Delete
Details
Import
7.2.6.1
Continued from previous page
The common name (CN) part of the distinguished name (DN)
found in the imported certificate’s subject.
The date of expiration for the certificate.
Click this icon to remove a certificate/key. You will be asked to
acknowledge the removal before it is actually executed.
Click this icon to display details regarding a certificate.
Click this button to import a certificate or key.
Import Certificates
Menu path: Maintenance ⇒ Certificates ⇒ Import
When clicking the Import button you will be presented to the certificate import
page where you can import PKCS#12 certificate bundles, certificates and private
key files in PEM format, or an OpenVPN static key.
Type
File
Type of
Certificate
128
Select the type of file to import (PKCS#12 bundle, PEM file or
OpenVPN static key file).
Browse your file system for the file to import by clicking the
Browse ... button.
(Only for PEM files) Declare the type of PEM file to upload:
Public (regular certificate), Private (a private key), or CA (a
CA certificate).
Continued on next page
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Label
Password
7.2.6.2
Continued from previous page
Enter a label for identification of the certificate/key. The filename (base part) will be used as label if left empty. E.g. if
uploaded file name is mycert.p12, the label will be mycert
(Only for PKCS#12 bundles) If your certificate bundle is password protected, you have to enter the password or the import
will fail.
Certificate Details
Menu path: Maintenance ⇒ Certificates ⇒
Label
Common Name (CN)
Certificate Dump
A unique label identifying the certificate.
The common name (CN) part of the distinguished
name (DN) found in the imported certificate subject.
A raw dump of the certificate.
To exit the details page, select a menu option in the navigation menu.
© 2017 Westermo Teleindustri AB
129
Westermo OS Management Guide
Version 4.21.1-0
7.2.7
Ping tool
Ping is useful as a basic diagnostic tool. The output on the web is displayed once
the ping command has completed. If the command takes too long to execute the
web page may time out.
Menu path: Tools ⇒ Ping
Address
Ping Count
Packet Size
130
The network host to send ICMP ECHO REQUEST packets to
Defines the number of ICMP packets to send.
Alters the default size of the ICMP packets.
This only only increases the empty payload
of the packet
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
7.2.8
Traceroute tool
Trace the route packets take to a network host. The output on the web is displayed once the ping command has completed. If the command takes too long
to execute the web page may time out.
Menu path: Tools ⇒ Trace
Address
Maximum Hops
Maximum Wait time
The network host
Max time-to-live (number of hops).
Set the delay, in seconds, before timing out
a probe packet
© 2017 Westermo Teleindustri AB
131
Westermo OS Management Guide
Version 4.21.1-0
7.2.9
IPConfig scan tool
Scan network for IPConfig neighbours. The output on the web is displayed once
the ping command has completed. If the command takes too long to execute the
web page may time out.
Menu path: Tools ⇒ IPConfig
Interface
Flash On LED.
132
The interface to scan
If enabled, this unit will flash the on LED,
while scanning
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
7.2.10
Wake-On-LAN
The Wake-On-LAN (WOL) allows computers to be turned on or woken up by a
network message (magic packet).
Menu path: Tools ⇒ WOL
Interface
MAC Addresses
The interface to send the magic packet on.
The MAC Addresses of the computers to
wake
© 2017 Westermo Teleindustri AB
133
Westermo OS Management Guide
Version 4.21.1-0
7.2.11
Tech support
The Tech support collects system information (hardware, status and configuration) and delivers it as a compressed file. Note: The configuration is included
with passwords. The file format is compressed tar archive(tar.gzip).The filename
has the format of
<LOCATION>_<HOSTNAME>_<YYYYMMDD>_<HHMMSS>.tar.gz, if the location
field is not set, the last three octets of the mac-address will be used.
Menu path: Tools ⇒ Tech Support
Clicking Create will create a Tech support file. Once the file is created you will be
presented with the following dialogue.
The Tech support file consist of a number of text files. Configuration files can be
found in the /cfg directory of the archive, and log files under the /var/log subdirectory.
134
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
© 2017 Westermo Teleindustri AB
135
Westermo OS Management Guide
Version 4.21.1-0
7.3
Maintenance via the CLI
Command
Firmware Upgrade
upgrade <pri|sec|boot>
<IPADDR FILENAME | URI://. . . >
show system-information
System Boot Options
boot
[no] cable-reset
[no] boot-order <flash|bootp>
[no] bootp
[no] timeout <0-1800>
[no] mac <offset <num> |
address <MACADDRESS>>
[no] vfs-target <flash|usb>
[no] console
[no] password-reset
[no] factory-reset
[no] usb
[no] enable
[no] timeout <1-60>
[no] loader
[no] login <password|hash> <STRING>
[no] rescue-port <UDPPORT>
[no] rescue-address <IPADDR>
[no] rescue-netmask <NETMASK>
[no] rescue-peer <IPADDR>
Default
Section 7.3.1
Section 7.3.2
N/A
Enabled
Flash
N/A
300
offset 114
Section
Section
Section
Section
Section
Section
7.3.3
7.3.4
7.3.5
7.3.6
7.3.7
7.3.8
Disabled
N/A
Enabled
Enabled
N/A
Enabled
Disabled
N/A
Disabled
6000
192.168.2.200
255.255.255.0
192.168.2.1
Section
Section
Section
Section
Section
Section
Section
Section
Section
Section
Section
Section
Section
7.3.9
7.3.10
7.3.11
7.3.12
7.3.13
7.3.14
7.3.15
7.3.16
7.3.17
7.3.18
7.3.19
7.3.20
7.3.21
File handling (Configuration, Log, etc.) and Reboot
dir <cfg:// | log:// | usb://>
copy <FROM_FILE> <TO_FILE>
erase <file>
show <running-config | startup-config |
14 See
136
Section
Section 7.3.22
Section 7.3.23
Section 7.3.24
Section 7.3.25
Continued on next page
command description for details and exceptions.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Command
factory-config | [<filesys>://]FILENAME>
backup
restore
reboot
Continued from previous page
Default
Section
Section 7.3.26
Section 7.3.27
Section 7.3.28
Certificate and Key Management
cert import <pkcs|pem|ovpn> [. . . ] <URI>
no cert [force] [LABEL]
show cert [LABEL]
Section 7.3.29
Section 7.3.29
Section 7.3.30
Maintenance and Diagnostic tools
ping <IPADDR>
traceroute <IPADDR>
ssh [[email protected]]<IPADDR|DNAME>[/PORT]
telnet <IPADDR|DNAME> [PORT]
show ipconfig <IFNAME>
wake-on-lan <IFNAME> <MACADDR>
[no] monitor
[no] enable
destination <PORT>
source <PORTLIST>
Section
Section
Section
Section
Section
Section
Section
Section
Section
Section
7.3.31
7.3.32
7.3.33
7.3.34
7.3.35
7.3.36
7.3.37
7.3.38
7.3.39
7.3.40
Section
Section
Section
Section
Section
Section
Section
Section
Section
Section
8.3.7
8.3.5
8.3.8
7.3.41
7.3.42
7.3.43
7.3.44
7.3.45
7.3.46
7.3.47
Other maintenance commands
date [[YYYY-MM-DD ]hh:mm[:ss]]
[no] timezone <TIMEZONE>
show timezone [QUERY|SUBSTRING]
show env
show uptime
show memory
show processes
show flash-table
show partitions
flash-table-update
© 2017 Westermo Teleindustri AB
admin/22
23
Disabled
137
Westermo OS Management Guide
Version 4.21.1-0
7.3.1
Upgrading firmware
Syntax upgrade <pri|sec|boot> <IPADDR> <FILENAME>
upgrade <pri|sec|boot> URI://<ADDRESS>/PATH/<FILENAME>
Context Admin Exec
Usage Upgrade primary, secondary, or bootloader firmware via FTP, TFTP or USB
stick. In the first form, upgrade attempts to download and install FILENAME
via FTP from a server at IPADDR. If no FTP server is available, the command
tries to download the file using TFTP instead.
Note
If you use TFTP for upgrading with ”pkg” files, make sure your TFTP
server supports large files as defined in RFC2347[27].
The second form uses a URI based format. The same format used in the
copy command, not all URIs are supported though, only ftp://, tftp:// and
usb://. In the usb:// case there is of course no need to give an ADDRESS, and
PATH is optional. Also, some units may not have a USB port.
In the second form of the command it is also possible use an Internet name
(FQDN), instead of just an IP address. For this to work you need to have first
setup a valid name server in the configuration.
Before the actual ”Flashing” starts, i.e. when upgrade is still downloading
or checking the downloaded image CRC, it is possible to abort the upgrade
using Ctrl-C (BREAK). However, once the actual flashing starts the BREAK
signal, and other blockable signals, is completely disabled to prevent accidental destruction of the device partition and image contents.
After installing a primary firmware, the switch will automatically be rebooted.
(More precisely: after installing a primary firmware, the switch will automatically be rebooted given that the system booted from the primary image.
Similarly, after installing a secondary firmware, the switch will automatically
be rebooted given that the system booted from the secondary image.)
Caution! Only conduct upgrades over a stable network connection. Ensure
that the switch is not powered off while the downloaded firmware is being
installed.
Default values N/A
Examples ”upgrade primary 192.168.1.1 WeOS-4.15.1.pkg” will download
138
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
and install a new primary image named WeOS-4.15.1.pkg, from FTP/TFTP
server at 192.168.1.1.
”upgrade boot 192.168.1.1 WeOS-4.15.1.pkg” will download and install
a new bootloader image included in the pkg file (WeOS-4.15.1.pkg) from a
FTP/TFTP server with 192.168.1.1.
”upgrade pri usb://WeOS-4.15.1.pkg” upgrades primary firmware on a
WeOS unit using pkg file WeOS-4.15.1.pkg present on a USB stick. Check if
the USB stick has been mounted first using the ”dir usb://” command.
© 2017 Westermo Teleindustri AB
139
Westermo OS Management Guide
Version 4.21.1-0
7.3.2
Show System Information
Syntax show system-information
Context Admin Exec
Usage List general system information such as serial number, firmware version,
contained hardware, etc.
Default values Not applicable
Example
example:/#> show system-information
System Information
===============================================================================
System
System
System
System
Name
Contact
Location
Timezone
Product Family
Architecture
Platform
Article number
Boot loader ver.
Main firmware ver.
Manufacturing date
: example
:
:
: Etc/UTC
:
:
:
:
:
:
:
RedFox
mpc85xx
Corazon
3641-4015
2014.06.0-1
4.15.2
Sep 24, 2014
Model
:
Base MAC Address
:
Class
:
Serial Number
:
Active firmware
:
Backup firmware ver:
RFIR-219-F4G-T7G-AC
00:07:7c:15:5f:20
Extended
1037
Main
4.15.2
Card #1 ======================================================================
Type
: CPU
Chipset
: MV88E6352 r1
Article no
: 5013-1010
Revision
: 0
Batch id
: 140915-01274960-00001
Channel interfaces : 2
Bandwidth limit
: Disabled (for CPU channels)
... (More info follows)
example:/#>
7.3.3
Manage Boot Options
Syntax boot
Context Admin Exec context
Usage Enter System Bootstrap context to configure device specific boot settings. These settings are stored separately, i.e., outside the regular config-
140
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
uration file.
Use ”show boot” to view a summary of the boot option settings.
Default values N/A
Example
example:/#> show boot
Device Bootstrap Configuration ===============================================
Boot order
: flash
Cable reset
: Enabled
Console Functions
Password reset : Enabled
Factory reset : Enabled
USB Functions
Status
Timeout
example:/#>
7.3.4
: Enabled
: Disabled
Enable/Disable Cable Factory Reset
Syntax [no] cable-reset
Context System Bootstrap context
Usage Enable/disable the cable factory reset mechanism.
The cable factory reset mechanism enables you to conduct a factory reset
without using the console port or being logged into the unit. When disabled,
the console port remains the last means to perform a factory reset.
Disabling cable factory reset speeds up the boot and, for some installations,
reduces the risk of inadvertently triggering a factory reset on devices where
only one looped cable is required to perform factory reset.
Use ”cable-reset” to enable and ”no cable-reset” to disable cable factory reset.
Use ”show cable-reset” to show the current setting.
Default values Enabled
© 2017 Westermo Teleindustri AB
141
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/#> boot
example:/boot/#> no cable-reset
example:/boot/#> show cable-reset
Disabled
example:/boot/#> end
example:/#>
7.3.5
Set Boot Order
Syntax [no] boot-order <flash|bootp|usb>
Context System Bootstrap context
Usage Select Boot Order for configuration file15 .
As of WeOS v4.21.1 the ”boot-order” has the following limitations:
ˆ ”boot-order” can only be used to select a single boot media, not a list.
That is, you can select either ”flash” or ”bootp”, but not both.
Note
The WeOS unit will fall-back to find its startup-configuration from
on-board flash when other methods such as ”bootp” fails.
ˆ The alternative ”boot-order usb” (referred to as ”boot from USB”) is
only available as technology preview. See WeOS release notes for more
information on WeOS technology previews in general and for specific
information on the ”boot from USB” function.
Use ”no boot-order” to reset the boot-order to the default setting.
Use ”show boot-order” to view the configured boot order. Flash will listed
as second choice if ”boot-order bootp” is set.
Default values Flash
15 Future
142
versions of WeOS may include support for boot order of software image files.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/#> boot
example:/boot/#>
flash
example:/boot/#>
example:/boot/#>
bootp, flash
example:/boot/#>
example:/#>
7.3.6
show boot-order
boot-order bootp
show boot-order
end
Manage BOOTP Bootstrap Settings
Syntax [no] bootp
Context System Bootstrap context
Usage Enter System Bootstrap BOOTP context to configure settings for BOOTP
boot services.
”no bootp” will reset the BOOTP bootstrap settings to default.
Use ”show bootp” to list BOOTP bootstrap settings (also available as ”show”
command within the System Bootstrap BOOTP context.
Default values N/A
7.3.7
BOOTP timeout
Syntax [no] timeout <0-1800>
Context System Bootstrap BOOTP context
Usage Set timeout in seconds to wait for BOOTP server response.
If no BOOTP response is received from the BOOTP/DHCP server, new BOOTP
Requests will be re-transmitted up to the given timeout interval.
To avoid congestion, the Requests are re-transmitted randomised around
an exponential back-off interval; the back-off interval is doubled for each
request up to 60 seconds.
The BOOTP client will wait one extra back-off interval after the last transmitted request, thus the actual timeout can be roughly 60 seconds longer than
configured.
© 2017 Westermo Teleindustri AB
143
Westermo OS Management Guide
Version 4.21.1-0
Use ”no timeout” to reset the timeout to default.
Default values 300 (seconds)
7.3.8
BOOTP source MAC address
Syntax [no] mac <offset <num> | address <MACADDRESS>>
Context System Bootstrap BOOTP context
Usage Set MAC address for BOOTP request. The source MAC-address used in
BOOTP request can be:
ˆ offset relative to system base MAC: Typically used this if you wish your
product to use a MAC match the MAC of a specific LAN interface on your
unit.
ˆ a statically configure MAC: Assign a specific MAC address to use for
BOOTP for this unit.
By default the source MAC is an offset to system base MAC, which would
match the MAC assigned to interface vlan1. On most WeOS products this
would mean ”mac offset 1” (exceptions are products with more than one
CPU channel; the offset equals the number of CPU channels by default).
Note
See sec. 7.3.2 and 15.4.14 for information on CPU base MAC and CPU
channels. For more information on how a LAN interface is assigned its
MAC address, see section 21.2.4.
Use ”no mac” to reset the BOOTP MAC setting to default.
Use ”show mac” to show the BOOTP MAC setting.
Default values offset 1 (or more generally, the offset equals the number of
CPU channels of the product.)
144
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/#> show iface
Press Ctrl-C or Q(uit) to quit viewer, Space for next page, <CR> for next line.
Interface Name
Oper Address/Length
MTU
MAC/PtP Address
---------------- ---- ------------------ ----- --------------------------lo
UP
127.0.0.1/8
16436 N/A
vlan1
UP
192.168.2.200/24
1500
00:07:7c:84:91:65
-----------------------------------------------------------------------------example:/#> boot
example:/boot/#> bootp
example:/boot/bootp/#> show mac
00:07:7c:84:91:65 (offset 1)
example:/boot/bootp/#>
7.3.9
Storage of BOOTP configuration file (VFS target)
Syntax [no] vfs-target <flash|usb>
Context System Bootstrap BOOTP context
Usage Set virtual file system (VFS) target for configuration file.
Use this setting to save the retrieved file in a non-volatile location. By default
all configuration files retrieved over BOOTP are temporary, and will be lost
when rebooting the system, unless an operator saves a copy with an explicit
”copy running-config cfg://mybackup.cfg” or similar (e.g., Web ’Apply’
or SNMP Set).
Set to ”vfs-target flash” to automatically save to built-in flash (startupconfig) , or ”vfs-target usb” to save to an external USB stick.
Use ”no vfs-target” to disable the setting to get the default behaviour
where the file is stored in RAM only.
Use ”show vfs-target” to show the VFS target setting.
Default values Disabled (i.e., store in RAM only)
7.3.10
Manage Console Settings
Syntax [no] console
Context System Bootstrap context
© 2017 Westermo Teleindustri AB
145
Westermo OS Management Guide
Version 4.21.1-0
Usage Enter System Bootstrap Console context to configure settings related to
the console, or functions only available from the console.
”no console” will reset all console settings to default.
Use ”show console” to list all console settings (also available as ”show”
command within the System Bootstrap Console context.
Default values N/A
7.3.11
Enable/Disable Console Password Reset
Syntax [no] password-reset
Context System Bootstrap Console context
Usage Enable or disable the function to reset the admin user’s password from
the console port.
Use ”no password-reset” to disable the password/reset login.
Use ”show password-reset” to show whether it is enabled or disabled.
Default values Enabled
Example
example:/#> boot
example:/boot/#> show console
Password reset : Enabled
Factory reset : Disabled
example:/boot/#> console
example:/boot/console/#> no password-reset
example:/boot/console/#> show
Password reset : Disabled
Factory reset : Disabled
example:/boot/console/#>
7.3.12
Enable/Disable Console Factory Reset
Syntax [no] factory-reset
Context System Bootstrap Console context
Usage Enable or disable the function to reset the device to factory defaults from
the console port.
146
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Use ”no factory-reset” to disable the factory/reset login.
Use ”show factory-reset” to show whether it is enabled or disabled.
Default values Enabled
Example
example:/#> boot
example:/boot/#> show console
Password reset : Disabled
Factory reset : Enabled
example:/boot/#> console
example:/boot/console/#> no factory-reset
example:/boot/console/#> show
Password reset : Disabled
Factory reset : Disabled
example:/boot/console/#>
7.3.13
Manage USB Bootstrap Settings
Syntax [no] usb
Context System Bootstrap context
Usage Enter System Bootstrap USB context to configure settings for USB boot
services.
”no usb” will reset the USB settings to default.
Use ”show usb” to list configured USB settings (also available as ”show”
command within the System Bootstrap USB context.
Default values N/A
7.3.14
Enable/disable USB Bootstrap Services
Syntax [no] enable
Context System Bootstrap USB context
Usage Enable or disable USB bootstrap services.
Use ”no enable” to disable USB bootstrap services: USB automatic backup/restore
© 2017 Westermo Teleindustri AB
147
Westermo OS Management Guide
Version 4.21.1-0
and USB deployment16 . It is still possible to perform manual ”backup” (see
section 7.3.26) and manual ”restore” see section 7.3.27).
Use ”show enable” to show whether USB bootstrap functionality is enabled
or disabled.
Default values Enabled
Example
example:/#> boot
example:/boot/#> show usb
Status
: Enabled
Timeout
: Disabled
example:/boot/#> usb
example:/boot/usb/#> no enable
example:/boot/usb/#> show
Status
: Disabled
Timeout
: Disabled
example:/boot/usb/#>
7.3.15
USB wait timeout
Syntax [no] timeout <1-60>
Context System Bootstrap USB context
Usage Set timeout in seconds for USB stick to settle at boot.
Some USB sticks cannot be accessed immediately at power-up. This setting
can be used to fine tune the time the system waits for a USB stick to settle.
The system bootup time will be prolonged up to the given timeout, unless
the system discovers the USB stick before.
Default values Disabled (no timeout)
16 ”no enable” also disables the technology preview feature ”boot from USB”, see also section 7.3.5
148
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/#> boot
example:/boot/#> usb
example:/boot/usb/#> timeout 10
example:/boot/usb/#> show
Status
: Enabled
Timeout
: 10 second(s)
example:/boot/usb/#> leave
example:/#>
7.3.16
Manage bootloader settings (Barebox)
Syntax [no] loader
Context System Bootstrap context
Usage Enter System Bootloader context to configure settings related to the
(Barebox) bootloader boot-menu. (You enter the boot-menu by pressing
Ctrl-C on the console port when a unit boots.
Note
The System Bootloader context is only available for products running
the Barebox bootloader.
”no loader” will reset all bootloader settings to default.
Use ”show loader” to list all bootloader settings (also available as ”show”
command within the System Bootloader context.
Default values N/A
Example
example:/boot/#> show loader
Device Bootloader Configuration:
Login Password: Disabled
Rescue Mode Settings:
Address: 192.168.2.200
Netmask: 255.255.255.0
Peer
: 192.168.2.1
Port
: 6000
example:/boot/#>
© 2017 Westermo Teleindustri AB
149
Westermo OS Management Guide
Version 4.21.1-0
7.3.17
Setting boot-menu password (Barebox)
Syntax [no] login <password|hash> <STRING>
Context System Bootloader context
Usage Configure a boot-menu login password. Setting a boot-menu password
is recommended to improve security. When a password is configured, a
user must provide the correct password to enter the boot-menu at system
bootstrap.
When setting the password, you can either enter it as is (”login password
<STRING>”), or provide a SHA1 hash of the password (”login hash <STRING>”).
Use ”no login” to disable the boot-menu login password.
Use ”show login” to see if a boot-menu login password is set or not.
Default values Disabled (no login)
Example
example:/boot/loader/#> login password TopSecret
example:/boot/loader/#> end
Saving bootloader configuration to FLASH
100% / [====================================================================]
example:/boot/#>
7.3.18
Setting rescue console UDP port (Barebox)
Syntax [no] rescue-port <UDPPORT>
Context System Bootloader context
Usage Configure UDP port for rescue-mode netconsole, e.g., ”rescue-port 12345”.
This is used as the local and remote port number for the UDP rescue console.
Defaults to UDP port 6000.
Use ”no rescue-port” to reset UDP port to the default (6000). Use ”show
rescue-port” to show the configured UDP port.
Default values 6000
150
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
7.3.19
Setting rescue console local IP address (Barebox)
Syntax [no] rescue-address <IPADDR>
Context System Bootloader context
Usage Configure local IP address for rescue-mode netconsole, e.g., ”rescue-address
10.0.1.1”. This is used as the local IP for rescue console. Defaults to address 192.168.2.200.
This address is also used as default local IP address when selecting TFTP
boot-image download (technology preview) within the boot-menu (at startup).
Use ”no rescue-address” to reset local IP for rescue console to 192.168.2.200.
Use ”show rescue-address” to show the configured address.
Default values 192.168.2.200
7.3.20
Setting rescue console netmask (Barebox)
Syntax [no] rescue-netmask <IPADDR>
Context System Bootloader context
Usage Configure local IP address netmask for rescue-mode netconsole, e.g.,
”rescue-netmask 255.255.0.0”. Defaults to netmask 255.255.255.0.
Use ”no rescue-netmask” to reset netmask for rescue console interface
to 255.255.255.0 Use ”show rescue-netmask” to show the configured netmask.
This netmask is also used as default rescue interface netmask when selecting TFTP boot-image download (technology preview) within the boot-menu
(at startup).
Default values 255.255.255.0
7.3.21
Setting rescue console peer IP address (Barebox)
Syntax [no] rescue-peer <IPADDR>
Context System Bootloader context
© 2017 Westermo Teleindustri AB
151
Westermo OS Management Guide
Version 4.21.1-0
Usage Configure peer IP address for rescue-mode netconsole, e.g., ”rescue-peer
10.0.1.2”. This is used as the peer IP for rescue console. Defaults to address 192.168.2.1.
This address is also used as default peer IP address when selecting TFTP
boot-image download (technology preview) within the boot-menu (at startup).
Use ”no rescue-peer” to reset local IP for rescue console to 192.168.2.1.
Use ”show rescue-peer” to show the configured address.
Default values 192.168.2.1
7.3.22
List Configuration and Log Files
Syntax dir [<cfg:// | log:// | usb://>]
Context Admin Exec
Usage List files in the configuration file directory, log file directory, or files on a
mounted USB memory. When listing configuration files you should be able
to see which of the present configuration files that is used as startup file. To
map a different configuration file as startup configuration, see the ”copy”
command (section 7.3.23).
Default values cfg://
Example
example:/#> dir
==============================================================================
Contents of Config File System
==============================================================================
config0.cfg --> startup-config
config1.cfg
example:/#>
7.3.23
Copy, Store, Restore or Paste Files
Syntax copy <FROM_FILE> <TO_FILE>
Several methods are available to specify <FROM_FILE> and <TO_FILE>. Local file access methods are listed below:
ˆ Configuration files (default): ”cfg://<FILENAME>”
152
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
ˆ Special configuration files: ”console”, ”running-config”, ”startup-config”,
and ”factory-config”.
ˆ Log files: ”log://<FILENAME>”
ˆ USB memory: ”usb://[DIRECTORY/]<FILENAME>”
Remote file access methods:
ˆ TFTP: ”tftp://location[/directory]/filename”
ˆ FTP: ”ftp://[username[:password]@]location[:PORT]
[directory]/filename”
If no username is provided, anonymous ftp login will be used. Default
password is ”[email protected]”.
ˆ SCP: ”scp://[[email protected]]location[:PORT][/directory]
/filename”
By default username ”admin” will be used.
ˆ HTTP: ”http://location[:PORT][/directory]/filename”
Context Admin Exec
Usage Copy files, save config, transfer to/from network locations. Copy localto-local, local-to-network and network-to-network. Special files are console,
running-config, startup-config and factory-config.
The variant ”copy <FROM> startup-config”, where ”FROM” is a file of the
form ”configN[.cfg]” or ”cfg://file.cfg”, changes which configuration
file is used as the startup-config. In effect only changing which file startupconfig points to. The contents of the previous file it pointed to remains
untouched.
This also means that you can not copy a file directly to startup-config from
any VFS. I.e., when copying a file from (T)FTP or USB you must first copy the
file to a configN[.cfg] file in the cfg:// VFS.
Please note, the use of the special file ”console” is very similar to the old
DOS style usage. Albeit limited to the usage: ”copy console <FILE>”.
When issuing this command you are presented with a paste area where
you can safely type in or paste parts of, or full, configuration files. However,
when pasting in partial ”.cfg” file snippets the system will use WeOS defaults
for unspecified settings.
Also, the destination file in ”copy console <FILE>” cannot be the console
© 2017 Westermo Teleindustri AB
153
Westermo OS Management Guide
Version 4.21.1-0
itself or factory-config, which is read-only. Hence we recommend using:
”copy console config<N>” or ”copy console running-config”.
Default values N/A
Examples
1. Restore factory default (to running configuration)
Example
example:/#> copy factory-config running-config
Using default factory.cfg found in firmware image.
Stopping Syslog daemon ..................................... [ OK ]
Starting Syslog daemon ..................................... [ OK ]
example:/#>
2. Store running configuration to startup configuration
Example
example:/#> copy running-config startup-config
example:/#>
3. Copy configuration file from USB to local configuration file config3.
Example
example:/#> copy usb://myconfig.cfg config3
Copying myconfig.cfg to config3 ...
Done.
example:/#>
4. Copy configuration file onto remote server using FTP.
Example
example:/#> copy cfg://config0.cfg ftp://mylogin:[email protected]/myconfig
example:/#>
7.3.24
Delete a Configuration File
Syntax erase [filesys://]<FILENAME>
filesys can be ”cfg”, ”log”, or ”usb”, with ”cfg” as default.
Context Admin Exec
154
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Usage Delete a configuration file, log file or a file on a mounted USB memory.
Default values ”cfg” is the default file system.
Example
example:/#> dir
==============================================================================
Existing Configurations on System
==============================================================================
config0 --> startup-config
config1
example:/#> erase config1
example:/#> dir
==============================================================================
Existing Configurations on System
==============================================================================
config0 --> startup-config
example:/#>
7.3.25
Show Configuration File (or other files)
Syntax show <running-config|startup-config|factory-config|
[<filesys>://]<FILENAME>
filesys can be ”cfg”, ”log”, or ”usb”, with ”cfg” as default.
Context Admin Exec
Usage Show content of a configuration file, log file, or file on a mounted USB
memory. Special files are running-config, startup-config and factory-config.
Use the ”dir” command to list files (section 7.3.22).
Default values ”cfg” is the default file system.
7.3.26
Activate Auto-Backup
Syntax backup (applicable on units with USB port)
Context Admin Exec
Usage This command activates WeOS automatic backup and restore for USB
media. The directory ”/usb/westermo/backup” is used for this purpose.
See section 7.1.6 for details.
© 2017 Westermo Teleindustri AB
155
Westermo OS Management Guide
Version 4.21.1-0
Default values Not applicable.
7.3.27
Manual Restore from USB
Syntax restore (applicable on units with USB port)
Context Admin Exec
Usage Force restore from USB to running-config.
This command can be used to force an auto-restore of backup files from a
USB stick to ”cfg://” and also activate the new startup-config in the system
running-config.
See section 7.1.6 for details.
Default values Not applicable.
7.3.28
Rebooting the Device
Syntax reboot
Context Admin Exec
Usage Reboot the device. The switch will boot up with its startup-config.
Default values Not applicable.
7.3.29
Import Certificate/Key
Syntax (for PKCS#12)
cert import pkcs [password <PASSWORD>] <URI> [label <LABEL>]
Syntax (for PEM)
cert import pem type <private|public|ca> <URI> [label <LABEL>]
Syntax (for OpenVPN key)
cert import ovpn <URI> [label <LABEL>]
Context Admin Exec
156
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Usage Import PKCS#12 certificate bundle, individual certificate files in PEM format, or an OpenVPN static key. An optional label name can be specified. By
default the label name is set from the file name.
Examples:
ˆ ”cert import pkcs password "secret" ftp://1.2.3.4/bundle.p12”
ˆ ”cert import pem type public usb://remote.crt”
ˆ ”cert import ovpn ftp://1.2.3.4/tls-auth.key label tls”
To remove/delete a certificate by label, use ’force’ to avoid questions:
ˆ ”no cert remote” (Remove certificate file with label ”remote”. There
can be different certificate files (of different types) with the same label.
If so, a separate question will be asked for each file before removal.)
ˆ ”no cert force remote”
Default values Not applicable.
7.3.30
List and show details of Certificates
Syntax show cert [LABEL]
Context Admin Exec
Usage List all certificates, or show details of a specific certificate.
Example to show all certificates, or display/dump a given label:
ˆ ”show cert” (lists all certificates)
ˆ ”show cert remote” (list details of certificate with label ”remote”. There
can be different certificate files (of different types) with the same label.
Then all are shown.
Default values Not applicable.
7.3.31
Ping
Syntax ping [-i <IFACE|IPADDR>] [-s <size>] [-c <count>] [-t <TTL>]
[-M <hint>] <HOST>
Context Admin Exec context
© 2017 Westermo Teleindustri AB
157
Westermo OS Management Guide
Version 4.21.1-0
Usage Ping a remote host.
Ping is useful as a basic diagnostic tool.
The -i option can be used to select the interface to send ICMP_ECHO on,
which is useful in, e.g., VPN setups. The -i option can also be used with an
IP address to spoof the source IP address.
The -M option is used to control where to set the DF (don’t fragment) bit in
the ICMP packet. If this bit is set, no one will be allowed to fragment this
packet and an error will be generated if the packet is to big to fit in the MTU.
Valid options for hint:
ˆ do: Set the don’t fragment bit, prohibit all fragmentation.
ˆ dont: Never set the don’t fragment bit.
ˆ want: Make a MTU discovery and fragment packet if it is too large to fit
in the MTU.
You can use use the domain name or IP address as the host argument, but
you need a valid name server setup for domain names to work, see section 21.7.5.
Default values Not applicable.
Example
example:/#> ping 192.168.131.1
Ctrl-C to abort PING 192.168.131.1
64 bytes from 192.168.131.1: seq=0
64 bytes from 192.168.131.1: seq=1
64 bytes from 192.168.131.1: seq=2
64 bytes from 192.168.131.1: seq=3
(192.168.131.1): 56 data bytes
ttl=64 time=4.832 ms
ttl=64 time=0.836 ms
ttl=64 time=0.810 ms
ttl=64 time=0.823 ms
--- 192.168.131.1 ping statistics --4 packets transmitted, 4 packets received, 0% packet loss
round-trip min/avg/max = 0.810/1.825/4.832 ms
example:/#>
7.3.32
Traceroute
Syntax traceroute <HOST>
Context Admin Exec context
Usage Trace the path the packets take to a remote host.
158
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Traceroute is useful as a basic diagnostic tool.
You can use use the domain name or IP address as the host argument, but
you need a valid name server setup for domain names to work, see section 21.7.5.
Default values Not applicable.
Example
example:/#> traceroute 192.168.130.41
traceroute to 192.168.130.41 (192.168.130.41), 30 hops max, 40 byte packets
1 192.168.131.1 1.116 ms 0.755 ms 0.806 ms
2 192.168.130.41 0.824 ms 0.705 ms 0.742 ms
example:/#>
7.3.33
Remote Login to another device (SSH Client)
Syntax ssh [[email protected]]<IPADDR|DOMAINNAME>[/PORT]
Context Admin Exec context.
Usage Login to remote device using SSH.
Default values Default user ”admin”, default (TCP) port number ”22”.
7.3.34
Remote Login to another device (Telnet Client)
Syntax telnet <IPADDR|DOMAINNAME>[:PORT]
Context Admin Exec context.
Usage Login to remote device using Telnet.
Default values Default (TCP) port number ”23”.
7.3.35
Show IPConfig Neighbours
Syntax show ipconfig [IFNAME]
Context Admin Exec context.
Usage The command has two purposes:
© 2017 Westermo Teleindustri AB
159
Westermo OS Management Guide
Version 4.21.1-0
ˆ Scan the network for IPConfig neighbours on the given interface, i.e.,
scan for other Westermo devices with the IPConfig service enabled (see
section 8.3.21).
ˆ Show status of the IPConfig process on the own device, if enabled.
Note: There is another ”show ipconfig” command available in the Global
Configuration context, which shows IPConfig server configuration settings,
see section 8.3.21.
Default values If no interface is given, a scan for IPConfig neighbours is tried
on interface vlan1 (if existing).
Example
example:/#> show ipconfig
Using default interface vlan1
MAC
IP
Ver. Type
Status
===============================================================================
00:07:7c:87:85:23 192.168.2.100/24
4.03 Lynx+
-------------SI
00:07:7c:87:85:13 192.168.2.200/24
4.03 Lynx+
------------RSI
00:07:7c:87:57:a3 192.168.2.201/24
4.03 Lynx+
FOC:RING:MN:RSI
00:07:7c:87:85:d3 192.168.2.225/24
4.03 Lynx+
MEM:RING:MN:RSI
===============================================================================
Process ipconfigd running as PID 475
example:/#>
Explanations to the output:
ˆ MAC: The base MAC address of the discovered device.
ˆ IP: The IP address of the discovered device.
ˆ Version: Software version on the discovered unit. In the example above, all
discovered devices are running some variant of 4.3.x software. The platform
generation number (4) and feature release (03) number are shown, but we
cannot determine if those units are running 4.3.0, 4.3.1 or some other 4.3.x
patch revision.
ˆ Type: The type of Westermo device discovered.
ˆ Status:
– If FRNT is enabled, the role is displayed as ”FOC” (focal point) or ”MEM”
(member switch), and one can also see whether the FRNT ports are up
or down: ”M” - FRNT port M is up, ”m” - FRNT port M is down, and so
on. Note: the ports ”M” and ”N” refers to the operational state of the
FRNT port, which can differ from their configured role if the ports are
connected in the wrong order (swapped).
160
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
– If RSTP/STP is enabled on the discovered device, the letter ”R” is shown.
– If SNMP is enabled on the discovered device, the letter ”S” is shown.
– If IGMP Snooping is enabled on the discovered device, the letter ”I” is
shown.
7.3.36
Send Wake-On-LAN packet
Syntax wake-on-lan <IFNAME> <MACADDRESS>
Context Admin Exec context.
Usage Send Wake-On-LAN packet on the specified interface to wake a unit with
specified MAC address.
Default values Not applicable.
7.3.37
Manage Port Monitoring
Syntax [no] monitor
Context Admin Exec context
Usage Use the ”monitor” command to enter the Port Monitoring context.
”no monitor” will disable port monitoring (in the same way as ”no enable”
within the Port Monitoring, see section 7.3.38).
Use ”show monitoring” to show port monitoring settings (also available as
”show” command within the Port Monitoring context).
Default values Not applicable.
7.3.38
Enable/disable Port Monitoring
Syntax [no] enable
Context Port Monitoring context
Usage Enable port monitoring. Use ”no enable” to disable port monitoring.
Use ”show enable” to list whether port monitoring is enabled or disabled.
© 2017 Westermo Teleindustri AB
161
Westermo OS Management Guide
Version 4.21.1-0
Default values no enable (disabled)
7.3.39
Set Mirror Port
Syntax [no] destination <PORT>
Context Port Monitoring context
Usage Set the monitor destination port, i.e., the mirror port.
Use ”show destination” to show currently configured port monitoring destination port.
Default values By default there is no destination port.
7.3.40
Set Monitored Ports
Syntax [no] source <PORTLIST> [ingress] [egress]
Context Port Monitoring context
Usage Add/delete/update monitor source port(s), i.e., the ports being monitored.
Use ”show source” to show current set of ports being monitored.
Default values By default there are no source ports. Commands apply both to
”ingress” and ”egress” if neither is specified.
7.3.41
Show System Environment Sensors
Syntax show env
Context Admin Exec context.
Usage List available environment sensors, their index, and their current value.
Examples of sensors are power (DC1 and DC2), Digital In, and Temperature
sensors.
If the uint has a Status Relay, the current status is shown here.
162
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
If the unit is equipped with DDM/DOM capable SFPs17 , the voltage, bias current, Tx power, Rx power and temperature parameters will be listed for each
SFP.
Default values Not applicable.
7.3.42
Show System Uptime
Syntax show uptime
Context Admin Exec context.
Usage Show system uptime.
Default values Not applicable.
7.3.43
Show Memory Usage
Syntax show memory
Context Admin Exec context.
Usage Show system memory usage.
Default values Not applicable.
7.3.44
Show Running Processes
Syntax show processes
Context Admin Exec context.
Usage Show a list of currently running processes.
Default values Not applicable.
17 DDM/DOM diagnostic information is only available for Westermo DDM SFPs, see the SFP
Transceiver Datasheet of your WeOS product (www.westermo.com).
© 2017 Westermo Teleindustri AB
163
Westermo OS Management Guide
Version 4.21.1-0
7.3.45
Show Flash Partition Table
Syntax show flash-table
Context Admin Exec context.
Usage Show information on the flash partition table.
Default values Not applicable.
7.3.46
Show Partition table
Syntax show partitions
Context Admin Exec context.
Usage Show information on the flash partition table. The ”show partitions” is
similar to the ”show flash-table” command (section 7.3.45), but presents
the partition table somewhat differently.
Default values Not applicable.
Examples
ˆ Example with a WeOS unit (Basis platform) with RedBoot boot-loader
(see partition mtd0).
Example
example:/#> show partitions
Partition Name
Size
===============================================================================
mtd0
RedBoot
512.0 KiB
mtd1
Linux_main
12.5 MiB
mtd2
Linux_backup
12.5 MiB
mtd3
JFFS2
4.0 MiB
mtd4
Branding
2.1 MiB
mtd5
RedBoot config
4.0 KiB
mtd6
FIS directory
128.0 KiB
example:/#>
ˆ Example with WeOS unit (Corazon platform) with U-boot boot-loader
(see partition mtd4).
164
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/#> show partitions
Partition Name
Size
===============================================================================
mtd0
Linux_main
56.0 MiB
mtd1
Linux_backup
56.0 MiB
mtd2
Config
15.0 MiB
mtd3
U-Boot Config
512.0 KiB
mtd4
U-Boot
512.0 KiB
example:/#>
ˆ Example with WeOS unit (Corazon platform) with Barebox boot-loader
(see partition mtd4).
Example
example:/#> show partitions
Partition Name
Size
===============================================================================
mtd0
Linux_main
56.0 MiB
mtd1
Linux_backup
56.0 MiB
mtd2
Config
15.0 MiB
mtd3
BareboxEnv
512.0 KiB
mtd4
Barebox
512.0 KiB
example:/#>
7.3.47
Update Flash Partition Table
Syntax flash-table-update
Context Admin Exec context.
Usage This command is used to update the flash partition table on early RedFox
units, in order to allow firmware upgrades to WeOS release 4.3.0 or later.
For details, see section 7.1.10.
Default values Not applicable.
© 2017 Westermo Teleindustri AB
165
Westermo OS Management Guide
Version 4.21.1-0
Chapter 8
General System Settings
This chapter describes functionality related to the general behaviour of the WeOS
unit. In particular it concerns settings for managing the unit, e.g., enabling/
disabling the unit’s Web server, or controlling remote CLI access by enabling/
disabling the SSH or Telnet server.
This chapter also covers settings for the unit’s identity, location, time, etc. System hostname, location and contact correspond to the associated system objects
of the original MIB-2 standard MIB (RFC 1213). For more information on WeOS
SNMP support, see chapter 6.
8.1
Overview of General System Features
Feature
Web
System Identity and Time
System Hostname
X
System Location
X
System Contact
X
System Time Zone
X1
System Date/Time
X
CLI
X
X
X
X
X
Controlling Management Services
Manage LLDP service
X
X
166
General Description
Section 8.1.1
Continued on next page
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Feature
Manage Web server
Manage IPConfig service
Manage SSH server
Manage Telnet server
Manage SNMP server
Other System settings
CPU bandwidth limitation
Continued from previous page
Web CLI General Description
X
X
Section 8.1.2
X
X
X
X
X
(See chapter 6)
X
Section 8.2 covers management of general system settings via the Web interface,
and section 8.3 describes the corresponding features in the CLI.
8.1.1
Managing LLDP
The Link Layer Discovery Protocol (LLDP) is a standardised layer 2 protocol (IEEE
802.1AB[16]), which advertises information about the device itself and its capabilities to other devices within a LAN. The LLDP protocol also advertises from
which port the LLDP packet was sent. This enables the unit to build up a local
view of the remote ports on neighbour devices it is connected to for each local
port. This information is then stored in an SNMP MIB (LLDP MIB[16]), which can
be used by NMS-systems to draw a topology map of the network.
Examples of information advertised by LLDP:
ˆ Remote port number
ˆ Port capabilities
ˆ IP address (see note below)
ˆ Hostname
ˆ MAC-address
ˆ VLAN ID
1 Web configuration of System Time Zone is done as part of the Network settings, see section 21.5.
© 2017 Westermo Teleindustri AB
167
Westermo OS Management Guide
Version 4.21.1-0
In WeOS, LLDP frames are by default advertised every 30 second (configurable).
If an interface stops receiving frames, the neighbour information is expired after
4 advertisement intervals (configurable on sender side).
Note
The advertised IP address is an address of the ports default VLAN, see section 15.1.2.
Note
LLDP can be enabled/disabled globally as well as on a per port basis. As
of WeOS v4.21.1, LLDP support is limited to RxTx and Disabled modes, i.e.,
when enabling LLDP on a port it is set both for transmission and reception
LLDP frames.
8.1.2
Managing Web Server
In WeOS the Web server runs the Web GUI. Configurable Web server settings are:
ˆ Session Timeout: A user logging in via the Web GUI will automatically be
logged out if inactive for this period of time. The timeout can be disabled.
Default: 10 min
ˆ HTTP port: The TCP port the Web Server listens to for HTTP.
Default: 80
ˆ HTTPS port: The TCP port the Web Server listens to for HTTPS.
Default: 443
ˆ Custom HTTPS certificate: A unique self-signed HTTPS certificate will be
generated for the Web server unless a custom certificate is used. You may
upload your custom certificate (see section 7.1.8) and apply it to the Web
server using the Custom HTTPS certificate setting, see section 8.2.4 (Web)
and 8.3.20 (CLI).
Default: Disabled (i.e., use generated certificate)
168
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
8.2
Managing General System Settings via the web
interface
Menu path: Configuration ⇒ System ⇒ Identity
Fig 8.1 shows the page where you can set hostname, location and contact information for your switch.
Figure 8.1: Switch identity settings example
Hostname
Location
Contact
A name to identify this unit. Max 64 characters. Valid characters are A-Z, a-z, 0-9, and hyphen (-). The first character
should be alphabetic (A-Z, a-z). Hyphen is not valid as first or
last character.
A description to identify where the unit is located. Max 64
characters. Valid characters are ASCII 32-126. ”Space” (ASCII
32) is not valid as first or last character.
A description identifying whom to contact regarding management of the unit. Max 64 characters. Valid characters are ASCII
32-126. ”Space” (ASCII 32) is not valid as first or last character.
Change the values to appropriate values for your switch and click the Apply
button.
8.2.1
Set System Date and Time
Menu path: Configuration ⇒ System ⇒ Date & Time
© 2017 Westermo Teleindustri AB
169
Westermo OS Management Guide
Version 4.21.1-0
Figure 8.2: Switch date and time settings, NTP server
Current Date/Time
Remote NTP Server
Timezone
170
Shows current date and time. Click the
icon to
manually set date/time .
The IP address of a time server to be used to keep
the units calendar time synchronised. Leave empty
if you do not want to use a time server.
Select a timezone region to get adjusted local time.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
8.2.2
Enable/disable LLDP via the web interface
Menu path: Configuration ⇒ LLDP
Edit
Click this icon to edit LLDP settings
Remove
Click this icon to disable LLDP and remove LLDP settings.
Menu path: Configuration ⇒ LLDP ⇒
Enabled
Tx Interval
Tx Hold Multiplier
Port
Check this box to enable/disable LLDP on the unit.
Set the interval between LLDP advertisement packets.
Set how many times the Tx Interval should be multiplied to calculate the TTL in the LLDP packet.
Uncheck the box to disable LLDP on a specific port.
© 2017 Westermo Teleindustri AB
171
Westermo OS Management Guide
Version 4.21.1-0
8.2.3
Show LLDP Status via the web interface
Menu path: Status ⇒ LLDP
172
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
8.2.4
Manage Web server configuration
Menu path: Configuration ⇒ System ⇒ Web
Fig 8.3 shows the page where you can change session timeout, HTTP port, HTTPS
port and custom HTTPS certificate for the Web server of your switch.
Figure 8.3: Web interface settings example
Session Timeout
HTTP Port
HTTPS Port
Custom certificate
Change the session timeout value. Default 10 min.
Change HTTP port. Default 80.
Change HTTPS port. Default 443.
Select a custom HTTPS certificate. The certificate
must have been previously imported (as described
in section Section 7.2.6.1).
© 2017 Westermo Teleindustri AB
173
Westermo OS Management Guide
Version 4.21.1-0
8.3
Managing General System Settings via CLI
Command
Configure General System Settings
system
[no] hostname <ID>
[no] location <ID>
[no] contact <ID>
[no] timezone <TIMEZONE>
[no] cpu-bandwidth-limit <auto|<64-100000>|
<7700-1488000> fps>
Default
”family”1
(empty)
(empty)
Etc/UTC
Auto
Section
Section
Section
Section
Section
Section
Section
8.3.1
8.3.2
8.3.3
8.3.4
8.3.5
8.3.6
Set date and time
date
Section 8.3.7
View available time zones
system
show timezone [QUERY|SUBSTRING]
Section 8.3.8
LLDP Management
[no] lldp
[no] enable
[no] tx-interval <SEC>
[no] tx-hold-multiplier <VALUE>
[no] port <PORTLIST|all>
[no] enable
Enabled
Enabled
30
4
Enabled2
Show LLDP status
show lldp
8.3.9
8.3.10
8.3.11
8.3.12
8.3.13
8.3.14
Section 8.3.15
Configure/View Management Service Settings
[no] web
[no] session-timeout <TIMEOUT>
[no] port <PORT>
[no] ssl-port <PORT>
[no] certificate <LABEL>
174
Section
Section
Section
Section
Section
Section
Enabled
10 Min
80
443
Section 8.3.16
Section 8.3.17
Section 8.3.18
Section 8.3.19
Section 8.3.20
Continued on next page
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Command
[no] ipconfig
[no] read-only
[no] ssh
[no] telnet
[no] snmp-server
8.3.1
Continued from previous page
Default
Section
Enabled
Section 8.3.21
Disabled Section 8.3.22
Enabled
Section 8.3.23
Disabled Section 8.3.24
Enabled
Section 6.3.1
Manage System Identity Information
Syntax system
Context Global Configuration context
Usage Enter General System Configuration configuration context.
Use ”show system” to show all general system configuration settings (also
available as ”show” command within the General System Configuration).
Default values Not applicable
8.3.2
System Hostname
Syntax hostname <STRING>
Context General System Configuration context
Usage Set system hostname string.
Max 64 characters. Valid characters are A-Z, a-z, 0-9, and hyphen (-). The
first character should be alphabetic (A-Z, a-z). Hyphen is not valid as first or
last character.
”no hostname” resets the hostname to its default value.
Use ”show hostname” to view the configured hostname setting.
1 The default hostname depends on the product family, e.g., Lynx products have default hostname lynx.
2 LLDP is enabled on all LAN ports by default, except for xDSL ports.
© 2017 Westermo Teleindustri AB
175
Westermo OS Management Guide
Version 4.21.1-0
Default values ”family” (The default hostname depends on the product family,
e.g., Lynx products have default hostname lynx.)
8.3.3
System Location
Syntax location <STRING>
Context General System Configuration context
Usage Set system location string.
Max 64 characters. Valid characters are ASCII 32-126. ”Space” (ASCII 32) is
not valid as first or last character.
”no location” resets the location string to its default value (empty).
Use ”show location” to view the configured location setting.
Default values (empty)
8.3.4
System Contact
Syntax contact <STRING>
Context General System Configuration context
Usage Set system contact string.
Max 64 characters. Valid characters are ASCII 32-126. ”Space” (ASCII 32) is
not valid as first or last character.
”no contact” resets the contact string to its default value (empty).
Use ”show contact” to view the configured contact setting.
Default values (empty)
8.3.5
Set System Time Zone
Syntax [no] timezone <TIMEZONE>
Context General System Configuration context.
176
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Usage Set system time zone string.
Note
For information of available time zone settings, see section 8.3.8.
”no timezone” resets the timezone to its default value (Etc/UTC).
Use ”show timezone” to view the configured timezone setting.
Default values Etc/UTC
8.3.6
CPU bandwidth limitation
Syntax [no] cpu-bandwidth-limit <auto|<64-1000000>|<7700-1488000> fps>
Context General System Configuration context
Usage Limit the traffic sent to the CPU in kbit/s or frames per second (traffic
from the CPU is not affected). It is also possible use ISO modifiers k/M/G,
e.g., 256k or 10M as specifiers for kbps and Mbps.
Note
CPU bandwidth limit in frames per second mode is available on all
WeOS products, with exceptions for some RedFox models. On RedFox, the frames per second mode is available for products based on
the ”Corazon” platform, including RedFox Industrial Rack, newer RedFox Rail (RFR-212-FB)[56] and newer RedFox Industrial[54]. But the
frames per second mode is not available on RedFox products based on
the (older) ”Atlas” platform, i.e, RedFox Industrial listed in [53] or older
RedFox Rail (RFR-12-FB).
Set values are rounded off to the nearest possible HW setting.
© 2017 Westermo Teleindustri AB
177
Westermo OS Management Guide
Version 4.21.1-0
Note
Default is ”auto”, which means that system will automatically reduce
CPU bandwidth when critical services are enabled. As of WeOS v4.21.1,
FRNT Ring Bridging or Multi-link Dual-Homing (see chapter 17) are considered critical, but the set of critical services may change in future
WeOS releases.
A user can override the default with ”no cpu-bandwidth-limit” or
any more specific setting (e.g., ”cpu-bandwidth 4M”). However, for
critical services it is recommended leave the default ”auto”.
On units with multiple CPU channels (see section 15.1.6), the setting will
apply for each of the channels..
Use ”no cpu-bandwidth-limit” to disable CPU bandwidth limitation.
Use ”show cpu-bandwidth-limit” to view the configured CPU bandwidth
limit setting.
Default values Auto (”cpu-bandwidth-limit auto”)
8.3.7
Set or show System Date and Time
Syntax date [[YYYY-MM-DD ]hh:mm[:ss]]
Context Admin Exec context.
Usage Set system date and time, or only time.
Use ”show date” to view the current date and time.
Default values If no date or time is given, the current date and time will be
displayed.
Example
example:/#> date 2013-05-31 10:18
Fri May 31 10:18:00 UTC 2013
example:/#> show date
Fri May 31 10:18:09 UTC 2013
example:/#>
178
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
8.3.8
Show System Time Zone
Syntax show timezone [QUERY|SUBSTRING]
Context General System Configuration context.
Usage Show system time zone setting/list available time zones.
When given without any argument (”show timezone”), the configured time
zone setting is presented.
When providing an argument, the available time zone settings matching
that argument is listed, e.g., issuing the command ”show timezone asia”
will list all possible time zone configuration settings for Asia (or more precisely, all available time zones containing the substring ’asia’.) See section 8.3.5 for information of how to set the system time zone.
Default values Not applicable.
8.3.9
Manage LLDP settings
Syntax [no] lldp
Context Global Configuration context.
Usage Enter LLDP Configuration context. Use command ”lldp” to enter the
LLDP Configuration context. The LLDP Configuration context is created if it
does not exist.
Use command ”no lldp” to disable LLDP and remove all configuration.
Use command ”show lldp” to view the current configuration. Alternatively,
you can enter the LLDP Configuration context and run ”show” (see example
in section 8.3.10).
Default values LLDP is enabled by default.
8.3.10
Activate/deactivate LLDP
Syntax [no] enable
Context LLDP Configuration context.
© 2017 Westermo Teleindustri AB
179
Westermo OS Management Guide
Version 4.21.1-0
Usage Activate/deactivate LLDP. Use ”enable” to activate and ”no enable” to
deactivate LLDP configuration. When deactivated, LLDP is disabled, but the
configuration is not removed.
Default values Enabled (i.e., activated)
Example
example:/config/#> lldp
example:/config/lldp/#> enable
example:/config/lldp/#> show
Status: Enabled
tx-multiplier: 4 (DEFAULT)
tx-interval: 30 sec (DEFAULT)
Per port:
Eth 1 enabled
Eth 2 enabled
Eth 3 enabled
Eth 4 enabled
Eth 5 enabled
Eth 6 enabled
Eth 7 enabled
Eth 8 enabled
Eth 9 enabled
Eth 10 enabled
example:/config/lldp/#>
8.3.11
LLDP Transmission Interval
Syntax [no] tx-interval <5-300>
Context LLDP Configuration context.
Usage Tx-interval controls how often an LLDP packet will be sent out to its neighbours, e.g., ”tx-interval 10” sets the interval to 10 seconds.
Use ”no tx-interval” to reset the Tx-interval to default.
Use ”show tx-interval” to show the current Tx-interval setting.
Default values 30 (seconds)
8.3.12
LLDP Transmission Hold Multiplier
Syntax [no] tx-multiplier <2-10>
180
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Context LLDP Configuration context.
Usage Tx-multiplier controls how many times the Tx-interval should be multiplied to calculate the TTL in the LLDP packet, e.g., setting ”tx-multiplier
3” and ”tx-interval 5” implies that transmitted LLDP packets carry a TTL
of 15 (seconds).
Use ”no tx-multiplier” to reset the Tx-multiplier to default.
Use ”show tx-multiplier” to show the current Tx-multiplier setting.
Default values 4
8.3.13
Manage LLDP per port settings
Syntax [no] port <PORTLIST|all>
Context LLDP Configuration context.
Usage Enter LLDP Port Configuration context for a port or a set of ports.
Use ”port <PORTLIST>” to enter LLDP Port Configuration for one or more
ports, or ”port all” to enter LLDP Port Configuration for all (LAN) ports.
As of WeOS v4.21.1 you can use ”no port <PORTLIST>” and ”no port
all” to disable LLDP on some or all ports. This behaviour may change
in future versions of WeOS. To disable LLDP on a port it is therefore recommended to use the ”no enable” command within theLLDP Port Configuration, see section 8.3.14.
Default values Not applicable.
8.3.14
Enable/disable LLDP per port
Syntax [no] enable
Context LLDP Port Configuration context.
Usage Enable/disable LLDP per port. Use ”enable” to enable LLDP on the port,
and ”no enable” to disable LLDP for the port.
As of WeOS v4.21.1 Rx and tx mode can not be set individually. A port can
either be enabled (enables both rx and tx) or disabled. If LLDP is deacti-
© 2017 Westermo Teleindustri AB
181
Westermo OS Management Guide
Version 4.21.1-0
vated globally and later activated again (”see section 8.3.10”) the per
port settings will be remembered.
Default values Enabled (LLDP is enabled on all LAN ports by default. The exception is xDSL ports, where LLDP is disabled by default.)
8.3.15
Show LLDP Status
Syntax show lldp
Context Admin Exec context.
Usage Show LLDP information about neighbouring devices.
Default values Not applicable.
Example
example:/#> show lldp
------------------------------------------------------------------------------LLDP neighbors:
------------------------------------------------------------------------------Interface:
Eth 10, via: LLDP, RID: 1, Time: 0 day, 01:32:31
Chassis:
ChassisID:
mac 00:07:7c:84:d7:44
SysName:
wolverine
SysDescr:
Wolverine WeOS v4.9.x
MgmtIP:
192.168.2.2
Capability:
Bridge, off
Capability:
Router, on
Capability:
Wlan, off
Port:
PortID:
mac 00:07:7c:84:d7:47
PortDescr:
10/100TX Eth 2/1
VLAN:
1 vlan1
LLDP-MED:
Device Type: Network Connectivity Device
Capability:
Capabilities
Capability:
Policy
Capability:
Location
Capability:
MDI/PSE
Capability:
MDI/PD
Capability:
Inventory
-------------------------------------------------------------------------------
8.3.16
Enable/disable Web Management Interface
Syntax [no] web
182
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Context Global Configuration context.
Usage Enable Web management interface, and enter Web Configuration context. Use ”no web” to disable the Web server.
Warning
Then the switch cannot be managed via the Web interface.
Use ”show web” to list current Web configuration settings (also available as
”show” command within the Web Configuration context).
Default values Enabled (”web”)
8.3.17
Set Web Server Session Timeout
Syntax [no] session-timeout <1-1440>
Context Web Configuration context.
Usage Configures the Web server session timeout in minutes. Use ”session-timeout
<TIMEOUT>” to set the timeout, e.g., ”session-timeout 5” to set the timeout to 5 minutes.
”no session-timeout” disables the session timeout.
”show session-timeout” shows the current timeout setting.
Default values 10 (minutes)
8.3.18
Set Web Server HTTP port
Syntax [no] port <PORT>
Context Web Configuration context.
Usage Configures the HTTP port for the Web server, e.g., ”port 8080” makes
the Web server use listen to HTTP on port 8080.
”no port” resets the Web server to listen to the default port (80).
”show port” shows the current HTTP port setting.
Default values 80
© 2017 Westermo Teleindustri AB
183
Westermo OS Management Guide
Version 4.21.1-0
8.3.19
Set Web Server HTTPS port
Syntax [no] ssl-port <PORT>
Context Web Configuration context.
Usage Configures the HTTPS (SSL) port for the Web server, e.g., ”ssl-port
4343” makes the Web server listen to HTTPS on port 4343.
”no ssl-port” resets the Web server to listen to the default port (443).
”show ssl-port” shows the current HTTPS port setting.
Default values 443
8.3.20
Set custom Web Server HTTPS certificate
Syntax [no] certificate <LABEL>
Context Web Configuration context.
Usage Configures a custom HTTPS certificate (has to be uploaded first). If no
certificate is chosen the default is to use a self-signed certificate which will
be uniquely generated by the device.
”certificate custom1” selects the HTTPS certificate with label ”custom1”
(must have been previously imported, see section 7.3.29).
”no certificate” resets the HTTPS certificate to the default, an automatically generated certificate.
Example
example:/#> show cert
Type
Label
Common Name
Expires
===========================================================
Pub
custom1
device.example.com
Apr 19 2017 GMT
Key
custom1
example:/#> configure
example:/config/#> web
example:/config/web#> certificate custom1
example:/config/web#> leave
Default values Disabled (use generated self-signed certificate)
184
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
8.3.21
Enable/disable IPConfig Service
Syntax [no] ipconfig
Context Global Configuration context.
Usage Enable IPConfig service interface (management of the unit via the Westermo IPConfig protocol), and enter IPConfig Configuration context. Use ”no
ipconfig” to disable the IPConfig server
Warning
After this the switch cannot be managed (or detected) via the IPConfig
protocol, used by the WeConfig tool.
Use ”show ipconfig” to list whether IPConfig is enabled or disabled. Note:
There is another ”show ipconfig” command available in the Admin Exec
context, which is used (1) to scan for neighbour Westermo units, and (2)
to list status information on the IPConfig server running on this device, see
section 7.3.35.
Default values Enabled (”ipconfig”)
Examples
1. How to check whether IPConfig service is enabled on my switch:
Example
example:/#> config
example:/config/#> show ipconfig
Ipconfig is enabled
Read only mode : Disabled
example:/config/#> end
2. How to enable/disable IPConfig service:
Enter Global Configuration context, check the current IPConfig configuration, and modify it if desired. Below is an example of how to disable
IPConfig.
© 2017 Westermo Teleindustri AB
185
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/#> config
example:/config/#> show ipconfig
Ipconfig is enabled
Read only mode : Disabled
example:/config/#> no ipconfig
Deactivating ipconfig service.
example:/config/#> end
8.3.22
Enable/Disable configuration and upgrade via IPConfig service
Syntax [no] read-only
Context IPConfig Configuration context.
Usage The IPConfig service (used by the WeConfig tool) can be used to discover
and view status of a unit, but also for some simple configuration (IP address,
netmask and default gateway) and firmware upgrade (primary firmware).
By setting IPConfig in ”read-only” mode, no configuration or firmware upgrade is possible via IPConfig service.
Use ”show read-only” to list whether ’read-only’ is enabled or disabled.
Use ”read-only” to activate ’read-only’ mode, and ”no read-only” to set
the mode to ’read/write’.
Default values Disabled (”no read-only”, i.e., configuration and upgrading
via IPconfig service is possible.)
Examples How to limit IPConfig service to ’read-only’. That is, disabling configuration and upgrading of the unit via IPConfig, while allowing use of IPConfig
to discover the unit and status information retrieval.
Example
example:/#> config
example:/config/#> show ipconfig
Ipconfig is enabled
Read only mode : Disabled
example:/config/ipconfig/#> read-only
Setting IPconfig read only mode Enabled
example:/config/ipconfig/#> end
186
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
8.3.23
Enable/disable SSH Service
Syntax [no] ssh
Context Global Configuration context.
Usage Enable SSHv2 management service, and enter SSH Configuration context. Use ”no ssh” to disable the SSHv2 server.
Warning
Then the switch cannot be managed via SSHv2.
Use ”show ssh” to list current SSH configuration settings (also available as
”show” command within the SSH Configuration context).
Default values Enabled (”ssh”)
8.3.24
Enable/disable Telnet Service
Syntax [no] telnet
Context Global Configuration context.
Usage Enable Telnet management service, and enter Telnet Configuration context. Use ”no telnet” to disable the Telnet server.
Warning
Then the switch cannot be managed via Telnet.
Use ”show telnet” to list current Telnet configuration settings (also available as ”show” command within the Telnet Configuration context).
Default values Disabled (”no telnet”)
© 2017 Westermo Teleindustri AB
187
Westermo OS Management Guide
Version 4.21.1-0
Chapter 9
AAA - Authentication,
Authorisation and Accounting
This chapter describes WeOS AAA support - Authentication, Authorisation and
Accounting. The AAA configuration gathers authentication methods and policies
in one place and is referenced from other subsystems in WeOS. Four uses of AAA
are currently supported:
ˆ WeOS unit login: The login password to the unit is part of AAA.
ˆ Port-based Network Access Control (PNAC): WeOS supports port access control with IEEE 802.1X and MAC based authentication. This is configured in
two different places, in AAA and as settings related to VLAN. The configuration in AAA specifies RADIUS backends and MAC filtering lists, and the
configuration in VLAN which ports are enabled for port access control. See
section 15.2 for an introduction and guidance to configure port-based network access control in WeOS.
ˆ PPP Peer Authentication: You can create and use local user database lists
to authenticate and authorise your PPP peers. This is typically used for PPP
connections in dial-in/server mode (see chapter 33), but you can also use
this to authenticate and authorise your peer in other PPP modes.
ˆ SSL VPN Client Authentication: When acting as SSL VPN server, a WeOS unit
can require clients to provide username and password (in addition to certificates) as part of the authentication step. The username and password
credentials are then verified using the WeOS AAA framework. See chapter 36 for more information on SSL VPN security.
188
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
9.1
Overview over AAA
Feature
AAA methods
Built-in accounts
Local database
Centralised authentication
Authentication chains
Services using AAA
Login
SSL VPN
PPP
IEEE 802.1X
Port-based Network Access Control
IEEE 802.1X Access
Control Instances
MAC authentication lists
9.1.1
Web
CLI
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
General Description
Section 9.1.1
Section 9.1.1.1
Section 9.1.1.2
Section 9.1.1.3
Section 9.1.1.4
Section 9.1.2
Section 9.1.2.1
Section 9.1.2.2
Section 9.1.2.3
Section 9.1.2.4
Section 15.2
Section 15.2.1
X
X
Section 15.2.2
Methods in the AAA framework
9.1.1.1
Built-in interactive user accounts
The table below lists the built-in interactive user accounts in WeOS. These accounts only apply to the login service, and some of them are limited to the console port.
Name
Purpose
admin
Administrative account for configuring and management of
the device. This is accessible from console, SSH and Web
(HTTP/HTTPS) by default. See additional information below and
section 9.1.2.1.
Password reset function. Only accessible from console. Can be
disabled. See sections 7.1.3.2 and 7.3.11.
Factory reset function. Only accessible from console. Can be
disabled. See sections 7.1.3.2 and 7.3.12.
password
factory
© 2017 Westermo Teleindustri AB
189
Westermo OS Management Guide
Version 4.21.1-0
Note
Accessing the device using the admin account allows full access to the device although the default shell is the WeOS CLI to simplify device management.
The factory default setting for the built-in admin account is as follows:
ˆ Login: admin
ˆ Password: westermo
The password of the built-in admin user can (and should!!) be changed.
Example
example:/#> configure
example:/config/#> aaa
example:/config/aaa/#> username admin YourSecr3t
example:/config/aaa/#> leave
Configuration activated. Remember "copy run start" to save to flash (NVRAM).
example:/#>
The admin account password can be at most 64 characters long. Printable ASCII1
characters except ”space” are allowed in the password, i.e., ASCII 33-126 are
allowed.
Section 7.1.3 provides information on how to proceed in case you forget the builtin admin account password.
9.1.1.2
Local Databases (User Authentication Lists)
Local user databases are useful for storing authentication credentials with no
need for any external infrastructure. The lists consist of username and password pairs, which are stored in plain text. (Future releases of WeOS may include
support for storing hashed passwords in the local databases.)
When a local database is created, a numeric ID is associated with it. This ID
will be used when mapping this database to, e.g., the login service. Additionally, a description string may be associated with the instance to make it easier to
remember its purpose, e.g., ”maintainers” for a local database of system maintainers.
1 American Standard Code for Information Interchange (ASCII), see e.g. http://en.wikipedia.
org/wiki/ASCII (accessed May 2009).
190
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/config/aaa/#> local-db 1
Creating new local db 1
example:/config/aaa/local-db-1/#>
example:/config/aaa/local-db-1/#>
example:/config/aaa/local-db-1/#>
example:/config/aaa/local-db-1/#>
example:/config/aaa/#>
description maintainers
username alice AliceSecreT
username bob BobSeCret
end
Usernames can be at most 63 characters long. Alphanumerical characters (a-z,
A-Z, and 0-9), underscore (_), dash (-) and period (.) are allowed. A username
cannot start with dash (-).
Passwords can be at most 64 characters long. Printable ASCII characters except
”space” are allowed in the password, i.e., ASCII 33-126 are allowed.
9.1.1.3
Centralised Authentication Servers
WeOS supports use of centralised authentication via RADIUS and TACACS+. Using
a central authentication server simplifies account administration, in particular
when you are maintaining a network with many (WeOS) units.
When you define a central authentication server in WeOS, a numeric ID is associated with it. This ID will be used when mapping this server to, e.g., the login
service.
Example
example:/config/aaa/#> remote-server 1
Creating new remote server 1
example:/config/aaa/remote-server-1/#>
example:/config/aaa/remote-server-1/#>
example:/config/aaa/remote-server-1/#>
example:/config/aaa/remote-server-1/#>
example:/config/aaa/remote-server-1/#>
example:/config/aaa/#>
password RADiuSseCret
address 192.168.5.1
type radius
auth-port 1812
end
Note
A remote server of type TACACS+ will use chap as authentication protocol,
which is not configurable.
If you are running two (or more) remote-servers for redundancy purposes (both
servers have identical user databases and are of the same type) you can add
them to a server group in WeOS.
© 2017 Westermo Teleindustri AB
191
Westermo OS Management Guide
Version 4.21.1-0
ˆ If the WeOS unit gets a response from one server (accept or reject), it will
not try the other server(s) in the server group.
ˆ If the request sent to one server times out, it tries to send the request to the
next server. This goes on until a response is received or until all servers in
the groups have been tried.
When you define a server group in WeOS, a numeric ID is associated with it. This
ID will be used when mapping the server group to, e.g., the login service.
Example
example:/config/aaa/#> remote-server 1
Creating new remote server 1
example:/config/aaa/remote-server-1/#> password TACacSseCret
example:/config/aaa/remote-server-1/#> address 192.168.5.1
example:/config/aaa/remote-server-1/#> type tacacs
example:/config/aaa/remote-server-1/#> auth-port 49
example:/config/aaa/remote-server-1/#> end
example:/config/aaa/#> remote-server 2
example:/config/aaa/remote-server-2/#> password TACacSseCret2
example:/config/aaa/remote-server-2/#> address 10.0.1.3
example:/config/aaa/remote-server-2/#> type tacacs
example:/config/aaa/remote-server-2/#> auth-port 49
example:/config/aaa/remote-server-2/#> end
example:/config/aaa/#> server-group 1
Creating new server group 1
example:/config/aaa/server-group-1/#> server 1,2
example:/config/aaa/server-group-1/#> type tacacs
example:/config/aaa/server-group-1/#> end
Note
In case the connection between the WeOS unit and the central authentication server traverses an insecure network, the use of VPN services (see
chapter 32) to protect the authentication communication is strongly recommended.
9.1.1.4
Authentication Chains
WeOS supports use of authentication chains. An authentication chain is an ordered list of authentication methods, and enables you handle more advanced
authentication scenarios. For example, you can create an authentication chain
which first contacts a RADIUS server, and then looks in a local database if the RADIUS server does not respond. As of WeOS v4.21.1, use of authentication chains
is limited to the login service, see also section 9.1.2.
192
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/config/aaa/#> auth-chain 1
Creating new auth-chain 1
example:/config/aaa/auth-chain-1/#> method server 1, local-db 1
example:/config/aaa/auth-chain-1/#> end
example:/config/aaa/#> login
example:/config/aaa/login/#> method auth-chain 1
example:/config/aaa/login/#> end
For an authentication chain, you may select up to MAX_AUTH_CHAIN_METHODS
(4) methods. You can create up to MAX_AUTH_CHAINS (3) chains.
When using an authentication chain, the methods in that chain will be evaluated in the configured order. The exact behaviour differs when the option
”continue-on-reject” is enabled or not.
ˆ Continue-on-reject enabled: (Default) With continue-on-reject enabled, the
chain is evaluated in order and the first method able to verify the user’s
credentials will result in a valid authentication. If a method returns reject
(wrong credentials or user not found) or times out (no response from a remote server or group of servers), the next method in the chain will be tried.
Only when no method in the chain responds ”success”, the result of the
authentication is failure.
Continue-on-reject is enabled by default. This gives a robust behaviour, as
it reduces the risk to you lock yourself out by mistake.
ˆ Continue-on-reject disabled: With continue-on-reject disabled, the chain is
evaluated in order, but stops at the first method responding (success or
reject). Only if a method is unavailable (no response from a remote server
or group of servers), the next method will be tried.
Example
example:/config/aaa/#> auth-chain 1
Creating new auth-chain 1
example:/config/aaa/auth-chain-1/#> method server 1, local-db 1
example:/config/aaa/auth-chain-1/#> no continue-on-reject
example:/config/aaa/auth-chain-1/#> end
The main purpose of disabling continue-on-reject is when you wish to use
a remote server (or server group) as primary authentication method, and
only fall-back to a local database (or built-in accounts for login service) if
the server(s) is unavailable.
Only remote server or server groups can be ”unavailable”. A lookup in a lo-
© 2017 Westermo Teleindustri AB
193
Westermo OS Management Guide
Version 4.21.1-0
cal user database will always respond ”success” or ”reject” (the exception is
empty local databases, which are ignored and therefore implicitly excluded
from the chain). Thus, with continue-on-reject disabled it is not meaningful
to put a remote server after a local database in a chain, as the authentication procedure will never continue passed the local database.
Warning
Be careful when applying an authentication chain with continue-on-reject
disabled, as it increases the risk to lock yourself out by mistake.
In particular, the use of a local database in chains with continue-on-reject
disabled will effectively prohibit fallback to the built-in accounts for the login
service.
Hint
To achieve high security for authentication, it is recommended to ensure
strong credentials on your accounts (including the built-in ”admin” account),
rather than disabling ”continue-on-reject” in an authentication chain.
There exist dedicated commands to disable the built-in ”factory” and ”password” accounts, see section 9.1.1.1.
9.1.2
Services utilising the AAA framework
The table below lists the methods available for the different services using the
AAA framework. The built-in account method is limited to the login service on
purpose. The central server, local database and authentication chain methods
also have limited applicability for different services, but these limitations may be
removed in future versions of WeOS.
Service
Login
SSL VPN
PPP
IEEE 802.1X
Central
Server
X
X
Local DB
X
X
X
Built-in
Accounts
X
Auth-Chains
X
X (only RADIUS)
More information on the services using the AAA framework is given is the subsections below.
194
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
9.1.2.1
Login Service
The login service can utilise all methods in the AAA framework (built-in account,
local database and centralised server) for authenticating the user.
ˆ Built-in accounts (implicit): The use of built-in accounts as login method is
implicit. The built-in accounts are tried after any configured login method
(local database, centralised server, or authentication chain; special rules
apply when ”continue-on-reject” is disabled, see section 9.1.1.4).
ˆ Local database, central server or authentication chain (optional): It is possible to configure the use of a local database, central server (single server
or server group), or authentication chain as login authentication method.
When enabled, the specified method is tried first, typically with fallback to
built-in accounts if the method fails (timeout or reject).
These optional methods are disabled by default.
Below an example is shown, where a remote-server configured with ID 1 is used
to authenticate users logging in to the WeOS unit. If the request sent to the
server results in reject or timeout, the WeOS unit falls back to using the built-in
accounts.
Example
example:/config/aaa/#> login
example:/config/aaa/login/#> method server 1
example:/config/aaa/login/#> end
example:/config/aaa/#>
A user which has successfully authenticated will be allowed to login to the unit.
As of WeOS v4.21.1, any user allowed to login will acquire the highest privilege
level2 . That is, in WeOS v4.21.1 users authenticated via a local database or
centralised server will have the same rights as a user logged in using the built-in
admin account.
Details regarding reserved usernames for login is given below:
ˆ Built-in account names: Use of built-in account names (admin, factory, password) in local databases or central servers is discouraged when using those
methods to login to a WeOS unit. In particular, the factory and password
usernames cannot be used to login, but triggers special behaviour when
used from the console port.
2 Future
versions of WeOS may include support for giving different users different authorisation
© 2017 Westermo Teleindustri AB
195
Westermo OS Management Guide
Version 4.21.1-0
ˆ Username ”guest”: As described in the WeOS release notes, support or a
guest user (username ”guest”) is provided as a technology preview. The
guest can be enabled as additional built-in account, and would then get
access to a limited set of diagnostic functions in the CLI (no Web). The builtin guest user is disabled by default, but the limitations of the guest user
applies to any user logging in with username ”guest”.
Note
Logging in with username ”guest” via local databases or central servers
is discouraged in production networks, as it makes use of functionality
provided as technology preview. See WeOS release notes for further
information.
ˆ Other reserved usernames: Logging in with other usernames commonly reserved in Unix systems (”root”, ”nobody”, etc.) is prohibited.
To limit login accessibility you can either restrict from where a user can access
the CLI or Web, or you can even disable some of the services.
ˆ Disabling access: It is possible to disable the Web service (HTTP/HTTPS),
see section 8.3.16 for details. The CLI can be accessed via console, SSH
and Telnet. The CLI cannot be disabled, but it is possible to disable SSH and
Telnet (see section 8.3.23 and 8.3.24), thereby limiting CLI access to the
console. Telnet is disabled by default.
ˆ Restricting access: It is possible to restrict access to Web (HTTP/HTTPS),
SSH and Telnet per network interface, using the WeOS management interface feature, see section 21.2.7. For units running WeOS Extended you can
configure more fine grain access to these services using the WeOS firewall
functionality, see chapter 31.
9.1.2.2
SSL VPN
The WeOS AAA framework can be used to authenticate SSL VPN clients. The
SSL VPN client will authenticate itself using a certificate, and optionally also via
username and password. When using this option, a WeOS unit acting as an SSL
VPN Server, can either use a local database or a central server (or server group)
to verify the client’s username and password.
For more information on SSL VPN in WeOS and the possibility to use username
and password for authentication, see section 36.1.4.1.
196
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
9.1.2.3
Point-to-Point Protocol (PPP)
The WeOS AAA framework can be for authenticating PPP peers. As of WeOS
v4.21.1 only the local database method can be used for PPP authentication.
A PPP peer can be a user connecting via an external modem (and in future releases of WeOS possibly also via a PPPoE, L2TP or PPTP VPN tunnel). The most
common configuration is to require the peer to authenticate itself when the WeOS
device has a server role, but it is also possible to require authentication in a client
configuration.
For more information on PPP support in WeOS, see chapter 33.
9.1.2.4
IEEE 802.1X
The WeOS AAA framework can be for authenticating IEEE 802.1X supplicants. As
of WeOS v4.21.1 only the central server method, of type RADIUS, can be used for
802.1X authentication.
For information on the IEEE 802.1X service in WeOS, see section 15.2.
© 2017 Westermo Teleindustri AB
197
Westermo OS Management Guide
Version 4.21.1-0
9.2
9.2.1
Managing AAA via the web interface
Login Account Management via the Web Interface
Menu path: Maintenance ⇒ Password
In this section the password for the built-in account admin can be changed.
New Password
Repeat New Password
198
Enter the new password for the admin account.
To minimise risk of typing error, enter the new
password for the admin account once again.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
9.2.2
Select Login Method via the Web Interface
It is possible to add a centralised authentication server/group (section 9.2.9) or
local database (section 9.2.3) as login method, in addition to the built-in admin
account.
Menu path: Configuration ⇒ AAA ⇒ Login
Method
Select login method from the drop-down box. Only configured local
databases and servers/groups, of type RADIUS and TACACS+, will
be visible in the box. If Disabled is selected, only the built-in admin
account will be enabled.
© 2017 Westermo Teleindustri AB
199
Westermo OS Management Guide
Version 4.21.1-0
9.2.3
Local User Databases
Menu path: Configuration ⇒ AAA ⇒ Local User DB
The main page for local user databases shows an overview of created databases.
ID
Description
Edit
Delete
New
200
A unique identifier for the local user database.
The users description of this database.
Click this icon to edit the user database. See section 9.2.5 for details.
Click this icon to remove the user database. You will
be asked to acknowledge the removal before it is
actually executed.
Click this button to add a new user database. See
section 9.2.4 for details. You can create at maximum
4 databases.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
9.2.4
New Local User Database
Menu path: Configuration ⇒ AAA ⇒ Local Users DB ⇒ New
ID
Description
The local user database identifier. This is generated
automatically in the web interface and can not be
changed.
Optional.
A user defined description for this
database that will be visible in listings.
After pressing the Apply button, users can be added to the database. See section 9.2.6.
9.2.5
Edit a local user database
Menu path: Configuration ⇒ AAA ⇒ Local Users DB ⇒
See section 9.2.4 for descriptions of the fields on this page.
© 2017 Westermo Teleindustri AB
201
Westermo OS Management Guide
Version 4.21.1-0
9.2.6
Users
Menu path: Configuration ⇒ AAA ⇒ Local Users DB ⇒
The users list is displayed on the edit page for the local user database.
Username
New User
202
A username unique in this database.
Press this button to create a new user in this
database. See section 9.2.7
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
9.2.7
New User
Menu path: Configuration ⇒ AAA ⇒ Local Users DB ⇒
Username
Password
9.2.8
⇒ New User
A username unique in this database.
The password for this user.
Edit User
Menu path: Configuration ⇒ AAA ⇒ Local Users DB ⇒
⇒
(Users table)
See section 9.2.7 for descriptions of the fields on this page.
© 2017 Westermo Teleindustri AB
203
Westermo OS Management Guide
Version 4.21.1-0
9.2.9
Remote Server overview
Menu path: Configuration ⇒ AAA ⇒ Remote Server
The main page for Remote Server shows an overview of configured server groups
and the remote server configurations.
204
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
9.2.9.1
Server Groups in the overview
ID
Description
Type
Servers
Edit
Delete
New Group
9.2.9.2
The group identifier
The user defined name of this group
The type of remote servers in the group (RADIUS or
TACACS+), which can not be mixed.
List of servers included in this group. Each server
is presented by its description name and the server
identifier inside parentheses
Click this icon to edit the RADIUS group. See section 9.2.10 for details.
Click this icon to remove the group. You will be
asked to acknowledge the removal before it is actually executed. Removing a group will not remove
the config of the included servers.
Click this button to add a new group. See section 9.2.11 for details. You can create at maximum
2 groups.
Remote servers in the overview
ID
Description
Type
Address
Auth Port
Edit
Delete
New Server
The remote server identifier
The user defined name on this server
The type of remote servers (RADIUS or TACACS+).
IP or FQDN of the server
The port used for authentication
Click this icon to edit the remote server setting. See
section 9.2.12 for details.
Click this icon to remove the remote server setting.
You will be asked to acknowledge the removal before
it is actually executed.
Click this button to add a new remote server configuration. See section 9.2.13 for details. You can
define at maximum 6 remote server configurations.
© 2017 Westermo Teleindustri AB
205
Westermo OS Management Guide
Version 4.21.1-0
9.2.10
Edit a server group
Menu path: Configuration ⇒ AAA ⇒ Remote Server ⇒
ID
Description
Type
Servers
206
(Server Groups)
The group identifier. This is generated automatically
in the web interface and can not be changed.
Optional. A user defined name for this group that
will be visible in listings.
The group type (RADIUS or TACACS+).
Remote servers that are included in this group. The
order of this list is important as it defines the order
that servers are queried. Select a server in the dropdown list and add it by clicking the plus icon. Only
remote servers of the same type as the group will be
added. Use the
icon to remove a server from the
group. You are limited to max 3 servers per group.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
9.2.11
Add a server group
Menu path: Configuration ⇒ AAA ⇒ Remote Server ⇒ New Group
See section 9.2.10 for descriptions of the fields on this page. You can have at
maximum 2 server groups.
© 2017 Westermo Teleindustri AB
207
Westermo OS Management Guide
Version 4.21.1-0
9.2.12
Edit a remote server
Menu path: Configuration ⇒ AAA ⇒ Remote Server ⇒
ID
Description
Type
Address
Auth Port
Secret
208
(Remote Servers)
The remote server identifier. This is generated automatically in the web interface and can not be
changed.
Optional. A user defined name for this server configuration that will be visible in listings.
The type of remote-servers (RADIUS or TACACS+).
Mandatory. The IP number or Fully Qualified Domain
Name (FQDN) to the remote server
Mandatory. The port number for server authentication requests. The default and standardised
port number for this is 1812 for RADIUS and 49 for
TACACS+ but can be changed here if needed. If port
number 0 is entered, the standardised port number
will be configured.
Optional. A shared secret (password) that should be
used to encrypt the communication with this server.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
9.2.13
Add a remote server
Menu path: Configuration ⇒ AAA ⇒ Remote Server ⇒ New Server
See section 9.2.12 for descriptions of the fields on this page. You can have at
maximum 6 remote server configurations.
© 2017 Westermo Teleindustri AB
209
Westermo OS Management Guide
Version 4.21.1-0
9.2.14
Authentication Chains Overview
Menu path: Configuration ⇒ AAA ⇒ Auth Chain
The main page for authentication chains shows an overview of created chains.
ID
Description
Methods
Edit
Delete
New
210
A unique identifier for the authentication chain.
The users description of the authentication chain.
An ordered list of the methods in the authentication
chain.
Click this icon to edit the authentication chain. See
section 9.2.16 for details.
Click this icon to remove the authentication chain.
You will be asked to acknowledge the removal before
it is actually executed.
Click this button to add a new authentication chain.
See section 9.2.15 for details. You can create at
maximum 4 authentication chains.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
9.2.15
New Authentication Chain
Menu path: Configuration ⇒ AAA ⇒ Auth Chain ⇒ New
ID
Description
Continue on Reject
Methods
The authentication chain identifier. This is generated automatically in the web interface and can not
be changed.
Optional. A user defined description for this authentication chain that will be visible in listings.
Default Enabled. If enabled continue to next method
if rejected, if disabled stop on reject, only continue
if method unavailable. See section 9.1.1.4 for more
information.
The list of methods. Select method and click the Add
to add the method to the list.
Press the Apply button to store changes.
9.2.16
Edit an authentication chain
Menu path: Configuration ⇒ AAA ⇒ Auth Chain ⇒
Click the pen icon in the overview page to edit a specific chain. See section 9.2.15
for descriptions of the fields on this page.
© 2017 Westermo Teleindustri AB
211
Westermo OS Management Guide
Version 4.21.1-0
9.2.17
IEEE 802.1X authentication
Menu path: Configuration ⇒ AAA ⇒ 802.1X
Here you see a listing of currently configured 802.1X instances.
ID
Enabled
Description
Method
Edit
Delete
New
212
The IEEE 802.1X instance identifier
If this instance is active, A green check-mark means
yes and a dash means no
The user defined name on this IEEE 802.1X instance
The server or group used for authentication, which
needs to be of type RADIUS.
Click this icon to edit the instance See section 9.2.18
for details.
Click this icon to remove the instance. You will be
asked to acknowledge the removal before it is actually executed. Removing an IEEE 802.1X instance
will not remove the referenced RADIUS group or
server.
Click this button to add a new IEEE 802.1X instance.
See section 9.2.19 for details. You can currently only
create one instance.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
9.2.18
Edit an IEEE 802.1X instance
Menu path: Configuration ⇒ AAA ⇒ 802.1X ⇒
ID
Enabled
Description
Method
The IEEE 802.1X instance identifier. This is generated automatically in the web interface and can not
be changed.
Check to enable this instance.
Optional. A user defined name for this instance.
Mandatory. Use this drop-down menu to select a
RADIUS group or a remote RADIUS server. Remote
servers and groups, of type RADIUS, are created
separately. See section 9.2.11 and section 9.2.13.
IMPORTANT: Creating an IEEE 802.1X instance does not in itself activate authentication. Port access is managed in the VLAN configuration. See sections 15.2
and 15.3.4. The instance here must be referenced from the port access configuration for it to be used!
© 2017 Westermo Teleindustri AB
213
Westermo OS Management Guide
Version 4.21.1-0
9.2.19
Add an IEEE 802.1X instance
Menu path: Configuration ⇒ AAA ⇒ 802.1X ⇒ New
See section 9.2.18 for descriptions of the fields on this page. You can currently
only configure one IEEE 802.1X instance.
214
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
9.2.20
MAC based authentication
Menu path: Configuration ⇒ AAA ⇒ MAC Auth
Here you see a listing of currently configured MAC authentication lists.
ID
Enabled
Description
Edit
Delete
New List
The MAC authentication list identifier
If this list is active, A green check-mark means yes
and a dash means no
The user defined name on this MAC authentication
list
Click this icon to edit the list See section 9.2.21 for
details.
Click this icon to remove the list. You will be asked
to acknowledge the removal before it is actually executed.
Click this button to add a new MAC authentication
list. See section 9.2.22 for details. You can create
up to 8 MAC authentication lists.
© 2017 Westermo Teleindustri AB
215
Westermo OS Management Guide
Version 4.21.1-0
9.2.21
Edit a MAC authentication list
Menu path: Configuration ⇒ AAA ⇒ MAC Auth ⇒
ID
Enabled
Description
MAC
The MAC authentication list identifier. This is generated automatically in the web interface and can not
be changed.
Check to enable this list.
Optional. A user defined name for this list.
Optional. A list of MAC addresses and MAC address patterns. Single MAC addresses are specified in the format: HH:HH:HH:HH:HH:HH. A wildcard * can be used at the end of the pattern to
match a block of addresses. Examples: 00:80:C8:*,
00:D8:AA:2C:85:01. Use the drop-down list to select
a port if you want the pattern to only be valid for
requests coming in through a specific port. The description field is optional. Add a pattern by clicking
on the plus icon. Use the
icon to remove a pattern. A list is limited to max 44 addresses/patterns.
IMPORTANT: Creating a MAC authentication list does not in itself activate filtering of addresses. Port access is managed in the VLAN configuration. See sections 15.2 and 15.3.4. The created MAC authentication list must be referenced
from the port access configuration for it to be used!
216
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
9.2.22
Add a new MAC authentication list
Menu path: Configuration ⇒ AAA ⇒ MAC Auth ⇒ New List
See section 9.2.21 for descriptions of the fields on this page.
© 2017 Westermo Teleindustri AB
217
Westermo OS Management Guide
Version 4.21.1-0
9.3
Managing AAA via the CLI
The table below shows AAA management features available via the CLI.
Command
Account management (Login)
aaa
username <USERNAME> [hash]
<PASSWORD>
login
[no] method <local-db <ID> | server <ID> |
server-group <ID> | auth-chain <ID> >
Default
Section
Section 9.3.1
Section 9.3.2
Section 9.3.3
Section 9.3.4
Local User Database Lists (PPP, . . . )
aaa
local-db <ID> [plain]
[no] username <USERNAME><PASSWORD>
[No] description <STRING>
Section 9.3.5
Section 9.3.6
Section 9.3.7
Configure Remote Server Connectors
aaa
[no] remote-server <ID> [type <TYPE>]
type <radius | tacacs>
[no] description <STRING>
[no] address <IP | FQDN>
[no] password <PASSWORD>
[no] auth-port <PORT>
Section
Section
Section
Section
Section
Section
Configure Server Groups
aaa
[no] server-group <GID> [type <TYPE>]
type <radius | tacacs>
[no] description <STRING>
[no] server <ID|ID,ID|ID,ID,ID>
218
radius
1812 | 49
9.3.8
9.3.9
9.3.10
9.3.11
9.3.12
9.3.13
Section 9.3.14
radius
Section 9.3.15
Section 9.3.16
Section 9.3.17
Continued on next page
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Command
Continued from previous page
Default
Section
Configure Authentication Chains
aaa
[no] auth-chain <ID>
[no] method <group <GID>| server <ID>|
local-db <ID>, . . . >
[no] description <STRING>
[no] continue-on-reject
Configure IEEE 802.1X Authentication
aaa
[no] dot1x-auth <ID>
[no] enable
[no] description <STRING>
[no] method <group <GID>|server <ID>>
Configure MAC Authentication Lists
aaa
[no] mac-auth <ID>
[no] enable
[no] description <STRING>
[no] mac match <MAC-PATTERN>
[limit <PORT>] [description <STRING>]
9.3.1
Section 9.3.18
Section 9.3.19
Enabled
Enabled
Enabled
Section 9.3.20
Section 9.3.21
Section
Section
Section
Section
9.3.22
9.3.23
9.3.24
9.3.25
Section 9.3.26
Section 9.3.27
Section 9.3.28
Section 9.3.29
Manage AAA Settings
Syntax aaa
Context Global Configuration context
Usage Enter AAA Configuration context (Authentication, Authorisation and Accounting). The AAA context is used for managing user account settings,
etc.
© 2017 Westermo Teleindustri AB
219
Westermo OS Management Guide
Version 4.21.1-0
Use ”show aaa” to show all configured AAA settings: list the local users and
any configured remote servers, server groups, IEEE 802.1X authentications
and MAC authentications.
Default values Not applicable.
9.3.2
Changing Built-in Account Password
Syntax username <USERNAME> [hash] <PASSWORD>
Context AAA Configuration context
Usage Change password of a certain user account, e.g., the ”admin” account.
By default, the password is entered as clear-text, and saved as a hash.
The ”hash” keyword is not intended to be used by regular users - instead
it is used by the switch itself when reading a configuration file including a
hashed password. By adding the ”hash” keyword, the system expects that
a hashed password is entered (as opposed to a clear-text password).
Use ”show username <USERNAME>” to show the hashed password for the
specified user.
Default values Password is entered in clear-text.
Example Setting the ”admin” password to ”foobar”.
Example
example:/config/aaa/#> username admin foobar
example:/config/aaa/#>
9.3.3
Manage AAA Login Settings
Syntax login
Context AAA Configuration
Usage Enter AAA Login Configuration context. The AAA Login Configuration context is currently only used for managing the login method, see section 9.3.4.
220
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
9.3.4
Manage AAA Login Method
Syntax [no] method <local-db <ID> | server <ID> | server-group <ID>
| auth-chain <ID> >
Context AAA Login Configuration
Usage Select login method. If the method is disabled, only the built-in admin
account (section 9.3.2) can be used to login to the Web/CLI.
ˆ local-db: If a local-db (section 9.3.5) is selected, then all configured
users in that database will be able to log in and get the same privileges
as the built-in admin account.
ˆ Server or Server Group: Select server (section 9.3.8) to use one remote server as authentication backend and select server-group (section 9.3.14) to use a group of remote servers as authentication backend.
Server users will also have the same privileges as the built-in admin
account. Server-group is used for redundancy. All servers inside one
group is supposed to have the same set of user accounts and must be
of the same type. Having different sets of accounts for servers inside
the same server-group is not supported.
ˆ Authentication Chain: Select auth-chain (section 9.1.1.4) to use a chain
of authentication methods, tried in order.
Use ”no method” to disable any configured login method.
Use ”show method” to show the configured login method.
Note
As of WeOS v4.21.1 it is only possible to select one login method, and
it is not possible to disable the built-in admin account. When a login
method is chosen the built-in admin account will still be there, last in
the authentication chain.
Default values Disabled
Examples
ˆ Using a remote server for login authentication:
© 2017 Westermo Teleindustri AB
221
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/config/aaa/#> remote-server 1
Creating new remote server 1
Error: Missing remote server password.
example:/config/aaa/remote-server-1/#> password RADiuSseCret
Error: Missing remote server address.
example:/config/aaa/remote-server-1/#> address 192.168.5.1
example:/config/aaa/remote-server-1/#> type radius
example:/config/aaa/remote-server-1/#> end
example:/config/aaa/#> login
example:/config/aaa/login/#> method server 1
example:/config/aaa/login/#> end
example:/config/aaa/#>
ˆ Using a server group, of type TACACS+, for login authentication:
Example
example:/config/aaa/#> remote-server 1
Creating new remote server 1
Error: Missing remote server password.
example:/config/aaa/remote-server-1/#> password TACacSseCret
Error: Missing remote server address.
example:/config/aaa/remote-server-1/#> address 192.168.5.1
example:/config/aaa/remote-server-1/#> type tacacs
example:/config/aaa/remote-server-1/#> end
example:/config/aaa/#> remote-server 2
Creating new remote server 2
Error: Missing remote server password.
example:/config/aaa/remote-server-2/#> password TACacSseCret2
Error: Missing remote server address.
example:/config/aaa/remote-server-2/#> address 10.0.1.3
example:/config/aaa/remote-server-2/#> type tacacs
example:/config/aaa/remote-server-2/#> end
example:/config/aaa/#> server-group 1
Creating new server group 1
example:/config/aaa/server-group-1/#> server 1,2
example:/config/aaa/server-group-1/#> type tacacs
example:/config/aaa/server-group-1/#> end
example:/config/aaa/#> login
example:/config/aaa/login/#> method group 1
example:/config/aaa/login/#> end
example:/config/aaa/#>
ˆ Using a local user database for login:
222
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/config/aaa/#> local-db 1
Creating new local db 1
example:/config/aaa/local-db-1/#> username alice AliceSecreT
example:/config/aaa/local-db-1/#> username bob BobSeCret
example:/config/aaa/local-db-1/#> end
example:/config/aaa/#> login
example:/config/aaa/login/#> method local-db 1
example:/config/aaa/login/#> end
example:/config/aaa/#>
9.3.5
Manage Local User Database Lists
Syntax [no] local-db <ID> [<TYPE>]
Context AAA Configuration context
Usage Enter Local User Database Configuration context to create, modify or remove a local user database.
Use ”local-db <ID>” to create a local database, or to enter the configuration context of an existing database. ”ID” must be a number greater or
equal to 0 and is referenced from other commands. As of WeOS v4.21.1,
you can specify up to 4 local databases.
An optional ”TYPE” parameter is used to specify how passwords within the
database are stored. The only supported type in the current version of WeOS
is ”plain”, which means that all passwords are stored as plain text.
Use ”no local-db <ID>” to remove a specific database, or
”no local-db” to remove all configured databases.
To list all configured databases, use ”show local-db”.
Default values The ”TYPE” parameter is ”plain” by default.
9.3.6
Add/Delete User in Local Database List
Syntax [no] username <USERNAME> <SECRET>
Context Local User Database Configuration context
© 2017 Westermo Teleindustri AB
223
Westermo OS Management Guide
Version 4.21.1-0
Usage Add or remove users to or from the database.
Use ”username <USERNAME> <SECRET>” to add a new user called ”USERNAME”,
whose password is ”SECRET”.
Use ”no username <USERNAME>” to remove a specific user from the database.
To list all the users in the database, use ”show username”. To show the
credentials of a particular user, use ”show username <USERNAME>”.
Default values Not Applicable.
Examples
Example
example:/config/aaa/local-db-0/#> username alpha foobar
example:/config/aaa/local-db-0/#>
9.3.7
Local Database List Description Setting
Syntax [no] description <STRING>
Context Local User Database Configuration context
Usage Set or remove the local user database description string.
Use ”description <STRING>” to set a description for this database.
Use ”no description” to remove the current description.
Use citation marks around the string if you want to have a description containing space characters.
To view the current description, use ”show description”.
Default values Empty.
Examples
Example
example:/config/aaa/local-db-0/#> description PPPUsers
or ...
example:/config/aaa/local-db-0/#> description ’’PPP Users’’
224
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
9.3.8
Manage Remote Server Connectors
Syntax [no] remote-server <ID> [type <TYPE>]
Context AAA Configuration context
Usage Enter Remote Server Configuration context to create, modify or remove
a server connector.
Use ”remote-server <ID>” to create a new connector, or to enter the configuration context of an existing connector. ”ID” must be a number greater
or equal to 0 and is referenced from other commands. As of WeOS v4.21.1,
you can specify up to 6 server connectors.
An optional ”type” parameter is used to specify the type of server. The
supported types are ”radius” and ”tacacs”.
Use ”no remote-server <ID>” to remove a specific server, or
”no remote-server” to remove all configured servers.
Use ”show remote-server” to list all configured connectors, or
”show remote-server <ID>” to show information on a specific connector.
Default values The ”type” parameter is ”radius” by default.
9.3.9
Set Remote Server Type
Syntax type <TYPE>
Context Remote Server Configuration context
Usage Set the remote server type.
Use this command to specify the type of a remote server connector. The
supported types are ”radius” and ”tacacs”
Use ”show type” to show the configured remote server type.
Default values ”radius”
9.3.10
Configure Remote Server Description
Syntax [no] description <STRING>
© 2017 Westermo Teleindustri AB
225
Westermo OS Management Guide
Version 4.21.1-0
Context Remote Server Configuration context
Usage Set or remove the remote server description string.
Use ”description <STRING>” to set a description for this server or
”no description” to remove the current description.
Use citation marks around the string if you want to have a description containing space characters.
Use ”show description” to show the configured remote server description.
Default values Empty.
Examples
Example
example:/config/aaa/remote-server-0/#> description MyRadius
or ...
example:/config/aaa/remote-server-0/#> description ’’Backup server’’
9.3.11
Configure Remote Server Address
Syntax [no] address <IP | FQDN>
Context Remote Server Configuration context
Usage Set or remove the remote server address.
Use this command to point out the remote server address. You can use an
IP address or a name. Names will be looked up via DNS.
Use ”show address” to show the configured remote server address.
Default values Empty. This will reject authentication for the services using this
server.
Examples
Example
example:/config/aaa/remote-server-0/#> address 1.2.3.4
or ...
example:/config/aaa/remote-server-0/#> address myserver.mydomain.se
226
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
9.3.12
Configure Remote Server Password
Syntax [no] password <PASSWORD>
Context Remote Server Configuration context
Usage Set or remove the remote server password.
Use this command to set the shared secret password to use with this server.
Use ”show password” to show the configured remote server password setting.
Default values Empty. Mandatory to configure.
9.3.13
Configure Remote Server Authentication Port
Syntax [no] auth-port <PORT>
Context Remote Server Configuration context
Usage Set the port number used when communicating with the remote server.
The default port value for authentication requests using RADIUS is 1812 and
49 for TACACS+. ”no auth-port” will reset the value to the standard port
number for the specific server type.
Use ”show auth-port” to show the configured port used for authentication
requests to the server.
Default values 0, automatically changed to the standard port number for the
chosen server type.
9.3.14
Manage Server Groups
Syntax [no] server-group <GID> [type <TYPE>]
Context AAA Configuration context
Usage Enter Server Group Configuration context to create, modify or remove a
server group.
© 2017 Westermo Teleindustri AB
227
Westermo OS Management Guide
Version 4.21.1-0
Use ”server-group <GID>” to create a new group, or to enter the configuration context of an existing group. ”GID” must be a number greater or
equal to 0 and is referenced from other commands.
An optional ”type” parameter is used to specify the type of server. The
supported types in the current version of WeOS are ”radius” and ”tacacs”.
You can specify up to 2 server groups in this version of WeOS.
Use ”no server-group <GID>” to remove a specific group, or
”no server-group” to remove all configured groups.
Use ”show server-group” to list all configured server groups, or ”show
server-group <GID>” to show information on a specific server group (also
available as ”show” command within the Server Group Configuration context).
Default values The ”type” parameter is ”radius” by default.
9.3.15
Set Server Group Type
Syntax type <TYPE>
Context Server Group Configuration context
Usage Set the server group type.
Use this command to specify the type of the servers included in the group.
The supported types in the current version of WeOS are ”radius” and ”tacacs”.
Use ”show type” to show the configured server group type.
Default values ”radius”
9.3.16
Configure Server Group Description
Syntax [no] description <STRING>
Context Server Group Configuration context
Usage Set or remove the server group description string.
Use ”description <STRING>” to set a description for this group or
”no description” to remove the current description. Use citation marks
228
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
around the string if you want to have a description containing space characters.
Use ”show description” to show the configured server group description.
Default values Empty.
Examples
Example
example:/config/aaa/server-group-0/#> description MyGroup
or ...
example:/config/aaa/server-group-0/#> description ’’Backup servers’’
9.3.17
Configure Server Group Members
Syntax [no] server <ID|ID,ID|ID,ID,ID>
Context Server Group Configuration context
Usage Set the server(s) that are included in the server group.
Use this command to specify which servers belong to this server group. You
can specify up to three servers comma separated by their remote server ID.
Each server must be configured separately before the group is set up and
need to be of the same type as the group. Mixing types is not allowed. See
section 9.3.8.
Note
The order of the servers IS important and is used as fall-back order. The
first (leftmost) defined server in the group is queried first. If the first
server returns an error or does not reply the second is queried and so
on.
Use ”show server” to show the configured members of the server group
(listed order is fall-back order).
Default values Empty. This will reject authentication for the services using this
group.
© 2017 Westermo Teleindustri AB
229
Westermo OS Management Guide
Version 4.21.1-0
9.3.18
Manage Authentication Chains
Syntax [no] auth-chain <ID>
Context Authentication Chain Configuration context
Usage Enter Authentication Chain Configuration context to create or modify an
authentication chain.
Use ”auth-chain <ID>” to create a new chain, or to enter the configuration
context of an existing chain. ”ID” must be a number greater or equal to 0
and is referenced from other commands.
Use ”no auth-chain <ID>” to remove a specific chain, or ”no auth-chain”
to remove all configured chains.
Use ”show auth-chain” to list all configured chains, or ”show auth-chain
<ID>” to show information on a specific chain (also available as ”show”
command within the Authentication Chain Configuration context).
Default values Not applicable
9.3.19
Set Authentication Chain Methods
Syntax [no] method <group <GID> | server <ID> | local-db <ID>, ...>
Context Authentication Chain Configuration context
Usage Set the ordered list of authentication methods.
Use this command to specify which methods belong to this chain. The supported types in the current version of WeOS are ”server”, ”group” and
”local-db”. You can specify up to 4 comma-separated methods. Each
method must be configured separately before the chain is set up.
Note
The order of the methods is important, since the methods are queried
in the order defined. The first (leftmost) method in the chain is queried
first. See section 9.1.1.4 for more information on how the methods are
queried.
Use ”show method” to show the configured methods of the authentication
chain.
230
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Use ”no method” to remove configured methods from this chain.
Default values Not applicable
Examples
Example
example:/config/aaa/auth-chain-0/#> method server 1, local-db 1
9.3.20
Configure Authentication Chain Description
Syntax [no] description <STRING>
Context Authentication Chain Configuration context
Usage Set or remove the authentication chain description string.
Use ”description <STRING>” to set a description for this chain or
”no description” to remove the current description. Use quotation marks
around the string if you wish to have a description containing space characters.
Use ”show description” to show the configured authentication chain description.
Default values Empty.
Examples
Example
example:/config/aaa/auth-chain-0/#> description MyChain
or ...
example:/config/aaa/auth-chain-0/#> description ’’The Chain’’
9.3.21
Authentication Chain Continue on Reject
Syntax [no] continue-on-reject
Context Authentication Chain Configuration context
Usage Enable/disable the continue-on-reject functionality. If enabled continue to
next method if rejected. If disabled stop on reject, only continue if a method
© 2017 Westermo Teleindustri AB
231
Westermo OS Management Guide
Version 4.21.1-0
is unavailable. See section 9.1.1.4 for more information.
Default values Disabled
9.3.22
Manage IEEE 802.1X authentication instances
Syntax [no] dot1x-auth <ID>
Context AAA Configuration context
Usage Enter 802.1X Configuration context to create, modify or remove an IEEE
802.1X authentication instance.
Use ”dot1x-auth <ID>” to create a new 802.1X authentication instance,
or to enter the configuration context of an existing instance. (As of WeOS
v4.21.1 you can only create one 802.1X authentication instance.) ”ID” must
be a number greater or equal to 0 and is referenced from other commands.
Important
Creating an IEEE 802.1X authentication instance does not in itself activate authentication. Port access is managed in the VLAN configuration.
See section 15.2. The created 802.1X instance must be referenced
from the port access configuration for it to be used!
Use ”no dot1x-auth <ID>” to remove a specific instance, or ”no dot1x-auth”
to remove all 802.1X instances.
Use ”show dot1x-auth” to list all 802.1X authentication instances, or ”show
dot1x-auth <ID>” to show information on a specific instance (also available as ”show” command within the 802.1X Configuration context).
Default values Not applicable.
9.3.23
Enable/Disable an IEEE 802.1X authentication instance
Syntax [no] enable
Context 802.1X Configuration context
Usage Enable or disable an 802.1X authentication instance.
Use ”no enable” to disable.
232
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Use ”show enable” to show whether this instance is enabled or disabled.
Default values Enabled.
9.3.24
Set IEEE 802.1X authentication instance description
Syntax [no] description <STRING>
Context 802.1X Configuration context
Usage Set or remove the description string for this 802.1X authentication instance.
Use ”description <STRING>” to set a description or ”no description” to
remove the current description. Use citation marks around the string if you
want to have a description containing space characters.
Use ”show description” to show the configured instance description setting.
Default values Empty.
Examples
Example
example:/config/aaa/dot1x-auth-0/#> description My_1X_net
or ...
example:/config/aaa/dot1x-auth-0/#> description ’’Employees only’’
9.3.25
Configure IEEE 802.1X authentication back-end servers
Syntax [no] method <group <GID>|server <ID>>
Context 802.1X Configuration context
Usage Set or remove the back-end method for the 802.1X authentication instance.
IEEE 802.1X commonly use the RADIUS protocol as back-end. A RADIUS
server connection or a server group must be configured separately before
you can use the method command. See sections 9.3.8 and 9.3.14.
Use the syntax ”method group <GID>” to select a specific RADIUS server
group as back-end.
© 2017 Westermo Teleindustri AB
233
Westermo OS Management Guide
Version 4.21.1-0
Use the syntax ”method server <ID>” to select a specific RADIUS server
as back-end.
Use ”no method” to remove the back-end selection setting.
Use ”show method” to show the ID/GID of the configured back-end server
or back-end server group.
Default values No backend. 802.1X authentication attempts will fail.
9.3.26
Manage MAC authentication lists
Syntax [no] mac-auth <ID>
Context AAA Configuration context
Usage Create, modify or remove a MAC authentication list.
Use ”mac-auth <ID>” to create a new list, or to enter the configuration
context of an existing list. ”ID” must be a number greater or equal to 0 and
is referenced from other commands. As of WeOS v4.21.1, you can create up
to 8 MAC authentication lists.
Important
Creating a MAC authentication list does not in itself activate filtering
of addresses. Port access is managed in the VLAN configuration. See
section 15.2. The created MAC authentication list must be referenced
from the port access configuration for it to be used!
Use ”no mac-auth <ID>” to remove a specific list, or ”no mac-auth” to
remove all configured MAC authentication lists.
Use ”show mac-auth” to list all MAC authentication lists, or ”show mac-auth
<ID>” to show information on a specific instance (also available as ”show”
command within the MAC Authentication List Configuration context).
Default values Not applicable.
9.3.27
Enable/Disable a MAC authentication list
Syntax [no] enable
Context MAC Authentication List Configuration context
234
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Usage Enable or disable a MAC authentication list.
Use ”no enable” to disable.
Use ”show enable” to show whether this list is enabled or disabled.
Default values Enabled.
9.3.28
Set MAC authentication list description
Syntax [no] description <STRING>
Context MAC Authentication List Configuration context
Usage Set or remove the description string for this list.
Use ”description <STRING>” to set a description or ”no description” to
remove the current description. Use citation marks around the string if you
want to have a description containing space characters.
Use ”show description” to show the configured list description setting.
Default values Empty.
Examples
Example
example:/config/aaa/mac-auth-0/#> description MyMACList
or ...
example:/config/aaa/mac-auth-0/#> description ’’Trusted MAC addresses’’
9.3.29
Configure MAC authentication list filters
Syntax [no] mac match <MAC-PATTERN> [limit <PORT>]
[description <STRING>]
Context MAC Authentication List Configuration context
Usage Add or remove MAC filter patterns.
A MAC Authentication List is built up by MAC filter patterns. Use the syntax ”mac match <MAC-PATTERN>” to create a new filter pattern. To match
a single MAC address specify the hardware Ethernet MAC in the format
HH:HH:HH:HH:HH:HH as <MAC-PATTERN>. You can also specify whole blocks
© 2017 Westermo Teleindustri AB
235
Westermo OS Management Guide
Version 4.21.1-0
of addresses by using a wild-card * at the end of the pattern. You can also
optionally filter on the port by using the ”limit” argument to this command.
A comment may also be added with the optional ”description” argument.
Use ”no mac match <MAC-PATTERN>” to remove a specific filter, or ”no mac”
to remove all filters in this list.
As of WeOS v4.21.1, you can create up to 44 MAC filter patterns per MAC
authentication list.
Use ”show mac” to show the defined MAC filter rules for this authentication
list.
Default values Empty, no filters.
Examples
Example
mac-auth-0/#> mac match 00:D8:AA:2C:85:01
or with wildcard...
mac-auth-0/#> mac match 00:80:C8:*
or with wildcard, limit filter, and description ...
mac-auth-0/#> mac match 00:D8:BB:C5:* limit 1/2 description ’’Laser printers on 1/2’’
236
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
9.4
Feature Parameters
MAX_AUTH_CHAIN_METHODS
MAX_AUTH_CHAINS
4
3
© 2017 Westermo Teleindustri AB
237
Westermo OS Management Guide
Version 4.21.1-0
Chapter 10
Ethernet Port Management
By default all ports on the switch are enabled. Section 10.1 provides general
information about the available port settings. Section 10.2 covers port settings
via the Web interface and section 10.3 port settings via the CLI.
10.1
Overview of Ethernet Port Management
The table below presents available port settings. The features are presented
further in the following sections.
Feature
Enable/disable port
Speed-duplex mode
Flow control
Port priority (level)
Port priority mode
Link alarm
Inbound rate limit
Rate Selection
Traffic Selection
Outbound traffic shaping
Bandwidth per port
MDI/MDIX
238
Web
X
X
X
X
X
X
X
X
X
X
X
CLI
X
X
X
X
X
X
X
X
X
X
X
X
General Description
Section 10.1.2
Section 10.1.3
Section 10.1.4
Section 10.1.4
Section 10.1.5
Section 10.1.6
-”-”Section 10.1.7
Section 10.1.8
Section 10.1.9
Continued on next page
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Feature
Fastlink
fallback default-VID
PHY fine tuning
Shielded/Unshielded TP cable
TX power mode
Continued from previous page
Web CLI General Description
(X)
(X) Section 10.1.10
X
Section 10.1.11
X
X
X
View port configuration
View port status
View SFP DDM/DOM diagnostics
10.1.1
X
X
X
X
X
X
Section 10.1.12
Port naming conventions
The convention to name communication ports such as Ethernet ports, DSL ports,
and Serial ports differs between WeOS products and product families.
10.1.1.1
Simple numbering
Lynx, Falcon, DDW-x42 (Wolverine), RedFox Industrial Rack, RedFox Rail, and
Viper all use a simple port ID to refer to their ports.
ˆ Lynx[52] and RedFox Industrial Rack[55]: Ethernet ports on Lynx and RedFox
Industrial Rack are named 1, 2, 3, . . .
ˆ Falcon[47], Lynx-DSS[49] and some Wolverine units (e.g., DDW-x42[42]):
These units have multiple port types; Ethernet, serial port(s) and xDSL/SHDSL
(Falcon/Wolverine), which are numbered individually per port type. For example, Falcon is equipped with:
– four Ethernet ports (numbered 1, 2, 3 and 4),
– one xDSL port (numbered 1), and
– one serial port (numbered 1).
As Ethernet and xDSL ports can be used in overlapping contexts, e.g., they
can be associated with the same VLAN, a port qualifier (”eth” or ”dsl”) is
© 2017 Westermo Teleindustri AB
239
Westermo OS Management Guide
Version 4.21.1-0
sometimes used to distinguish Ethernet port 1 (”eth 1”) from xDSL port 1
(”dsl 1”).
ˆ RedFox Rail and Viper: Ethernet ports on RedFox Rail and Viper are named
X1, X2, X3, . . .
10.1.1.2
Slot based numbering
RedFox Industrial[53, 54] and some Wolverine products (DDW-225[45] and DDW226[46]) use a slotted architecture, and ports are named according to the slot ID
and the port’s position within that slot. For example, port 1/2 would denote the
second port in the first slot.
This name convention is used irrespective of port type, e.g., DDW-226 (Wolverine)
has two SHDSL ports (1/1-1/2), 4 Ethernet ports (2/1-2/4), and one Serial port
(1/1). Details on the name convention and the slotted architecture is described
further below, using RedFox Industrial as example.
Figure 10.1: Three-slot RedFox Industrial switch equipped with a 8-port Gigabit/SFP card (middle slot), and an 8-port 10/100BaseTX card (right slot).
The RedFox Industrial switches come in a two-slot and a three-slot version. Fig. 10.1
shows a sample three-slot RedFox Industrial equipped with a 4-port Gigabit/SFP
240
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
card (middle slot) and an 8-port 10/100BaseTX card (right slot). The leftmost slot
contains the Power/CPU card, which is present on all RedFox Industrial switches.
RedFox Industrial makes use of a slotted architecture with different combinations
of interface modules. As mentioned above WeOS numbers the ports based on
slotID/portID, where the
ˆ the slotID denotes the slot’s position within the rack (left to right), and
ˆ the portID denotes the port’s position within the slot (left to right, up to
down).
For example, the three Ethernet ports in the leftmost slot (slot 1) are named 1/1
(top), 1/2 (middle). and 1/3 (bottom). The ports in the second slot are named
2/1-2/4 (left side) and 2/5-2/8 (right side), and ports in slot 3 are named 3/1-3/4
(left side) and 3/5-3/8 (right side).
10.1.2
Port speed and duplex modes
By default ports are configured to auto-negotiate speed (10/100/1000 Mbit/s) and
duplex modes (half/full) to the ”best” common mode when a link comes up. When
configured for auto-negotiation, the resulting speed and duplex mode agreed is
shown as part of the port status information.
It is possible to disable auto-negotiation and instead use a static speed and duplex mode setting. When using a static speed and duplex setting, the operator
should ensure that the ports on both ends of the link are configured with the
same static speed and duplex settings.
Depending on Ethernet port type, the available port speeds will differ:
ˆ Fast Ethernet copper ports: Fast Ethernet copper ports are capable to operate at 10 or 100 Mbit/s.
ˆ Gigabit Ethernet copper ports: Gigabit Ethernet copper ports are capable to
operate at 10, 100 or 1000 Mbit/s.
ˆ Gigabit Ethernet fibre ports: Gigabit Ethernet fibre ports are capable to operate at 1000 Mbit/s.
© 2017 Westermo Teleindustri AB
241
Westermo OS Management Guide
Version 4.21.1-0
10.1.3
Flow control
The ports can be configured to use flow control, i.e., to dynamically limit inbound
traffic to avoid congestion on outbound ports.
When flow control is enabled on a full duplex port, the switch will send pause
frames (IEEE 802.3x) to limit inbound traffic on this port, if that traffic is causing
congestion when sent out on another switch port.
When flow control is enabled on a half duplex port, the switch will use a technique known as back-pressure to limit inbound traffic on this port, if that traffic
is causing congestion when sent out on another switch port. (The back-pressure
technique enables a switch to force its neighbour to slow down by sending jamming signals on that port, thus emulating a packet collision.)
10.1.4
Layer-2 priority support
Each Ethernet port has four output queues, enabling layer-2 priority support with
four traffic classes. The queues are serviced according to strict priority scheduling, i.e., when there is traffic in multiple queues, the packets in the queue with
higher priority is serviced first.
A packet’s priority is determined when it enters on a port, and can be classified
based on:
ˆ VLAN ID: The switch can be configured to give specific priority to certain
VLANs. This can be used, e.g., when providing IP telephony via a dedicated
VLAN. Priority based on VLAN ID has precedence over all priority classifications described below.
VLAN ID priority settings are further described in chapter 15.
ˆ VLAN tag: For packets carrying a VLAN tag, the packet’s priority can be
based on content of the priority bits inside the VLAN tag. The VLAN tag is
useful to carry packet priority information on inter-switch links.
Use of VLAN tag priority can be configured per port (see sections 10.2
and 10.3).
ˆ IP ToS/DiffServ: For IP packets the priority can be classified based on the
content of the IP ToS bits (IPv4) or the IP TC bits (IPv6). Classification based
on the IP ToS/Diffserv bits can be used to provide higher priority to delay
sensitive applications, such as IP telephony and remote login, than to bulk
242
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
data applications, such as file transfer. However, it requires that those applications can set the IP ToS/Diffserv bits appropriately.
Use of IP ToS/DiffServ priority can be configured per port (see sections 10.2
and 10.3).
ˆ Port Priority: Priority can be classified based on the inbound port.
Use of port priority can be configured per port (see sections 10.2 and 10.3).
Furthermore, when priority classification is configured to be based on VLAN
tag (or IP ToS/DiffServ), priority will be based on the port priority for untagged (or non-IP respectively) packets.
When priority is classified based on VLAN ID, VLAN tag, or port priority, the priority assigned to a packet will take a value in range 0-7, and be represented by
3 bits (IEEE 802.1p). The mapping of 802.1p priority (8 values) to traffic class
(4 output queues) is shown in table 10.2. The rationale behind this mapping is
described in Annex G of IEEE 802.1Q-2005[18].
IEEE 802.1p
priority
0
1
2
3
4
5
6
7
Queue number/
Traffic Class
0 (lowest)
0
1
1
2
2
3
3 (highest)
Table 10.2: Mapping of IEEE 802.1p priority to Queue/Traffic Class.
When priority is classified based on IP ToS/DiffServ, the priority assigned to a
packet will take a value in range 0-63, and be represented by 6 bits (DSCP Differentiated Services Code Point). The mapping of DSCP priority (64 values)
to traffic class (4 output queues) is shown in table 10.3. This mapping is in line
with the use of IP Precedence fields (RFC 1349), and IP DiffServ for best effort
and control traffic (RFC 2474), assured forwarding (RFC 2597) and expedited
forwarding (RFC 3246).
Packets sent out on a port with a VLAN tag will carry priority information (802.1p)
within their VLAN tag.
© 2017 Westermo Teleindustri AB
243
Westermo OS Management Guide
Version 4.21.1-0
5
IP Priority
bits
4 3 2 1
0
0
1
1
0
1
0
1
-
-
-
Queue number/
Traffic class
0
Queue
bits
1
0
-
0
0
1
1
0 (lowest)
1
2
3 (highest)
0
1
0
1
Table 10.3: Mapping of IP priority bits to Queue/Traffic Class.
ˆ For packets where priority was classified based on VLAN ID, VLAN tag, or
port priority, the outbound priority (3 bits) will be equal to the determined
inbound priority (3 bits).
ˆ When priority is classified based on IP ToS/DiffServ, determining the outbound priority (3 bits) is more complex: the two most significant bits of the
outbound priority will be equal to the queue number (i.e., queue bits in table 10.3), while the least significant bit of the outbound priority is equal to
the least significant bit of the inbound port’s configured port priority.
E.g., if the packet is put in priority queue 2 (binary ’10’), and the port priority
of the inbound port has an odd value (least significant bit is ’1’), the packet
will carry priority value 5 (’101’) in its VLAN tag when sent on the outbound
port.
Warning
Configuration of layer-2 priority should be handled with care. In particular,
mapping user traffic to the highest priority queue is discouraged, since that
may affect time critical control traffic, such as FRNT traffic, already mapped
to the highest priority queue.
10.1.5
Link alarm
Each Ethernet port on the switch can be configured to indicate alarm when the
link comes up or goes down. The alarm is indicated in multiple ways:
ˆ SNMP trap: An SNMP trap will be sent when a link changes state, i.e., both
when the link comes up, or when it goes down. This assumes that SNMP is
244
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
enabled, and that a trap host is configured. See chapter 6 for more information.
ˆ Front panel LEDs: A link alarm may effect both the individual LED of the port,
as well as the common status LED for the switch (for definite information
about what functions affect the common status LED, see chapter 24):
– Individual LED: Each Ethernet port has a LED, which generally indicates
’green’ if the link is up. If there is no link, the LED will indicate ’yellow’
when link alarm is configured.
– Common status LED: The switch has a common status LED, labelled
’ON’ on the front panel. This LED will generally indicate ’green’ if all
associated functions are OK, and ’red’ if one or more of the associated
alarm sources are ’NOT OK’. E.g., if one of the ports configured with link
alarm indicates link down, the common status LED will be ’red’.
ˆ Web interface: Link alarms (link down) are indicated on the main Web page,
and the port configuration/status page.
ˆ CLI: A link alarm (link down) is indicated by an exclamation mark (’!’) when
displaying the port’s status in the CLI.
ˆ Digital I/O: A link alarm can affect the output level of the status relay in the
same way as it will affect the common status LED.
For more information on the functionality of the Digital I/O port, see chapter 24.
10.1.6
Inbound/Ingress rate limiting
The switch can be configured to limit the rate of a port’s incoming traffic - inbound
rate limiting (also referred to as ingress rate limiting). By default a port will accept
packets at a rate up to the link speed, but with inbound rate limiting activated
the switch will start dropping packets when data arrives above the given rate
threshold.
The inbound rate limiting feature can be useful as a complement to layer-2 priority handling (see section 10.1.4) when congestion within the network is to be
avoided.
There are two configuration settings for inbound rate limiting:
© 2017 Westermo Teleindustri AB
245
Westermo OS Management Guide
Version 4.21.1-0
ˆ Rate: Defines the threshold data rate. The web interface provides a predefined set of rates (drop-down list). The CLI allows for more fine-grained rate
settings:
– in steps of 64 kbit/s in range 64-1000 kbit/s
– in steps of 1 Mbit/s in range 1-100 Mbit/s
– in steps of 10 Mbit/s in range 100-1000 Mbit/s (on Gigabit Ethernet
ports.)
Rate limiting calculations consider the layer-2 bits, i.e., from Ethernet destination MAC address to CRC (interframe gap and preamble bits are not
counted).
ˆ Traffic Type: Defines the kind of traffic subject to inbound rate limiting. By
default, a configured rate limit will apply to all traffic, however, it is possible
to restrain the rate limit to specific (layer-2) traffic types: broadcast, multicast and/or unknown1 unicast. As of WeOS v4.21.1 selection of traffic types
can only be done via the CLI.
1 Unknown unicast traffic is traffic with a unicast destination MAC address not present in the
switch forwarding database (see section 15.4.19). Unknown unicast traffic is flooded onto all ports
within the (V)LAN.
246
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
10.1.6.1
Restrictions on inbound rate limiting
On RedFox units, some of the interface modules have hardware dependent restrictions regarding the inbound rate limit function. These restrictions are described in this section.
Which Ethernet ports on RedFox have the restrictions described here?
The restrictions apply to Ethernet ports of switchcores MV88E6095 and
MV88E6185. Please see Detailed System Overview page in the Web (section 4.4.2) or use the ”show system-information” in the CLI (section 7.3.2)
to find definite information about what switchcore(s) is used in your product.
An informative list of products/modules where the restrictions apply is given
below:
ˆ RedFox Industrial (RFI) with Corazon platform[54]: Only Ethernet ports
on modules ”F4G” and ”F4G-T4G” (MV88E6185) have these restrictions.
ˆ RedFox Industrial (RFI) with Atlas platform[53]: Ethernet ports on all
modules except module ”F8” have these restrictions.
ˆ RedFox Industrial Rack (RFIR)[55]: Only Ethernet ports in the 8-port
group/module with Gbit/s ports (4 Gbit/s SFP and 4 Gbit/s Copper ports;
MV88E6185) have these restrictions.
ˆ RedFox Rail (RFR) with Corazon platform[56]: No Ethernet ports on the
RFR-212 have these restrictions.
ˆ RedFox Rail (RFR) with Atlas platform (not for sale): All Ethernet ports
on the RFR-12 have these restrictions.
ˆ TCP traffic: When the data rate rises above the given threshold on these
Ethernet ports, packets will be dropped in a manner ”punishing” TCP traffic
relatively hard. Thus, activating inbound rate limiting applicable to unicast
traffic may have an undesired impact on your TCP traffic,.
ˆ Traffic types: When restricting the inbound rate limit to a certain traffic
type (broadcast, multicast and/or unknown unicast) on these Ethernet ports,
there are dependencies between the settings:
– Unknown unicast: Selecting ”unknown unicast” implies that ”unknown
unicast”, ”multicast” and ”broadcast” traffic will be subject to inbound
rate limiting.
– Multicast: Selecting ”multicast” implies that ”multicast” and ”broadcast” traffic will be subject to inbound rate limiting.
© 2017 Westermo Teleindustri AB
247
Westermo OS Management Guide
Version 4.21.1-0
– Broadcast: Selecting ”broadcast” simply means that ”broadcast” traffic
will be subject to inbound rate limiting.
ˆ Rate limiting on Gigabit ports: Maximum rate limit on (MV88E6185) Gigabit
ports is 250 Mbit/s. Setting a higher rate limit (e.g., 300 Mbit/s) will result in
a rate limit of 250 Mbit/s.
Due to these restrictions, it is recommended that inbound rate limiting is primarily used as a means of storm prevention, on the ports where these restrictions
apply.
10.1.7
Outbound/Egress traffic shaping
The switch can be configured to limit the outbound data rate on a port (outbound
traffic shaping). By default each port will send at the maximum speed of the link,
but with outbound traffic shaping activated the switch will limit the outbound rate
to a given threshold. Above that threshold the switch will buffer packets - bursty
traffic will be shaped. In case the output buffer is full, additional packets destined
for that port will be dropped.
When configuring the threshold rate for outbound traffic shaping, the same settings as for inbound rate limiting (see section 10.1.6) applies. For outbound traffic
shaping it is also possible to specify rate in frames per second. The web interface provides a predefined set of rates (drop-down list). The CLI allows for more
fine-grained rate settings:
ˆ Bits per second:
– in steps of 64 kbit/s in range 64-1000 kbit/s
– in steps of 1 Mbit/s in range 1-100 Mbit/s
– in steps of 10 Mbit/s in range 100-1000 Mbit/s (on Gigabit Ethernet
ports)
ˆ Frames per second: in range 7700-1488000 frames per second
Traffic shaping calculations consider the layer-2 bits, i.e., from Ethernet destination MAC address to CRC (interframe gap and preamble bits are not counted).
248
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Note
Outbound traffic shaping in frames per second mode is available for Ethernet ports on all WeOS products, with exceptions for ports on some RedFox
and RedFox Industrial Rack models. The Ethernet ports listed to have restrictions for ingress rate limiting (see section 10.1.6.1) also lack support for the
frames per second mode.
Furthermore, outbound traffic shaping in frames per second mode is not
available available for DSL ports (ADSL/VDSL or SHDSL) ports.
10.1.8
Bandwidth per port
The switch can be configured to monitor the inbound and outbound bandwidth
per port. By default this feature is disabled, but can be enabled for any Ethernet
and DSL port. The bandwidth calculation is split into inbound and outbound traffic
for the last 10 seconds, 1 minute, 10 minutes and 1 hour using moving average.
These statistics are recalculated every 10 seconds.
10.1.9
MDI/MDIX crossover
By default a switch is able to sense which pin to use for reception and which
to use for transmission (auto MDI/MDIX crossover), thus no external crossover
cable is necessary. In addition, a port can be configured statically in MDI (Media
Dependent Interface) or MDIX (crossover) mode.
10.1.10
Fastlink - Fast link-up/link-down on fixed 10/100 Ethernet copper ports
Default port settings in WeOS are aimed at being conformant and compatible with
as many devices as possible. Therefore the ports are setup to auto-negotiate
speed, duplex and automatically agree with the link partner on which end should
cross RX and TX when a straight cable is used. Naturally this takes a bit of time,
despite all current products today do this in dedicated PHY circuitry.
To speed things up considerably, a feature called ”Fastlink” can be activated on
fixed 10/100 Mbit/s Ethernet copper ports2 . This feature basically disables any
2 Fastlink
does not apply to Gigabit Ethernet ports, or to any SFP ports. Another exception
© 2017 Westermo Teleindustri AB
249
Westermo OS Management Guide
Version 4.21.1-0
IEEE back-offs and timeouts in place to protect against glitches and temporary
link loss that otherwise prevent the port from going UP or DOWN. Westermo has
put a great deal of effort into making sure that, when enabling Fastlink, glitches
and link loss still do not occur.
Enable Fastlink by configuring the port(s) with the following two settings:
ˆ Fixed speed/duplex mode, preferably 100 Mbit full-duplex. That is, autonegotiation of speed/duplex mode is disabled. See section 10.1.2 for information on port speed/duplex.
ˆ Fixed MDI/MDIX crossover mode, i.e., auto-MDI/MDIX is disabled. See section 10.1.9 for information on port crossover mode.
In most use-cases auto-negotiation of speed-duplex and MDI/MDIX is still preferable, but enabling Fastlink can improve failover performance in some redundancy
applications – we refer to this as the fastlink mode:
ˆ RedFox Rail [56] bypass relay ports[56]: RedFox Rail routers equipped with
a bypass relay are typically used in train backbones. The four backbone
ports, two in each direction, are controlled by a relay, ensuring connectivity
between routers on the backbone when one or more routers experience
power-loss. The fastlink mode minimises disruption when the bypass relay
changes state.
ˆ Layer-2 redundancy: the fastlink mode can improve failover performance for
various layer-2 redundancy mechanisms, e.g., when using static link aggregation (section 19).
Note
The fastlink mode requires more precise knowledge of cabling and devices
used because all automatic detection is disabled. E.g., on the RedFox
Rail[56] bypass relay ports, make sure to setup 100 Mbps Full-Duplex, with
MDI/MDIX mode set to either:
ˆ MDIX in both directions and crossover cables between switches, or
ˆ MDI in one direction, MDIX in the other, with a straight cable
The latter case does however not work when a train car is turned 180°, but
may be useful in other setups since straight cables are more commonplace.
is the specific 10/100 Mbit/s Ethernet ports on ”Viper-12A/20A”[59, 60, 61, 62] attached to an
MV88E6352 switchcore. Please see section 4.4.2 (Web) or section 7.3.2 (CLI) to find definite information about what switchcore(s) is used in your product.
250
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
10.1.11
Fallback default VID
The fallback default VLAN ID is generally unnecessary to configure.
The purpose of the fallback default-VID is to control what should happen with ”untagged” packets entering a port only configured ”tagged” on a set of VLANs. For
more information on VLAN features and the VLAN related terms used throughout
this section, see chapter 15.
Every port needs to have a ”default VID”. The default VID specifies the VLAN ID
an ”untagged” packet should be associated with as it enters that port. A port’s
default VID is determined as follows:
ˆ If a port is associated ”untagged” with a VLAN, that VID will be the port’s
default VID. E.g., if a port is associated ”untagged” to VID 10, the port will
have VID 10 as its ”default VID”.
ˆ If a port is not associated ”untagged” with any VLAN, the port’s default VID
is determined as:
– the port’s fallback default VID, given that a fallback default-VID is configured, or
– the default VLAN (VID 1), if no fallback default-VID is configured
The fallback default VID can be used to control whether ”untagged” packets
should be accepted on a port (only) associated ”tagged” with a set of VLANs.
If the port’s default VID is represented within that set of VLANs, the packet will
be accepted. Otherwise it will be dropped.
10.1.12
SFP DDM/DOM Diagnostics
Digital diagnostics monitoring (DDM), also known as digital optics monitoring
(DOM), is a function enabling the user to monitor diagnostic parameters of the
SFP.
WeOS provides diagnostic information for the following DDM parameters.
ˆ Optical TX power
Measures the optic power when transmitting, which can be used for detecting a deteriorating link3 . The accuracy is better than +/-3dB and the total
3 By comparing the TxPower on a unit with the RxPower on the unit it is connected to, the user
can find out the amount of "signal strength" that is lost over the optic link. When the gap between
© 2017 Westermo Teleindustri AB
251
Westermo OS Management Guide
Version 4.21.1-0
range of -40 to +8.2 dBm (0–6.5535 mW).
ˆ Optical RX power
Measures the optic power when receiving, which can be used for detecting
a deteriorating link. The accuracy is better than +/-3dB and the total range
of -40 to +8.2 dBm (0–6.5535 mW).
ˆ Temperature
The temperature of the SFP should be very close or equal to the temperature
of the unit. The temperature accuracy is better than 3 degrees Celsius (°C)
and the total range of -128 °C to +128 °C.
ˆ Bias current
The transmitting bias current can be used to determine how fast an SFP is
aging. The accuracy is better than +/- 10% and the total range of 0 - 131
mA.
ˆ Voltage The voltage should always be 3.3V since the SFP’s power supply
line is the same as the unit. The accuracy is better than +/-3% and the total
range of 0–6.55 V.
DDM/DOM information will only be listed for enabled ports.
Note
DDM support in WeOS is limited to Westermo DDM SFPs, see the SFP
Transceiver Datasheet of your WeOS product (www.westermo.com).
TxPower and RxPower is increasing, the optic link’s capability to transfer the signal decreases.
252
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
10.2
Managing port settings via the web interface
10.2.1
List Port Settings
Menu path: Configuration ⇒ Port ⇒ Port
When entering the port configuration page you will be presented to a list of all
ports available on your switch, see fig. 10.2. Here you get an overview of the
settings for all ports, and in addition two items of dynamic information - alarms
and link status.
Figure 10.2: Port configuration settings overview (this example is from a RedFox
Industrial switch)
Alarm
Port
Enabled
Link
Type
There is an active link alarm associated with the port.
Only shown if link alarm is enabled and the link is down.
The port label.
Shows if the port is enable or disabled
Link status for the port. Up or down.
The port type: Gigabit Ethernet Fibre optic, Gigabit Ethernet, Fast Ethernet Fibre optic or Fast Ethernet.
Continued on next page
© 2017 Westermo Teleindustri AB
253
Westermo OS Management Guide
Version 4.21.1-0
Speed/Duplex
Link Alarm
Enabled
Edit
Continued from previous page
The speed duplex setting. Auto means speed and duplex
will be automatically negotiated. Otherwise the current
setting will be shown as speed in Megabit and duplex as
FDX for full duplex and HDX for half duplex. Note! This is
not the negotiated speed, it is the configuration setting!
When link alarm is enabled an alarm will be generated if
port link is down. Alarms trigger an SNMP trap message
to be sent and alarms to be shown on the administration web. In the ports overview table a green check-mark
means enabled, and a dash means disabled.
Click this icon to edit a port’s settings.
To change the settings for a specific port you will have to click the edit icon which
will take you to the port setting edit page see section 10.2.2.
10.2.2
Edit Port Settings
Menu path: Configuration ⇒ Port ⇒ Port ⇒
On this page you can change the settings for the port.
254
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Type
Enable
Speed/Duplex
MDIX mode
Priority Mode
Priority
Inbound Rate
Limit
Outbound
Traffic Shape
Bandwidth
Statistics
Link Alarm
The port type: Gigabit Ethernet Fibre optic, Gigabit Ethernet,
Fast Ethernet Fibre optic or Fast Ethernet.
Enable/disabled the port
The speed duplex setting. Auto means speed and duplex will
be automatically negotiated. Otherwise the current setting
will be shown as speed in Megabit and duplex as FDX for
full duplex and HDX for half duplex. Note! This is not the
negotiated speed, it is the configuration setting!
How to handle crossover cables. If you connect two units
with different port settings (one with mdi and one with
mdix) you need a straight-through twisted pair cabling. If
you connect two units with the same setting you will need
a crossover cabling.
Auto Automatic detection
mdi
Medium dependent interface
mdix mdi crossover
Here you select on what information priority will be based:
Port Based Based on the port’s priority. See the next
item (Priority).
IP
Based on the content of the IP ToS bits (IPv4)
or the IP TC bits (IPv6).
VLAN Tag
Based on the content of the (802.1p) priority
field inside the received packet’s VLAN tag.
The port’s priority level. Zero (0) is low priority and seven
(7) high priority.
Bandwidth limit for inbound traffic. Disabled means no limiting.
Bandwidth limit for outbound traffic. Disabled means no limiting.
Enable or disable bandwidth monitoring per port. See Section 11.2.2 for how to view the statistics that is gathered
when this function is enabled.
When link alarm is enabled an alarm will be generated if
port link is down. Alarms trigger an SNMP trap message to
be sent and alarms to be shown on the administration web.
© 2017 Westermo Teleindustri AB
255
Westermo OS Management Guide
Version 4.21.1-0
10.2.3
List SFP DDM/DOM diagnostics
For information on how to view SFP DDM/DOM diagnostics, see section 4.4.3.
256
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
10.3
Managing port settings via the CLI
The port configuration context can be entered using the ”port <PORT|PORTLIST>”
command from the Global Configuration context. When providing a list of ports,
the scope of the configuration commands becomes all ports in the list. There is
also a specific command, ”ports”, to enter the port context with the scope of all
Ethernet ports of the device.
Command
port [eth|. . . ] <PORTLIST>
ports [eth|. . . ]
[no] enable
[no] speed-duplex <auto|10-half|10-full|
100-half|100-full|. . . >
[no] flow-control
[no] priority <0-7>
[no] priority-mode <tag|ip|port>
[no] link-alarm
[no] rate-limit <64-1000000>
[match <TYPE>[,<TYPE>,...]]
[no] traffic-shaping <<64-1000000>|
<7700-1488000> fps>
[no] bandwidth-statistics
[no] mdix-mode <auto|mdi|mdix>
[no] unshielded
[no] low-power
[no] default-vid <VLAN_ID>
Default
Ethernet
Ethernet
Enabled
auto
Section
Section 10.3.1
Section 10.3.2
Section 10.3.3
Section 10.3.4
Disabled
0
tag
Disabled
Disabled
Section
Section
Section
Section
Section
Disabled
Section 10.3.10
Disabled
auto
Unshielded
Low Power
Disabled
Section
Section
Section
Section
Section
10.3.5
10.3.6
10.3.7
10.3.8
10.3.9
10.3.11
10.3.12
10.3.13
10.3.14
10.3.15
Show port status
show ports
Section 10.3.16
Show SFP DDM/DOM diagnostics
show environment
Section 7.3.41
© 2017 Westermo Teleindustri AB
257
Westermo OS Management Guide
Version 4.21.1-0
10.3.1
Managing Ports
Syntax port [eth|...] <PORT|PORTLIST>
The ”port” command is used for many port types, thus the full command
syntax is
”port [eth|dsl|shdsl|xdsl|serial] <PORT|PORTLIST>”.
Context Global Configuration context
Usage Enter Port Configuration context of the given PORT (or PORTLIST) and port
type.
A ”PORTLIST” is a comma separated list of ranges of ports without intermediate spaces, e.g., ”1/1,1/2” on a slotted product, or ”1-3,5” on a nonslotted product.
The port qualifier keyword ”eth|...” is not needed if the numbers in the
”PORTLIST” are unique to a single type of port. If there are multiple port
with the same number (but different types), the port qualifier is needed,
e.g., ”port eth 1” and ”port dsl 1”.
For information on using the ”port” command to enter:
ˆ xDSL Port Configuration context, see section 13.3.1.
ˆ SHDSL Port Configuration context, see section 12.3.1.
ˆ Serial port context, see section 38.3.1.
Use ”show port [eth|...] [PORT|PORTLIST]” to list port configuration
information on one or more ports. Also available as ”show” command within
thePort Configuration.
Default values Not applicable for configuration. For listing configuration ”show
port” information on all ports are listed by default.
Entering port configuration context for Ethernet ports 1-4:
Example
example:/config/#> port 1-4
example:/config/port-eth1-4/#>
This unit has two ports with number 1 (”eth 1” and ”dsl 1”) thus the port qualifier is needed to determine which port to configure:
258
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/config/#> port 1
Ambiguous or bad port range or port type: 1
example:/config/#> port dsl 1
example:/config/port-dsl1/#>
10.3.2
Managing all Ports
Syntax ports [eth|dsl|shdsl|xdsl|serial]
Context Global Configuration context
Usage Enter Port Configuration context with the scope of all ports of a specific
type (Ethernet, xDSL, etc.).
Default values ”Ethernet” for configuration (i.e., ”ports” will enter Ethernet
Port Configuration for all Ethernet ports), and ”All” for showing configuration
(i.e., ”show ports” will list information on all port types).
Listing information on all ports.
Example
example:/config/#> show ports
Ethernet ---------------------------------------- Priority ---- Limit - Default
Port
Ena Aneg Speed DPX Flow MDI/X Alarm Mode Level
In | Out
Vid
===============================================================================
Eth 1
YES YES --NO
auto
NO tag
0
None None
Auto
Eth 2
YES YES --NO
auto
NO tag
0
None None
Auto
Eth 3
YES YES --NO
auto
NO tag
0
None None
Auto
Eth 4
YES YES --NO
auto
NO tag
0
None None
Auto
===============================================================================
xDSL -------------------------------------------- Priority ---- Limit - Default
Port
Ena Mode Filter Encap PVC
Annex Alarm Mode Level
In | Out
Vid
===============================================================================
DSL 1
YES adsl
YES
llc 8/35
A
NO tag
0
None None
Auto
===============================================================================
Serial --------------------- Data ------- Stop RTS XON -----------------------Port
Ena Type
Speed
bits Parity bits CTS XOFF Terminate
===============================================================================
Ser 1
YES rs232
115200
8
None
1
OFF OFF
N/A
===============================================================================
example:/config/#>
Listing information on a all ports of a specific type
© 2017 Westermo Teleindustri AB
259
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/config/#> show ports dsl
xDSL -------------------------------------------- Priority ---- Limit - Default
Port
Ena Mode Filter Encap PVC
Annex Alarm Mode Level
In | Out
Vid
===============================================================================
DSL 1
YES adsl
YES
llc 8/35
A
NO tag
0
None None
Auto
===============================================================================
example:/config/#>
10.3.3
Port enabling and disabling
Syntax [no] enable
Context Ethernet Port Configuration context (also available in SHDSL Port Configuration and xDSL Port Configuration for products with DSL ports)
Usage Use ”enable” to enable and ”no enable” disable a port.
Use ”show enable” to show whether the port is enabled or disabled.
Default values Ports are enabled by default.
10.3.4
Speed and duplex setting
Syntax [no] speed-duplex <auto|10-half|10-full|100-half|100-full|
1000-half|1000-full>
Context Ethernet Port Configuration context.
Usage Set port speed and duplex modes. ”auto” means auto-negotiate, other
modes are static configurations specifying 10, 100 or 1000 Mbit/s, and half
or full duplex.
”no speed-duplex” will revert to default configuration for the speed-duplex
setting, i.e., ”speed-duplex auto”.
Use ”show speed-duplex” to show the port’s speed/duplex setting.
Default values auto
Error messages An attempt to set a port speed not available for this specific
port type will render an error message, including information of available
port speeds.
260
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
10.3.5
Flow-control setting
Syntax [no] flow-control
Context Ethernet Port Configuration context.
Usage Enable or disable IEEE 802.3 flow-control. For full duplex links, flow control will utilise IEEE 802.3 pause frames, and for half duplex links a technique
known as back-pressure is used.
The flow control setting is only valid when the speed-duplex mode is set to
”auto”, see section 10.1.2.
Use ”show flow-control” to show the port’s flow-control setting.
Default values Disabled (no flow-control)
10.3.6
Port priority setting
Syntax [no] priority <0-7>
Context Ethernet Port Configuration context (also available in SHDSL Port Configuration and xDSL Port Configuration for products with DSL ports)
Usage Set the (IEEE 802.1p) priority associated with the port. Packets coming
in on this port will receive this priority unless priority is based on VLAN ID,
VLAN tag or IP ToS/DiffServ bits.
”no priority” will revert to default configuration for the port priority setting, i.e., ”priority 0” (zero).
Use ”show priority” to show the port’s priority setting.
Default values 0 (zero)
10.3.7
Set port priority mode
Syntax [no] priority-mode <tag|ip|port>
Context Ethernet Port Configuration context (also available in SHDSL Port Configuration and xDSL Port Configuration for products with DSL ports)
© 2017 Westermo Teleindustri AB
261
Westermo OS Management Guide
Version 4.21.1-0
Usage Base priority classification for this port on content of VLAN tag (IEEE
802.1p priority bits), content of IP ToS/Diffserv bits, or the port priority configured for this port.
Note
VLAN priority settings (see section 15.4) will have precedence over port
priority mode settings.
tag (Default) The packet’s priority is based on the content of the VLAN tag
(802.1p priority bits) of the incoming packet. For packets coming in
untagged, the priority is based on the priority associated with the port,
see section 10.3.6.
ip The packet’s priority is based on the content of the IP ToS/Diffserv bit of
the incoming packet. For non-IP packets coming in on the port (e.g.,
ARP packets), the priority is based on the priority associated with the
port, see section 10.3.6.
port The packet’s priority is based on the priority associated with the port,
see section 10.3.6.
Use ”show priority-mode” to show the port’s ”priority mode” setting.
Default values tag
10.3.8
Link alarm
Syntax [no] link-alarm
Context Ethernet Port Configuration context (also available in SHDSL Port Configuration and xDSL Port Configuration for products with DSL ports)
Usage Use ”link-alarm” to enable and ”no link-alarm” disable link-alarm for
this port. When enabled, an alarm indication is activated when the link is
down.
”show link-alarm” to show the port’s link-alarm setting.
Default values Disabled (”no link-alarm”)
262
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
10.3.9
Inbound rate limiting
Syntax [no] rate-limit <64-1000000> [match <TYPE>[,<TYPE>,...]]
Context Ethernet Port Configuration context (also available in SHDSL Port Configuration and xDSL Port Configuration for products with DSL ports)
Usage Configure inbound rate limit in kbit/s. It is also possible use ISO modifiers
k/M/G, e.g., 256k or 10M as specifiers for kbit/s and Mbit/s.
Note
Set values are rounded off to the nearest possible HW setting.
Optionally packet TYPE may be specified using one or more of the specifiers
”all” (all types), ”bc” (broadcast), ”mc” (multicast) or ”u-uni” (unknown
unicast) in any combination. If no TYPE is specified (or if the specifier ”all”
is given) all packets will be rate limited.
Note
All WeOS products except RedFox and RedFox Industrial Rack support
any combination of types. As stated in section 10.1.6.1, the traffic type
selection on RedFox and RedFox Industrial Rack implicitly adds ”bc” if
”mc” is specified, and adds both ”bc,mc” if ”u-uni” is specified.
Use ”no rate-limit” to disable inbound rate limiting.
Use ”show rate-limit” to show the port’s inbound rate limit setting.
Default values Disabled (”no rate-limit”)
10.3.10
Outbound traffic shaping
Syntax [no] traffic-shaping <<64-1000000>|<7700-1488000> fps>
Context Ethernet Port Configuration context (also available in SHDSL Port Configuration and xDSL Port Configuration for products with DSL ports, albeit
not fps)
Usage Configure outbound traffic shaping in kbit/s or frames per second. It is
also possible use ISO modifiers k/M/G, e.g., 256k or 10M as specifiers for
kbit/s and Mbit/s.
© 2017 Westermo Teleindustri AB
263
Westermo OS Management Guide
Version 4.21.1-0
Note
Set values are rounded off to the nearest possible HW setting.
Use ”no traffic-shaping” to disable outbound traffic shaping.
Use ”show traffic-shaping” to show the port’s outbound traffic shaping
setting.
Default values Disabled (”no traffic-shaping”)
10.3.11
Bandwidth statistics
Syntax [no] bandwidth-statistics
Context Ethernet Port Configuration context (also available in SHDSL Port Configuration and xDSL Port Configuration for products with DSL ports)
Usage Use ”bandwidth-statistics” to enable and ”no bandwidth-statistics”
to disable bandwidth-statistics for this port. When enabled, bandwidth monitoring for inbound and outbond traffic is calculated.
”show bandwidth-statistics” to show whether monitoring is enabled or
disabled.
Please see Section 11.3.5 for how to view the statistics that is gathered
when this function is enabled.
Default values Disabled (”no bandwidth-statistics”)
10.3.12
Cable crossover setting
Syntax [no] mdix-mode <auto|mdi|mdix>
Context Ethernet Port Configuration context.
Usage Configuration of Cable Crossover setting. ”auto” means automatic crossover
mode, ”mdix” sets port to crossover mode (MDIX) and ”mdi” sets port to
MDI mode. This command is not valid for fibre ports.
”no mdix-mode” resets the MDIX mode to the default setting (”auto”).
Use ”show mdix-mode” to show the port’s cable crossover setting.
264
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Default values auto.
10.3.13
Adapting PHY Receiver to Shielded or Unshielded Cable
Syntax [no] shielded
Context Ethernet Port Configuration context.
Usage Fine tune the PHY receiver to the cable characteristics of shielded or unshielded TP cables. This setting applies to 10/100 Base-TX ports, excluding
SFP/SFF ports as well as ports also capable of 1000 Mbit/s speeds.
Use ”shielded” to adapt the PHY receiver to the use of shielded TP cables.
Use ”no shielded” to adapt the PHY receiver to the use of unshielded TP
cables.
Note
This setting is only expected to be used by customers with special requirements - the default setting should be sufficient for most use cases.
Use ”show shielded” to show the port’s ”shielded” setting.
Default values Unshielded (no shielded).
10.3.14
Enable/disable Low Power Mode on TX Data Signalling
Syntax [no] low-power
Context Ethernet Port Configuration context.
Usage It possible to select between two signal power modes on the Ethernet
data signalling pins for 10/100 Base-TX ports. (This setting applies to 10/100
Base-TX ports, excluding SFP/SFF ports as well as ports also capable of 1000
Mbit/s speeds.)
The low-power mode is sufficient in most use cases, but for long cables or
cables with specific characteristics it may be necessary to disable low-power
mode.
Use ”low-power” and ”no low-power” respectively to enable/disable lowpower mode on this Ethernet port.
© 2017 Westermo Teleindustri AB
265
Westermo OS Management Guide
Version 4.21.1-0
Note
This setting is only expected to be used by customers with special requirements - the default setting should be sufficient for most use cases.
Use ”show low-power” to show whether the PHY (TX Data Signalling) lowpower mode is enabled or disabled.
Default values Low-Power (low-power).
10.3.15
Fallback default VLAN
Syntax [no] default-vid <VLAN_ID>
Context Ethernet Port Configuration context (also available in SHDSL Port Configuration and xDSL Port Configuration for products with DSL ports)
Usage Configuration of (fallback) default-VID for this port. The default-VID configuration is only valid when this port is not configured ”untagged” on any
VLAN.
Use ”no default-vid” to clear the (fallback) default VID setting (the defaultVID setting will also be cleared whenever the port is associated ”untagged”
with any VLAN).
When cleared (”no default-vid”), VLAN ID 1 will be used as the port’s
fallback default-VID.
For more information see section 10.1.11.
Use ”show default-vid” to show the port’s ”fallback default-VID” setting.
Default values Disabled/cleared (no default-vid).
10.3.16
Show port status (all ports)
Syntax show ports
Context Admin Exec context
Usage Show Port status information for all ports.
Default values Not applicable.
266
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Chapter 11
Ethernet Statistics
A set of per port Ethernet statistic counters are available via the Web and via the
CLI. Most of these counters correspond to standard SNMP MIB Ethernet statistics
counters from the RMON MIB (RFC 2819), the Interface MIB (RFC 2863) and the
Ether-Like MIB (RFC 3635). For more information about WeOS SNMP support, see
chapter 6.
Section 11.1 gives a general introduction to the Ethernet statistic counters available via Web and CLI. Sections 11.2 and 11.3 present use of Ethernet statistics
via the Web and CLI respectively.
11.1
Ethernet Statistics Overview
The table below provides a summary of the available Ethernet statistics counters.
Sections 11.1.1-11.1.8 give more detailed information on their meaning.
Feature
Inbound
Total Bytes
Bytes Good
Bytes Bad
Mean rate
Total Good Packets
Unicast
© 2017 Westermo Teleindustri AB
Web
CLI
Description
(X)2 Section 11.1.1
X
-”X
-”X
-”2
(X)
Section 11.1.2
X
X
-”Continued on next page
X
267
Westermo OS Management Guide
Version 4.21.1-0
Feature
Multicast
Broadcast
Pause frames
Size statistics
Dropped
Filtered
Discarded
Erroneous
Undersize
Oversize
Fragments
Jabber
Checksum
PHY Error
Bandwidth Statistics
10 second average
1 minute average
10 minute average
1 hour average
Continued from previous page
Web CLI Description
X
X
-”X
X
-”X
-”X
-”X
X
Section 11.1.3
X
-”X
-”(X)2 Section 11.1.4
X
X
-”X
X
-”X
X
-”X
X
-”X
X
-”X
-”X
X
Section 11.1.9
X
X
-”X
X
-”X
X
-”X
X
-”-
Outbound
Total Bytes
Mean rate
Total Packets
Unicast
Multicast
Broadcast
Pause frames
Dropped
Filtered
Collisions and Busy Medium
Single
Multiple
Excessive
268
X
X
X
(X)2
X
X
X
X
Section 11.1.5
”
2
(X)
Section 11.1.6
X
-”X
-”X
-”-”Section 11.1.7
X
-”2
X
(X)
Section 11.1.8
X
-”X
-”X
-”Continued on next page
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Feature
Late
Other collisions
Deferred
Bandwidth Statistics
10 second average
1 minute average
10 minute average
1 hour average
11.1.1
Continued from previous page
Web CLI Description
X
X
-”X
-”X
-”X
X
Section 11.1.9
X
X
-”X
X
-”X
X
-”X
X
-”-
Inbound Byte Counters
A set of byte counters (i.e., octet counters) are provided. The number of good
bytes is also used to compute a rough estimation of the current inbound data
rate.
Bytes Good The number of good bytes/octets received on a port, i.e., the sum
of the length of all good Ethernet frames received.
Bytes Bad The number of bad bytes/octets received on a port, i.e., the sum of
the length of all bad Ethernet frames received.
Total Bytes The sum of good and bad bytes received on a port (see above).
This would correspond to the RMON MIB etherStatsOctets and the Interface
MIB ifHCInOctets objects.
Mean Rate Rough estimation of the current data rate based on the number of
good bytes received during a time interval (2 seconds).
11.1.2
Inbound Counters of Good Packets
The following per port counters for good inbound Ethernet packets are provided.
Unicast packets The number of good packets with a unicast MAC address received on the port.
This would correspond to the Interface MIB ifInUcastPkts object.
2 Counters
listed within parenthesis (i.e., as ’(X)’) are provided implicitly.
© 2017 Westermo Teleindustri AB
269
Westermo OS Management Guide
Version 4.21.1-0
Multicast packets The number of good packets with a group MAC address (excluding broadcast) received on the port.
This would correspond to the RMON MIB etherStatsMulticastPkts and the
Interface MIB ifInMulticastPkts objects, except that Pause frames (see below)
are not included.
Broadcast packets The number of good packets with a broadcast MAC address
received on the port.
This would correspond to the RMON MIB etherStatsBroadcastPkts and the
Interface MIB ifInBroadcastPkts objects.
Pause Frames The number of good flow control packets received.
Packet Size Statistics Counters for good Ethernet packet of the following size
intervals are provided: 64 bytes, 65-127 bytes, 128-255 bytes, 256-511
bytes, 512-1023 bytes, and 1024-MAXPKTSIZE bytes, where MAXPKTSIZE is
1632.
These size intervals match the corresponding RMON statistics counters, except for the MAXPKTSIZE (1632 instead of 1518).
11.1.3
Dropped Inbound Packets
Counters for two types of dropped inbound packets are provided. Note, these
packets are good Ethernet packets, but are dropped due to the reasons given
below.
Filtered Inbound packets dropped due to VLAN mismatch or because the port
was in LEARNING, LISTENING or BLOCKING state.
Discarded Packets dropped due to lack of buffer space.
11.1.4
Erroneous Inbound Packets
The following counters for received erroneous packets are provided:
Undersized packet Number of packets smaller than 64 bytes, and with a valid
FCS.
This corresponds to the RMON MIB etherStatsUndersizePkts object.
270
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Oversized packet Number of packets larger than 1632 bytes, and with a valid
FCS.
This corresponds to the RMON MIB etherStatsOversizePkts object, except
for the used MAXPKTSIZE (1632 instead of 1518 bytes).
Fragmented packet Number of packets smaller than 64 bytes, with an invalid
FCS.
This corresponds to the RMON MIB etherStatsFragments object.
Jabber Number of packets larger than 1632 bytes, and with an invalid FCS.
This corresponds to the RMON MIB etherStatsJabbers object, except for the
used MAXPKTSIZE (1632 instead of 1518 bytes).
Checksum/FCS Error Packets of valid length (64-1632), but with an incorrect
FCS.
This corresponds to the RMON MIB etherStatsCRCAlignErrors object, except
for the used MAXPKTSIZE (1632 instead of 1518 bytes).
PHY Error Signal Number of received packets generating a receive error signal
from the Ethernet PHY. (Referred to as InMacRcvErr in the CLI port statistics
list)
11.1.5
Outbound Byte Counters
A single outbound byte/octet counter, Outbound Bytes, is provided. It represents the sum of the length of all Ethernet frames sent on the port. This would
correspond to the Interface MIB ifHCOutOctets object.
The number of Outbound bytes is also used to calculate a rough estimation of
the current sending data rate (Mean Rate, i.e., the number of bytes sent during
a time interval (2 seconds).
11.1.6
Outbound Packets Counters
The following per port counters for outbound Ethernet packets are provided.
Unicast packets The number of packets with a unicast destination MAC address
sent on the port.
This would correspond to the Interface MIB ifOutUcastPkts object.
© 2017 Westermo Teleindustri AB
271
Westermo OS Management Guide
Version 4.21.1-0
Multicast packets The number of packets with a group destination MAC address (excluding broadcast) sent on the port.
This would correspond to the Interface MIB ifOutMulticastPkts objects, except that Pause frames (see below) are not included.
Broadcast packets The number of packets with a broadcast destination MAC
address sent on the port.
This would correspond to the Interface MIB ifOutBroadcastPkts objects.
Pause Frames The number of flow control packets sent.
11.1.7
Dropped Outbound Packets
The counter for a single type of dropped outbound packets is described here
(there is also a second kind, see excessive collisions in section 11.1.8).
Filtered Outbound packets dropped outbound policy rules or because the port
was in LEARNING, LISTENING or BLOCKING state.
11.1.8
Outbound Collision and Busy Medium Counters
The collision and busy medium counters described here are only relevant for
half-duplex links.
Single Collisions The number of packets involved in a single collision, but then
sent successfully.
This would correspond to the Ether-like MIB dot3StatsSingleCollisionFrames
object.
Multiple Collisions The number of packets involved in more than one collision,
but finally sent successfully.
This would correspond to the Ether-like MIB dot3StatsMultipleCollisionFrames
object.
Excessive Collisions The number of packets failing (i.e., dropped) due to excessive collisions (16 consecutive collisions).
This would correspond to the Ether-like MIB dot3StatsExcessiveCollisions object.
272
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Late Collisions The number of collisions detected later than a 512-bits time into
the packet transmission.
This would correspond to the Ether-like MIB dot3StatsLateCollisions object.
Other Collisions Other collisions than single, multiple, excessive or late collisions discovered on a port.
Total Collisions Computed as the sum of single, multiple, excessive, late and
other collisions.
Deferred (busy medium) The number of packets experiencing a busy medium
on its first transmission attempt, and which is later sent successfully, and
without experiencing any collision.
This would correspond to the Ether-like MIB dot3StatsDeferredTransmissions
object.
11.1.9
Inbound and outbound bandwidth statistics
The bandwidth statistics counters are described here.
These counters need to be enabled to give any statistics data. This is configured
per port. Please see Section 10.3.11 for CLI and Section 10.2.2 for web.
The counters are split into inbound and outbound traffic and shown as bits, kbits,
Mbits or Gbits per second.
10 second average The average bandwidth over the last 10 seconds.
1 minute average The average bandwidth over the last minute.
10 minute average The average bandwidth over the last 10 minutes.
1 hour average The average bandwidth over the last hour.
© 2017 Westermo Teleindustri AB
273
Westermo OS Management Guide
Version 4.21.1-0
11.2
Statistics via the web interface
Statistics shown in the web administration tool has two views. An overview with
a selection of statistics for all ports, including some status information (e.g. if
port is blocking or forwarding), and a detailed page with a larger set of statistics.
Note that collection of statistics is started by the first access to the statistics
page, and will be halted after a short period of time (to save resources) if no one
requests the statistic data. This has the effect that you may need to enter the
page once again, by e.g. clicking the menu item, to ensure you are presented to
updated statistics data.
11.2.1
Statistics Overview
Menu path: Status⇒Port
On the port statistics overview page you will be presented to a selection of static
data for each port. Additional statistic numbers are presented on the detailed
view page.
274
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Alarm
Port
Link
State
Speed / Duplex
Total Bytes In
Total Bytes Out
FCS Errors
Details
Auto Refresh
Refresh
Clear All
An alarm icon appears at the start of a line if there is a
link alarm on a port.
The port label.
The status of the link. Up or down.
FORWARDING Unit forwards packets. Normal operation.
LEARNING
The port is preparing itself for entering FORWARDING state.
BLOCKING
Unit does not forward any packets.
DISABLED
Port does not participate in operation.
The current speed and duplex negotiated or set on the
port.
Total number of bytes received on the port.
Total number of bytes sent out on the port.
Total number of inbound packets with check sum error
received on the port.
Click this icon to view more detailed statistics for the
port.
Click on a value to make the page reload with updated
statistics automatically every 5, 15, 30 or 60 seconds.
Click Off to turn off auto refresh.
Click on this button to reload with updated statistics.
Clear all statistics counters for all ports.
© 2017 Westermo Teleindustri AB
275
Westermo OS Management Guide
Version 4.21.1-0
11.2.2
Detailed Statistics
Menu path: Status ⇒ Port ⇒
When clicking the details-icon in the overview page you will be presented to the
detailed statistics page for the port.
276
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Link Status
Total Bytes
Broadcast Packets
Multicast Packets
Unicast Packets
Dropped Packets
Fragments
Oversize
Undersize
Jabber
Frame Checksum
Traffic Size, Inbound
Total Collisions
Single Collisions
Multiple Collisions
Excessive Collisions
Status of link (Up/Down). If a link-alarm is associated with this port, an alarm icon is displayed if the
link-alarm is active.
Total number of bytes received (inbound) or transmitted (outbound) on this port.
Total number of good broadcast packets received
(inbound) or transmitted (outbound) on this port.
Total number of good multicast packets received (inbound) or transmitted (outbound) on this port.
Total number of good unicast packets received (inbound) or transmitted (outbound) on this port.
Total number of packets received that have been
discarded.
Total number of fragmented packets received on this
port.
Total number of oversized packets received on this
port.
Total number of undersized, but otherwise well
formed, packets received on this port.
Total number of packets received on this port larger
than the network segment’s maximum transfer unit
(MTU).
Total number of packets received on this port with
checksum error.
Number of octets received in different size categories.
Total number of collisions detected on this port (sum
of single, multiple, excessive, late, and other collision counters).
The number of packets involved in a single collision,
but then sent successfully.
The number of packets involved in more than one
collision, but finally sent successfully.
The number of packets failing (i.e., dropped) due to
excessive collisions (16 consecutive collisions).
Continued on next page
© 2017 Westermo Teleindustri AB
277
Westermo OS Management Guide
Version 4.21.1-0
Late Collisions
Other collisions
Deferred
Filtered
Bandwidth Statistics
Auto Refresh
«Previous
Next»
Refresh
Clear Port
278
Continued from previous page
The number of collisions detected later than a 512bits time into the packet transmission.
Other collisions than single, multiple, excessive or
late collisions discovered on a port.
The number of packets experiencing a busy medium
on its first transmission attempt, and which is later
sent successfully, and without experiencing any collision.
Outbound packets dropped outbound policy rules or
because the port was in LEARNING, LISTENING or
BLOCKING state.
Shows bandwidth statistics for the port in different
time scales. Note that you must enable bandwidth
statistics for the port in order to see any data here.
See Section 10.2.2.
Click on a value to make the page reload with updated statistics automatically every 5, 15, 30 or 60
seconds. Click Off to turn off auto refresh.
Goto statistics for previous port.
Goto statistics for next port.
Click on this button to reload with updated statistics.
Clear all statistics counters for the port shown.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
11.3
Statistics via the CLI
The table below shows statistic features available via the CLI.
Command
rmon
statistics [PORT]
clear-stats [PORT]
show rmon [PORT]
show port bandwidth [PORT]
11.3.1
Default
Section
Section 11.3.1
Section 11.3.2
Section 11.3.3
Section 11.3.4
Section 11.3.5
Managing Ethernet Statistics
Syntax rmon
Context Admin Exec context
Usage Enter Ethernet statistics context (RMON Statistics context). WeOS starts
gathering statistics when this command is issued, thus there is a 2 seconds
delay before the RMON context is entered.
Default values Not applicable.
11.3.2
List Current Ethernet Statistics
Syntax statistics [PORT]
Context RMON Statistics context
Usage Show Ethernet statistics. If no PORT is given (”statistics”, a summary
of statistics for all Ethernet ports is presented.
If a PORT is given as argument (e.g., ”statistics 1/1”) detailed statistics
for that port is presented.
For information about what the different statistics counters represent, see
section 11.1.
Default values If no PORT argument is given, a summary of statistics for all
Ethernet ports is presented.
© 2017 Westermo Teleindustri AB
279
Westermo OS Management Guide
Version 4.21.1-0
11.3.3
Clear Ethernet Statistics
Syntax clear-stats [PORT]
Context RMON Statistics context
Usage Clear Ethernet statistic counters. If no PORT is given (”clear-stats”,
counters for all Ethernet ports are cleared.
If a PORT is given as argument (e.g., ”clear-stats 1/1”) the counters for
that port are cleared.
Default values If no PORT argument is given, counters for all Ethernet ports are
cleared.
11.3.4
Show Ethernet Statistics
Syntax show rmon [PORT]
Context Admin Exec context. Also available as ”show [PORT]” command within
the RMON Statistics context.
Usage Show Ethernet statistics. This command provides the same information
as the ”statistics” command (section 11.3.2). The only difference is that
the ”show rmon [PORT]” command is available from the Admin Exec context.
If no PORT is given (”show rmon”), a summary of statistics for all Ethernet
ports is presented.
If a PORT is given as argument (e.g., ”show rmon 1/1”) detailed statistics
for that port is presented.
For information about what the different statistics counters represent, see
section 11.1.
Default values If no PORT argument is given, a summary of statistics for all
Ethernet ports is presented.
11.3.5
Show port bandwidth statistics
Syntax show port bandwidth [PORT]
280
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Context Admin Exec context.
Usage Show port bandwidth statistics for the port.
If no PORT is given (”show port bandwidth”), bandwidth statistics is shown
for all ports.
If a PORT is given as argument (e.g., ”show port bandwidth 1/1”) bandwidth statistics for that port is presented.
Note that you must enable bandwidth statistics for the port in order to see
any data here. See Section 10.3.11.
Default values If no PORT argument is given, bandwidth statistics for all ports
is presented.
© 2017 Westermo Teleindustri AB
281
Westermo OS Management Guide
Version 4.21.1-0
Chapter 12
SHDSL Port Management
Wolverine family switches (DDW225/DDW-226/DDW-x42/DDW-x42-4851 ) are equipped
with two SHDSL ports (Symmetric High-speed Digital Subscriber Line), enabling
LAN networks to be extended over legacy copper cabling.
12.1
12.1.1
Overview of SHDSL Port Management
SHDSL overview
With SHDSL Ethernet LANs can be extended over legacy copper cabling. Switches
can be connected in a simple point-to-point setup, but also in multi-drop and ring
topologies, as shown in fig. 12.1.
In a SHDSL connection, the port on one unit shall be configured as Central Office (CO) and the port on the other unit as Customer Premises Equipment (CPE).
SHDSL ports are named according to the name convention described in section 10.1.1). By default 1/1 or DSL 1 is configured as CPE while the 1/2 (or DSL 2)
is configured as CO.
SHDSL support in WeOS is based on Ethernet First Mile (EFM) technology, and
SHDSL can to a large extent be treated in the same way as Ethernet ports, e.g.,
you can add SHDSL ports to VLANs (chapter 15), you can run link-layer redundancy protocols such as FRNT (chapter 16) and RSTP (chapter 18) over them,
1 DDW-x42 refers to DDW-142 and DDW-242 products. DDW-x42-485 refers to DDW-142-485
and DDW-242-485 products.
282
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Feature
CO/CPE mode selection
DSL link rate
DSL noise margin
G.HS threshold
PAF
Low-Jitter
EMF
Web
X
X
X
X
X
X
X
CLI
X
X
X
X
X
X
X
Settings in common with Ethernet ports
Enable/disable port
X
X
Port priority (level)
X
X
Port priority mode
X
X
Link alarm
X
X
Inbound rate limit
X
X
Outbound traffic shaping
X
X
Bandwidth Statistics
X
X
Fall-back default-VID
X
View DSL port configuration
View DSL port status
X
X
General Description
Section 12.1.1-12.1.2
Section 12.1.1-12.1.2
Section 12.1.1-12.1.2
Section 12.1.2
Section 12.1.2
Section 12.1.2
Section 12.1.2
Section
Section
Section
Section
Section
Section
Section
Section
12.1.3
12.1.3
12.1.3
12.1.3
12.1.3
12.1.3
12.1.3
12.1.3
X
X
etc. Settings specific to SHDSL ports are described in section 12.1.2 while port
settings of more general nature is covered in section 12.1.3.
12.1.2
Settings specific to SHDSL ports
ˆ Port role: One unit shall be configured as Central Office (CO) and the other
unit as Customer Premises Equipment (CPE). CO is the answering central
unit. CPE (Customer Premises Equipment) is the unit that initiates the connection. In WeOS the SHDSL ports are named 1/1 and 1/2 in products with
slot based numbering and DSL 1 and DSL 2 in products with simple port
numbering: by default 1/1 (or DSL 1) is configured as CPE and 1/2 or DSL 2
configured as CO.
ˆ Data rate: For a regular SHDSL connection, data rates can be achieved in the
range from 192 kbit/s up to 5696 kbit/s depending on cable characteristics
© 2017 Westermo Teleindustri AB
283
Westermo OS Management Guide
Version 4.21.1-0
1/1 DSL 1/2
1/1 DSL 1/2
Ethernet
Ethernet
a) point−to−point topology
DSL
1/2
1/1
1/1
1/2
Ethernet
DSL
Ethernet
b) point−to−point topology, 2 bonded SHDSL channels
1/1 DSL 1/2
1/1 DSL 1/2
1/1 DSL 1/2
1/1 DSL 1/2
Ethernet
Ethernet
Ethernet
Ethernet
c) multi−drop topology
1/1 DSL 1/2
1/1 DSL 1/2
Ethernet
Ethernet
1/2 DSL 1/1
1/2 DSL 1/1
Ethernet
Ethernet
d) ring topology (redundancy)
Figure 12.1: SHDSL topologies: Point-to-point (a), point-to-point, 2 bonded SHDSL
channels (b), multi-drop (c) and ring (d).
and communication distance. For products supporting turbo-SHDSL, data
rates from 32 kbit/s up to 15304 kbit/s are possible. When using PAF in
DDW-x42 (and DDW-x42-485), data rates up to 30608 kbit/s are possible.
284
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Products with Turbo-speed support
Turbo-speed is supported on all DDW-226 and DDW-x42/DDW-x42-485
devices, and on all but the earliest DDW-225 devices. To see if your
DDW-225 unit supports turbo-speed SHDSL, inspect its article number,
either by reading its attached label, or remotely by viewing the ”Status
⇒ System” Web page or by using the ”show system-information”
command in the CLI. If the article number says ”3642-0230” the product lacks turbo-speed support. If it says ”3642-0250”, the product supports turbo-speed.
Turbo-speed data rates can only be achieved if the SHDSL devices at
both ends of the connection have turbo-speed support.
The operator can either specify a fixed data rate to be used, or let the CO
and CPE discover the achievable data rate automatically.
Using Auto mode will optimise the data rate for the current SNR conditions.
ˆ Noise margin: The noise margin is the difference between the required SNR
for a certain bit rate, and the actual SNR.
When the SHDSL connection data rate is set to auto-negotiation mode, the
operator can configure an administrative noise margin (also referred to as
target noise margin or target SNR margin). A large administrative noise
margin gives robustness against SNR fluctuations. But as the required SNR
increases with data rate, specifying a a large administrative noise margin
may imply that a low data rate is negotiated.
Thus, when configuring the administrative noise margin the operator can optimise the connection for reliability (noise margin 10dB), high speed (noise
margin 3dB) or as a tradeoff thereof (normal mode, i.e., noise margin 6dB).
To monitor the quality of the connection, WeOS enables the operator to read
the current noise margin.
ˆ G.HS Threshold: The G.HS threshold setting is only needed if the units are
located in a noisy environment with SHDSL line cables of good quality, and
where a connection can not even be established at SHDSL rate 192kbit/s.
The setting configures a higher threshold of the G.HS idle parameter in order to detect idle. The SHDSL line length capability will be affected, since
the G.HS idle threshold and the G.HS signals meet earlier when the G.HS
Threshold is raised.
When enabling GHS threshold, possible settings include ’low(750)’, ’medium
© 2017 Westermo Teleindustri AB
285
Westermo OS Management Guide
Version 4.21.1-0
(1500)’, ’high(3000)’ and a custom configured value.
Corresponding values to the fixed value settings are [low-750; medium1500; high-3000]. The custom configured value could be set in the range
[0-32767] in steps of 1.
ˆ PAF - PME Aggregation Function: PAF functionality is used to aggregate the 2
SHDSL ports on DDW-x42 (and DDW-x42-485) to achieve higher bandwidth.
The 2 "bonded" ports can reach rates from 64 kbit/s to 30,6 Mbit/s.
ˆ Low Jitter function: Low Jitter is a SHDSL port specific function that can be
used in applications where high accuracy of the Ethernet packet jitter is
needed. If enabled the jitter of the latency over the SHDSL link will be minimized.
This functionality is using a different SHDSL mode compared to default setting, thus the Low Jitter configuration must be set on both SHDSL ports sharing the physical cable.
ˆ EMF - Emergency Freeze function: EMF enabled makes the unit detect exception situations on the SHDSL links. The detection will freeze the SHDSL
transceiver parameters temporarily to keep the link up. With this function
enabled the unit might avoid a complete SHDSL retrain that could take up
to a minute. The unit may lose data even with EMF enabled, but only for a
short period of time.
Note
Only the data rate and noise margin settings of the CO are used in the SHDSL
connection. These parameters are passed to the CPE during the connection
establishment phase.
12.1.3
General port settings
The following parameters can be configured for SHDSL ports in the same way as
for Ethernet ports. The SHDSL uses Ethernet First Mile (EFM) encapsulation, thus
many Ethernet settings apply to the SHDSL ports. More detailed information is
found in chapter 10.
ˆ Port enable/disable: Ports can be disabled and enabled administratively.
ˆ Port priority mode: Define whether incoming packets should be prioritised
based on VLAN tag, VLAN ID, port ID, IP ToS, etc. See also section 10.1.4.
286
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
ˆ Port priority (level): The inbound priority associated with this port. See also
section 10.1.4.
ˆ Link alarm: Link status can be configured as an alarm source. See also
section 10.1.5.
ˆ Inbound rate limit: Setting the inbound rate limit is possible on DSL ports,
but is likely of less interest than on Ethernet ports, since the DSL data rates
are primarily limited by the rate of the DSL line. See also sections 10.1.6
and 12.1.2.
ˆ Outbound traffic shaping: Setting the outbound rate limit (traffic shaping)
is possible on DSL ports, but is likely of less interest than on Ethernet ports,
since the DSL data rates are primarily limited by the rate of the DSL line.
Furthermore, outbound traffic shaping in frames per second mode is not
availble on DSL ports. See also sections 10.1.7 and 12.1.2.
ˆ Bandwidth Statistics: Enable or disable bandwidth statistics per port. See
also section 10.1.8.
ˆ Fall-back default-VID: The fall-back default VID setting is only of interest
for the special case when untagged packets are received over a link only
associated with tagged VLANs.
Ethernet settings for port speed/duplex mode, and MDI/MDIX mode do not apply
to SHDSL ports, thus are not configurable.
Note
As of WeOS v4.21.1, enabling/disabling flow control (as described in section
section 10.1.3) has no effect on SHDSL ports.
© 2017 Westermo Teleindustri AB
287
Westermo OS Management Guide
Version 4.21.1-0
12.2
Managing SHDSL ports via the web interface
The Web interface provides configuration of SHDSL ports as well as listing of
SHDSL port statistics.
The SHDSL statistics is provided in two views – an overview with a selection of
statistics for all SHDSL ports, including some status information, and a detailed
page with a larger set of statistics.
12.2.1
List and Edit SHDSL Port Settings
Menu path: Configuration ⇒ Port ⇒ SHDSL
On this page you can list and change the settings for the SHDSL ports.
PAF
Port
CO/CPE
288
PAF aggregates the 2 SHDSL ports to achieve higher bandwidth. The functionality demands that the rate do not differ
more the 4 times between port 1 and 2 to ensure good performance. Note: This functionality is only available on DDW-x42
and DDW-x42-485. Check to enable, un-check to disable.
Default is Disabled.
The SHDSL port label.
To establish a connection between two DSL-ports, one has to
be configured as Central Office (CO) and one has to be configured as Customer Premises Equipment (CPE).
Default for port 1/1 is CPE, and default for port 1/2 is CO.
Continued on next page
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
DSL Rate
Mode
Link Alarm
Edit
Continued from previous page
Speed setting is only valid if the port is configured as CO (the
CPE rate setting is not used, since the CPE speed automatically follows the CO to which it becomes connected).
See section 12.1.2 for information on using SHDSL-turbo
speed data rates.
Default is Auto.
The noise-margin mode. The noise-margin mode setting is
only valid if the port is configured as CO (the CPE setting is not
used, since the CPE noise-margin mode automatically follows
the CO to which it becomes connected).
The CO can be configured to choose a faster less reliable
speed (High Speed), a slower more reliable speed (Reliable),
or a tradeoff between these two objectives (Normal).
Default is Normal.
When link alarm is enabled an alarm will be generated if port
link is down. Alarms trigger an SNMP trap message to be sent
and alarms to be shown on the administration web.
Click this icon to edit a port’s settings.
© 2017 Westermo Teleindustri AB
289
Westermo OS Management Guide
Version 4.21.1-0
12.2.2
Edit Port Settings
Menu path: Configuration ⇒ Port ⇒ SHDSL ⇒ PortNo ⇒
On this page you can change the settings for the port.
G.HS Threshold
290
The G.HS Threshold setting is only needed if the unit are
located in a noisy environment with SHDSL line cables of
good quality and where a connection can not even be established at SHDSL rate 192kbit/s. The setting configures
a higher threshold of the G.HS idle parameter in order
to detect idle. The SHDSL line length capability will be affected, since the G.HS idle threshold and the G.HS signals
meet earlier when the G.HS Threshold is raised.
When enabling G.HS Threshold, possible settings include
’low’, ’medium’ and ’high’.
Corresponding values to the fixed value settings are [low750; medium-1500; high-3000]
If a custom value is configured in CLI, it will be displayed
in the drop-down list.
Default is Disabled
Continued on next page
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Low Jitter
Link Fault
Forward (LFF)
Emergency
Freeze EMF
Priority Mode
Continued from previous page
The Low Jitter mode can be enabled to minimize the jitter
of the latency over the SHDSL link in applications where
high accuracy of the Ethernet packet jitter is needed. This
functionality is using a different SHDSL mode compared
to default setting, thus the Low Jitter configuration must
be set on both SHDSL ports sharing the physical cable.
Check to enable, un-check to disable.
Notice: Make sure that you have both line partners configured enabled or disabled.
Default is Disabled.
On devices with SHDSL ports, alarms can be triggered
when the remote SHDSL switch indicates it has link down
on its Ethernet port. That is, this feature can be used in
topologies where an Ethernet is extended over an SHDSL
link, and where the remote SHDSL switch (e.g. a DDW120) is able to signal that the Ethernet link is down on its
side.
Check to enable, un-check to disable.
Default is Disabled.
EMF enabled makes the unit detect exception situations
on the SHDSL links. The detection will freeze the SHDSL
transceiver parameters temporarily to keep the link up.
With this function enabled the unit might avoid a complete SHDSL retrain that could take up to a minute. The
unit may lose data even with this functionality enabled,
but only for a short period of time.
Check to enable, un-check to disable.
Default is Enabled.
Here you select on what information priority will be
based:
Port Based Based on the port’s priority. See the next
item (Priority).
IP
Based on the content of the IP ToS bits
(IPv4) or the IP TC bits (IPv6).
VLAN Tag
Based on the content of the (802.1p) priority field inside the received packet’s
VLAN tag.
Continued on next page
© 2017 Westermo Teleindustri AB
291
Westermo OS Management Guide
Version 4.21.1-0
Priority
Inbound Rate
Limit
Outbound
Traffic
Shape
Bandwidth
Statistics
292
Continued from previous page
The port’s priority level. Zero (0) is low priority and seven
(7) high priority.
Bandwidth limit for inbound traffic. Disabled means no
limiting.
Bandwidth limit for outbound traffic. Disabled means no
limiting.
Enable or disable bandwidth monitoring per port. See
Section 11.2.2 for how to view the statistics that is gathered when this function is enabled.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
12.2.3
SHDSL statistics Overview
Menu path: Status ⇒ Port ⇒ SHDSL
On the SHDSL port statistics overview page you will be presented to a selection
of static data for each port. Additional statistic numbers are presented on the
detailed view page.
Alarm
Port
Negotiation
State
State
Data Rate
Total Bytes In
Total Bytes Out
Details
Auto Refresh
Refresh
An alarm icon appears at the start of a line if there is a link
alarm on a port.
The port label. If PAF is configured, the background color of
the port identifier is pink
Current state of the DSL-line negotiation.
Possible
values are UP_DATA_MODE, INITIALISING, DOWN_READY
and DOWN_NOT_READY. Note: if no link is established
the normal state for a CO-mode configured port is
DOWN_NOT_READY, for a CPE-configured port the normal
state is DOWN_READY.
FORWARDING Unit forwards packets. Normal operation.
LEARNING
The port is preparing itself for entering
FORWARDING state.
BLOCKING
Unit does not forward any packets.
DISABLED
Port does not participate in operation.
Negotiated DSL data rate in bit/s.
Total number of bytes received on the port.
Total number of bytes sent out on the port.
Click this icon to view more detailed statistics for the port.
Click on a value to make the page reload with updated
statistics automatically every 5, 15, 30 or 60 seconds. Click
Off to turn off auto refresh.
Click on this button to reload with updated statistics.
© 2017 Westermo Teleindustri AB
293
Westermo OS Management Guide
Version 4.21.1-0
12.2.4
Detailed SHDSL Port Statistics
Menu path: Status ⇒ Port ⇒ SHDSL ⇒
When clicking the details-icon in the overview page you will be presented to the
detailed statistics page for the SHDSL port.
Link Status
Link Uptime
294
Status of link, (Up/Down). If a link-alarm is associated with this port, an alarm icon is displayed if the
link-alarm is active.
The time since link was established.
Continued on next page
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Negotiation State
Data Rate
Current SNR
Margin
Negotiations
Total Bytes
Broadcast Packets
Multicast Packets
Unicast Packets
Dropped Packets
Traffic Size, Inbound
Auto Refresh
<<Previous
Next>>
Refresh
Clear Port
Continued from previous page
Current state of the DSL-line negotiation.
Possible values are UP_DATA_MODE, INITIALISING,
DOWN_READY and DOWN_NOT_READY. Note: if no
link is established the normal state for a CO-mode
configured port is DOWN_NOT_READY, for a CPEconfigured port the normal state is DOWN_READY.
Negotiated DSL data rate in bit/s.
Signal to Noise Ratio in dB on this link.
Number of negotiations since unit startup.
Total number of bytes received (inbound) or transmitted (outbound) on this port.
Total number of good broadcast packets received
(inbound) or transmitted (outbound) on this port.
Total number of good multicast packets received (inbound) or transmitted (outbound) on this port.
Total number of good unicast packets received (inbound) or transmitted (outbound) on this port.
Total number of packets received that have been
discarded.
Number of octets received in different size categories.
Click on a value to make the page reload with updated statistics automatically every 5, 15, 30 or 60
seconds. Click Off to turn off auto refresh.
Goto statistics for previous port.
Goto statistics for next port.
Click on this button to reload with updated statistics.
Clear all statistics counters for the port shown.
© 2017 Westermo Teleindustri AB
295
Westermo OS Management Guide
Version 4.21.1-0
12.3
Managing SHDSL ports via the CLI
The table below shows SHDSL port management features available via the CLI.
Command
Configure SHDSL port settings
port [dsl|shdsl|. . . ] <PORTLIST>
[no] co
[no] speed <auto|auto-5696k|0-15304k>
[no] noise-margin
[no] ghs-threshold <low|medium|high>
[no] paf
[no] low-jitter
[no] emf
Default
Section
Auto
Normal
Disabled
Disabled
Disabled
Enabled
Section
Section
Section
Section
Section
Section
Section
Section
Port settings in common with Ethernet ports (chapter
[no] enable
Enabled
[no] priority <0-7>
0
[no] priority-mode <tag|ip|port>
tag
[no] link-alarm
Disabled
[no] rate-limit <70-2560>
Disabled
[no] traffic-shaping <70-2560>
Disabled
[no] bandwidth-statistics
Disabled
[no] default-vid <VLAN_ID>
Disabled
Show SHDSL related status and statistics
show <dsl|shdsl>
show ports
show rmon
12.3.1
10)
Section
Section
Section
Section
Section
Section
Section
Section
12.3.1
12.3.2
12.3.3
12.3.4
12.3.5
12.3.6
12.3.7
12.3.8
10.3.3
10.3.6
10.3.7
10.3.8
10.3.9
10.3.10
10.3.11
10.3.15
Section 12.3.9
Section 10.3.16
Section 11.3
Managing SHDSL port settings
Syntax port [dsl|shdsl|...]
<PORTLIST>
Context Global Configuration context
296
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Usage Enter the SHDSL Port Configuration context.
A ”PORTLIST” is a comma separated list of ranges of SHDSL ports without
intermediate spaces, e.g., ”1/1,1/2” on a slotted product, or ”1-3,5” on a
non-slotted product.
The port qualifier keyword ”shdsl” (or ”dsl”) is not needed if the numbers
in the ”PORTLIST” are unique to DSL ports.
For a more general description of the ”port” command, see section 10.3.1.
Use ”show port [dsl|shdsl] [<PORT|PORTLIST>]” port configuration information of the given PORT or PORTLIST. Alternatively, the command ”show”
can be run within the SHDSL Port Configuration context, to show the configuration of a port (or list of ports).
Default values Not applicable.
12.3.2
Setting SHDSL port mode (CO/CPE)
Syntax [no] co
Context SHDSL Port Configuration context
Usage Set the SHDSL port to operate in central office (CO) or customer premises
equipment (CPE) mode.
When connecting switches via SHDSL it is important that one side puts its
SHDSL port in CO mode (”co”) while the other side puts its SHDSL port in
CPE mode (”no co”).
Default values Factory default for DDW-225/226 is to have port 1/1 in CPE mode
(”no co”), and port 1/2 in CO mode (”co”). Factory default for DDW-x42
(and DDW-x42-485) is to have port DSL 1 in CPE mode (”no co”), and port
DSL2 in CO mode (”co”).
Use ”show co” to show whether the SHDSL port is configured to operate as
Central Office or Customer Premises Equipment.
12.3.3
Setting SHDSL port rate
Syntax [no] speed <auto|auto-5696k|0-5696k|0-15304k>
Context SHDSL Port Configuration context
© 2017 Westermo Teleindustri AB
297
Westermo OS Management Guide
Version 4.21.1-0
Usage Set SHDSL port rate, either bye specifying that auto-negotiation should
be used, or that a specific fixed rate should be used. Only the ”speed”
setting on the CO has affect on the established connection.
ˆ Auto-negotiate: Use ”speed auto”, ”speed 0”, or ”no speed” to let
the rate be auto-negotiated between the SHDSL nodes in the extended
SHDSL range 32-15288 kbps on Turbo HW; if not Turbo HW the range is
192-5696 kbps.
Use ”speed auto-5696k” to let the rate be auto-negotiated in the standard SHDSL range 192-5696 kbit/s.
ˆ Fixed rate: Use ”speed RATE”, where RATE is in range ”1k-15304k” on
products with Turbo-HW support, and in range ”1k-5096k” on products
without Turbo-HW support, to specify a fixed data rate in kbit/s. Alternatively, specify ”speed 1-15304000” and ”speed 1-5696000” respectively, to specify a fixed data rate in bit/s.
The following fixed rates are supported on all SHDSL products: 192k,
384k, 512k, 768k, 1024k, 1280k, 2048k, 2304k, 2688k, 3072k, 3456k,
3840k, 4224k, 4608k, 4992k, 5376k, and 5696k.
Products with Turbo-HW support the following additional fixed data rates:
32k, 64k, 128k, 6200k, 6712k, 7224k, 7736k, 8248k, 8760k, 9272k,
9784k, 10296k, 10808k, 11320k, 11832k, 12344k, 13112k, 13880k,
14648k and 15304k.
If other rates are specified, WeOS will round the value upwards to the
nearest supported rate.
Use ”show speed” to show the SHDSL port’s rate setting.
Default values ”speed auto”
12.3.4
Setting SHDSL port noise-margin
Syntax [no] noise-margin <reliable|normal|high-speed [nonstrict]>
Context SHDSL Port Configuration context
Usage Set SHDSL port noise-margin. Note: The noise-margin setting is only
relevant when the data rate is set to auto-negotiate (”rate 0”), see section 12.3.3).
Available noise-margin modes:
298
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
ˆ Reliable: Select ”noise-margin reliable” to let the rate auto-negotiation
optimise for reliability (rather than high data rate).
ˆ High-Speed: Select ”noise-margin high-speed” to let the rate autonegotiation optimise for high data rate (rather than reliability).
ˆ Normal: ”noise-margin normal” is the default setting for the noisemargin, which gives a tradeoff between reliability and high-speed. Alternatively, the command ”no noise-margin” can be used.
Using the parameter nonstrict after the selected noise-margin mode
will configure the unit to a less strict algorithm during the connection
phase. The resulting current SNR will not necessary match the configured noise-margin mode.
Use ”show noise-margin” to show the SHDSL port’s noise-margin setting.
Default values ”noise-margin normal”
Error messages None defined yet.
12.3.5
Setting SHDSL port G.HS Threshold
Syntax [no] ghs-threshold <low|medium|high>
Context SHDSL Port Configuration context
Usage Set SHDSL port to operate with new G.HS Threshold value. The G.HS
Threshold setting is only needed if the unit are located in a noisy environment with SHDSL line cables of good quality and where a connection can
not even be established at SHDSL rate 192kbit/s, see section 12.1.2. The
setting configures a higher threshold of the G.HS idle parameter in order to
detect idle. The SHDSL line length capability will be affected, since the G.HS
idle threshold and the G.HS signals meet earlier when the G.HS Threshold is
raised.
When enabling G.HS Threshold, possible settings include”low”, ”medium”,
”high” and a custom configured value. Corresponding values to the fixed
value settings are [low-750; medium-1500; high-3000].
The custom configured value could be set in the range [0-32767] in steps of
1.
Use ”no ghs-threshold” to disable the G.HS threshold.
© 2017 Westermo Teleindustri AB
299
Westermo OS Management Guide
Version 4.21.1-0
Use ”show ghs-threshold” to show the SHDSL port’s G.HS Threshold setting.
Default values Disabled (”no ghs-threshold”)
12.3.6
Setting SHDSL PAF mode
Syntax [no] paf
Context SHDSL Port Configuration context
Usage Set the SHDSL unit to operate in paf mode.
PAF aggregates the 2 SHDSL ports to achieve higher bandwidth. The functionality demands that the rate do not differ more than 4 times between
port 1 and 2 to ensure good performance. Port 2 must be configured to the
same role (CO/CPE) as port 1 to get the functionality working.
Note
This functionality is only available on DDW-x42 and DDW-x42-485.
Use ”show paf” to show whether PAF is enabled or not.
Default values Disabled
12.3.7
Setting SHDSL low jitter mode
Syntax [no] low-jitter
Context SHDSL Port Configuration context
Usage Set the SHDSL unit to operate in low jitter mode.
Low Jitter can be enabled to minimize the jitter of the latency over the
SHDSL link in applications where high accuracy of the Ethernet packet jitter
is needed. This functionality is using a different SHDSL mode compared to
default setting, thus the Low Jitter configuration must be set on both SHDSL
ports sharing the physical cable.
Use ”show low-jitter” to show the SHDSL port’s low-jitter setting.
Default values Disabled
300
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
12.3.8
Setting SHDSL emergency freeze mode
Syntax [no] emf
Context SHDSL Port Configuration context
Usage Set the SHDSL unit to operate in emf mode.
EMF enabled makes the unit detect exception situations on the SHDSL links.
The detection will freeze the SHDSL transceiver parameters temporarily to
keep the link up. With this function enabled the unit might avoid a complete
SHDSL retrain that could take up to a minute. The unit may lose data even
with this functionality enabled, but only for a short period of time.
Use ”show emf” to show the SHDSL port’s emergency freeze setting.
Default values Enabled
12.3.9
Show SHDSL port status
Syntax show shdsl
Context Admin Exec context.
Usage Show the status of all SHDSL ports.
Default values Not applicable.
© 2017 Westermo Teleindustri AB
301
Westermo OS Management Guide
Version 4.21.1-0
Chapter 13
ADSL/VDSL Port Management
The Falcon-206 is equipped with a xDSL port, i.e., a port capable of operating
in either ADSL or VDSL mode. Thus, the Falcon-206 can be used as customer
premises equipment (CPE), acting either as switch or router, when connecting to
an ISP over an ADSL or VDSL line.
This chapter describes how to setup and manage your xDSL port, as well as the
most common configuration steps to connect to your ISP.
13.1
Overview of ADSL/VDSL Port Management
Feature
xDSL mode (ADSL/VDSL)
xDSL carrier (POTS/ISDN)
External splitter (filter)
ADSL/ATM specific settings
ATM VPI/VCI
ATM Encapsulation
Restart/retrain xDSL link
View xDSL port configuration
View xDSL port status/statistics
xDSL settings in common
with Ethernet ports
ISP and network settings
302
Web
X
X
X
CLI
X
X
X
General Description
Section 13.1.1
Section 13.1.1
Section 13.1.1
X
X
X
X
X
X
X
X
X
X
X
X
Section 13.1.1-13.1.2
Section 13.1.1-13.1.2
X
X
Section 13.1.1, 13.1.4
Section 13.1.3
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
13.1.1
ADSL/VDSL overview
A Falcon xDSL router is typically used as a broadband router (fig. 13.1a, when
connecting a private company network to the Internet via xDSL. An alternative is
to use Falcon as a xDSL/Ethernet bridge (fig. 13.1b), to connect a single PC or an
external (non-”xDSL capable”) router to the Internet.
Internet
(via ISP DSLAM)
ADSL/VDSL
vlan1
Internet
(via ISP DSLAM)
Switch
(Bridge)
Falcon
vlan1
ADSL/VDSL
Ethernet
vlan1006 (pppoe0)
Router,
FW, NAT
Falcon
Ethernet
vlan1
Company Private
Network
a) Using Falcon as broadband router
Router
OR single
end device
External
Router
Company Private
Network
b) Using Falcon as switch (bridge)
Figure 13.1: Common ADSL/VDSL topologies: a) Using Falcon as broadband
router, or b) using Falcon as on xDSL/Ethernet switch (bridge).
When connection your Falcon xDSL unit to your ISP, you may have to configure
settings related to the xDSL port as well as IP settings specific to your xDSL
provider. To configure your Falcon router for the first time, it is recommended to
use the Web based Basic Setup Page, see section 13.2.1.
More information on xDSL settings is found below and in sections 13.1.2-13.1.4.
ˆ xDSL settings:
– ADSL or VDSL: As the Falcon can be used both for ADSL and VDSL connections, you may have to configure the xDSL mode.
Default: ADSL
© 2017 Westermo Teleindustri AB
303
Westermo OS Management Guide
Version 4.21.1-0
– POTS or ISDN carrier: Depending on the kind of telecom network used
to carry your xDSL connection, you should configure the Annex setting
accordingly: ”Annex A, I, L, and/or M” for POTS carrier networks,
or ”Annex B or J” for ISDN carrier networks.
Further details on configuration of the Annex setting:
* Annex setting for ADSL over POTS: For ADSL over POTS carrier,
”Annex A” is base annex implicitly available. Other annexes for
POTS (I, L, M, N) are extensions to Annex A. You can specify to only
use Annex A, or you can specify to use additional extensions:
· ”Annex A”: The WeOS unit announces capability of Annex A (ADSL
over POTS). This is the default setting of ADSL.
· ”Annex I”: The WeOS unit announces capability of Annex A and I.
Annex I allows for additional encoding techniques for ADSL over
POTS.
· ”Annex L”: The WeOS unit announces capability of Annex A and L.
Annex I allows for additional frequency bands (at lower POTS frequency), and longer reach.
· ”Annex M”: The WeOS unit announces capability of Annex A and M.
Annex I allows for additional frequency bands (at higher POTS
frequency).
· ”Annex L-M”: The WeOS unit announces capability of Annex A,
L and M.
* Annex setting for ADSL over ISDN: For ADSL over ISDN carrier,
”Annex B” is base annex implicitly available. Annex J is an extension
to Annex B for ADSL over ISDN. You have the following configuration
options:
· ”Annex B”: The WeOS unit announces capability of Annex B (ADSL
over ISDN).
· ”Annex J”: The WeOS unit announces capability of Annex B and J.
Annex I allows for additional encoding techniques for ADSL over
ISDN.
* Annex setting for VDSL: For VDSL it is possible to let the WeOS
unit automatically probe what carrier network is used (by choosing
”Annex A-B”); if this does not work to bring up the VDSL line, one
304
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
can manually try the individual settings ”Annex A” and ”Annex B”
respectively.
Default: ADSL Annex A (ADSL over POTS)
– Use of external Filter (Splitter) or not: If you wish to use your xDSL
connection for regular phone calls, the Falcon xDSL port should (1) be
connected to a splitter which in turn connects to the (first) telephone
jack, and (2) the Falcon xDSL ”filter” setting should be enabled.
Otherwise, the Falcon xDSL port should (1) be connected directly to the
(first) telephone jack, and (2) the Falcon xDSL ”filter” setting should
be disabled.
Default: Filter enabled (i.e., it is assumed the Falcon xDSL port is
connected via a splitter)
ˆ ADSL specific settings: When using Falcon for ADSL (as opposed to VDSL),
a few settings related to ADSL/ATM encapsulation and VPI/VCI may have to
be set, see section 13.1.2 below.
ˆ Use of PPPoE, DHCP or Static IP address assignment: xDSL providers
use different schemes to assign IP addresses to their customers. These
methods to assign an IP address are not specific to xDSL connections, thus
are explained in detail in other chapters: chapter 33 describes use of PPPoE,
and chapter 21 covers use of DHCP as well as static IP address assignment.
To simplify configuring IP settings appropriate for your xDSL subscription,
the Falcon Web interface has a Basic Setup Page, see section 13.2.1.
For those who wish to configure Falcon via the CLI, section 13.1.4 below
provides useful information.
Default: DHCP (i.e., acquire your IP address from your ISP via DHCP)
ˆ VLAN settings: By factory default, the xDSL port will belong to VLAN 1006
(untagged), while all Ethernet ports will be belong to VLAN 1 (untagged).
If the Falcon is configured to act as xDSL/Ethernet Bridge via the Basic Setup
Page (see section 13.2.1), all ports (xDSL and Ethernet) will be mapped to
VLAN 1.
© 2017 Westermo Teleindustri AB
305
Westermo OS Management Guide
Version 4.21.1-0
13.1.2
ADSL specific settings
There are two types of ADSL specific xDSL settings, and both of them concern
the use of ATM as ADSL carrier.
ˆ VPI and VCI: In WeOS you need to define the identifier of your ATM permanent virtual circuit (PVC) to your ADSL provider. This identifier contains two
parts, the virtual path identifier (VPI) and the virtual circuit identifier (VCI).
What values to use depends on your ISP provider.
Default: VPI 8 and VCI 35
ˆ ATM Encapsulation: Falcon units support two ATM encapsulation modes:
”bridged LLC” and ”bridged VC-MUX” ([9]). Which setting to use depends
on your ISP provider.
Default: bridged LLC
There is also an additional ADSL related setting to specify if the ADSL is carried
over a POTS telecom network (Annex A, I, L, or M), or an ISDN telecom network
(Annex B or J). However, as of WeOS v4.21.1 the ”annex” setting also applies to
VDSL, see section 13.1.1.
Default: Annex A (POTS)
13.1.3
xDSL settings in common with Ethernet ports
The following parameters can be configured for xDSL ports in the same way as
for Ethernet ports. In WeOS, VDSL uses Ethernet First Mile (EFM) encapsulation,
and ADSL uses ”bridged” LLC or VC-MUX encapsulation (see section 13.1.2), thus
many Ethernet settings apply to xDSL ports. More detailed information is found
in chapter 10.
ˆ Port enable/disable: Ports can be disabled and enabled administratively.
ˆ Port priority mode: Define whether incoming packets should be prioritised
based on VLAN tag, VLAN ID, port ID, IP ToS, etc. See also section 10.1.4.
ˆ Port priority (level): The inbound priority associated with this port. See also
section 10.1.4.
ˆ Link alarm: Link status can be configured as an alarm source. See also
section 10.1.5.
306
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
ˆ Inbound rate limit: Setting the inbound rate limit is possible on DSL ports,
but is likely of less interest than on Ethernet ports, since the DSL data rates
are primarily limited by the rate of the DSL line. See also sections 10.1.6
and 13.1.1.
ˆ Outbound traffic shaping: Setting the outbound rate limit (traffic shaping)
is possible on DSL ports, but is likely of less interest than on Ethernet ports,
since the DSL data rates are primarily limited by the rate of the DSL line.
Furthermore, outbound traffic shaping in frames per second mode is not
available on DSL ports. See also sections 10.1.7 and 13.1.1.
ˆ Bandwidth Statistics: Enable or disable bandwidth statistics per port. See
also section 10.1.8.
ˆ Fall-back default-VID: The fall-back default VID setting is only of interest
for the special case when untagged packets are received over a link only
associated with tagged VLANs.
Ethernet settings for port speed/duplex mode, and MDI/MDIX mode do not apply
to xDSL ports, thus are not configurable.
Note
As of WeOS v4.21.1, enabling/disabling flow control (as described in section
section 10.1.3) has no effect on xDSL ports.
13.1.4
Connecting to your ISP over an xDSL line
Recommendation: Use Basic Setup in Web
The simplest way to configure your Falcon unit to connect to your ISP is to
use the Basic Setup web page, see section 13.2.1.
This section is intended (1) for those who wish to configure the Falcon via
the CLI, and (2) for those looking for more background details on how to
configure Falcon as an xDSL router or bridge.
This section describes the most common steps to configure your Falcon xDSL
router to connect to your ISP. Although many configuration settings are affected,
setting up your ISP should be straight-forward:
ˆ The factory default configuration of xDSL are adapted to using the Falcon as
an xDSL router.
© 2017 Westermo Teleindustri AB
307
Westermo OS Management Guide
Version 4.21.1-0
ˆ On the Falcon xDSL router, the web interface includes a basic setup page,
for easy configuration of the most common use cases, see section 13.2.1.
A common setup is use Falcon as broadband router when connecting your company network towards the Internet, see fig. 13.2a.
An alternative is to use the Falcon as a xDSL/Ethernet bridge to connect a single
end device (such as a PC), or to use a separate router to connect your local
network, as shown in fig. 13.2b.
Internet
(via ISP DSLAM)
ADSL/VDSL
vlan1
Internet
(via ISP DSLAM)
Switch
(Bridge)
Falcon
vlan1
ADSL/VDSL
Ethernet
vlan1006 (pppoe0)
Router,
FW, NAT
Falcon
Ethernet
vlan1
Router
OR single
end device
Company Private
Network
a) Using Falcon as broadband router
External
Router
Company Private
Network
b) Using Falcon as switch (bridge)
Figure 13.2: Common ADSL/VDSL topologies: a) Using Falcon as broadband
router, or b) using Falcon as an xDSL/Ethernet switch (bridge) with an external
router (or single end-device such as a PC) behind.
Section 13.1.4.1 focus on using Falcon as a router, while section 13.1.4.1 covers
on how to use Falcon as a switch (bridge). Both sections assume you have configured the xDSL port settings appropriately for your xDSL subscription (see also
sections 13.1.1 and 13.1.2).
13.1.4.1
Using Falcon as a Router
By factory default, Falcon is configured as a router:
308
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
ˆ Port Segmentation: The xDSL and Ethernet ports are mapped to two
VLANs.
– WAN port: The xDSL port is mapped to VLAN 1006. VLAN 1006 had IGMP
snooping disabled, thereby avoiding sending IGMP queries towards your
ISP.
Example
vlan 1006
untagged dsl 1
no igmp
...
– LAN ports: The Ethernet ports are used as LAN ports, and are all mapped
to the default VLAN, i.e., VLAN 1. VLAN 1 has the same factory default
settings as other WeOS products.
Example
vlan 1
untagged eth 1-4
igmp
...
ˆ Network Interface Settings:
ˆ WAN interface: There are three methods to assign the IP address of the WAN
interface, and which method to use depends on your xDSL Internet Service
Provider (ISP): (1) acquire it via DHCP, (2) configure a static IP address, or
(3) acquire the IP address via PPPoE. Each method is described below.
By default, Falcon is configured to acquire the WAN interface address via
DHCP.
1. Address via DHCP: The WAN interface will by default use DHCP to get
its IP address automatically from the ISP. In addition, interface vlan1006
is assigned admin distance ”1” (section 21.2.6), in order to dynamically learn default gateway, DNS server and other global information
via DHCP. Management services such as SSH, HTTP (Web), etc. are by
default disabled to avoid unauthorised access from the public Internet.
Example
iface vlan1006 inet dhcp
distance 1
no management
end
© 2017 Westermo Teleindustri AB
309
Westermo OS Management Guide
Version 4.21.1-0
2. Static IP address: The WAN interface can be configured to get its IP
address assigned statically. If your ISP provides this option, the ISP
will inform you what address to use for your subscription. The example below uses address 192.168.5.4 and netmask 255.255.255.192 to
illustrate the method.
Example
iface vlan1006 inet static
distance 1
no management
address 192.168.5.4/26
end
With static IP assignment you would also need to set the IP address
of the default gateway and DNS server(s) (information provided by your
ISP). In the example below the default gateway has address 192.168.5.1,
and a DNS server at 192.168.5.2.
Example
ip
route default 192.168.5.1
name-server 192.168.5.2
...
end
3. Address via PPPoE: Some ISPs use PPPoE for authorisation of, and IP address assignment to, their customers. To configure a WAN interface to
use PPPoE, a PPPoE instance is created and mapped to the associated
VLAN interface (here vlan1006). This will in turn create a PPPoE interface (here pppoe0), which now acts as our WAN interface. The example below shows the default setting for the PPPoE interface; the admin
distance and management settings are automatically copied from the
configuration of interface vlan1006.
310
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Example
pppoe 0
iface vlan1006
ppp-advanced
identity [email protected] password sEcReT
end
end
...
iface pppoe0 inet dynamic
mtu 1492
tcp-mss 1412
distance 1
no management
end
As interface pppoe0 is typically used as upstreams interface, the NAT
settings should be adapted, see Routing, Firewall and NAT below.
ˆ LAN interface: The LAN interface vlan1 is by default assigned IP address
192.168.2.200. All management services are enabled on the LAN interface.
Example
iface vlan1 inet static
distance 16
management ssh http https ipconfig snmp
address 192.168.2.200/24
end
ˆ Routing, Firewall and NAT: Falcon by default has IP forwarding (routing)
and NAT enabled. Thereby Falcon can to route packets between a private
network on its LAN interface (vlan1) and the public Internet on its WAN interface.
The default firewall and NAT rules will block all incoming traffic on the WAN
interface, except for packets belonging to established connections. (Such
connections are in turn initiated from the private network, i.e., from the LAN
side.) These settings are chosen to limit the risk for security attacks when
connecting the Falcon to a public network such as the Internet.
Special firewall deny rules are set up for TCP and UDP port 53 (DNS). These
are to prevent the Falcon to become an open DNS relay on the WAN side.
Open DNS relay is considered to be a security problem and can be used for
remote attacks of the ISP’s DNS server. DNS relay is enabled on all interfaces
and should be filtered away on all interfaces facing public networks. Normal
DNS traffic originating from the inside (from the LAN) will work as expected
© 2017 Westermo Teleindustri AB
311
Westermo OS Management Guide
Version 4.21.1-0
and is not affected by these rules.
Example
ip
forwarding
firewall
policy input DROP
policy forward DROP
filter allow in vlan1 proto icmp
filter deny in vlan1006 dport 53 proto udp
filter deny in vlan1006 dport 53 proto tcp
nat type napt out vlan1006 addfilter
enable
end
Adapting Firewall and NAT rules when using PPPoE
When PPPoE is used for WAN IP address assignment (see above), the
firewall and NAT rules must be adapted accordingly, i.e., ”vlan1006”
should be replaced by ”pppoe0” as shown in the example below.
Example
ip
forwarding
firewall
policy input DROP
policy forward DROP
filter allow in vlan1 proto icmp
filter deny in pppoe0 dport 53 proto udp
filter deny in pppoe0 dport 53 proto tcp
nat type napt out pppoe0 addfilter
enable
end
ˆ Other Configurations: The items above cover the most important configuration settings when connecting a Falcon to your ISP. Notes on a few more
settings are given below:
– RSTP: Westermo switches running WeOS typically have RSTP enabled on
all Ethernet and DSL ports. However, the xDSL port on Falcon have RSTP
disabled by default. For more information on RSTP, see chapter 18.
– VPN: Its possible to use the Falcon as a VPN gateway. For more information on configuring VPNs in WeOS, see part IV.
– DHCP Server: For information on how to make your Falcon act as DHCP
server on your local network (vlan1), see chapter 22.
312
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
13.1.4.2
Using Falcon as a Switch (Bridge)
As shown in fig. 13.2b, it is possible to use the Falcon as a xDSL/Ethernet bridge.
That is, the xDSL port does not have to be used as a dedicated router port;
instead the Falcon could switch packets between Ethernet and xDSL ports, given
that they are mapped to the same VLAN (see chapter 15).
Although it is possible to make the Falcon work as a regular WeOS switch, there
are some differences:
ˆ Falcon is a router by default: All WeOS devices can be configured to act as
router or switch. The difference is that Falcon is configured as router in its
factory default setting (able to route between the WAN interface vlan1006
and the LAN interface vlan1, while other WeOS devices act as switches by
default (all ports on VLAN 1).
ˆ Layer-2 Redundancy (RSTP/FRNT): As the xDSL port is used to connect to a
xDSL provider (ISP), the remote end is managed by an external organisation.
Thus, layer-2 redundancy protocols such as RSTP and FRNT should not be
used on the xDSL port; for FRNT this is prohibited, and for RSTP it is disabled
by default.
The simplest way to configure your Falcon to act as a switch is by using the Basic
Setup Page in the Web interface (section 13.2.1). This way, all ports (Ethernet
and xDSL) will be mapped to VLAN 1. The Falcon will then be accessible via the
default IP address (IP address 192.168.2.200, netmask 255.255.255.0) unless you
have changed the IP settings of interface vlan1. As an alternative to using the
Basic Setup Page, you could achieve the corresponding result by removing VLAN
1006, either via the Web interface (section 15.3) or via the CLI (section 15.4) as
shown below.
Example
falcon:/#> show vlan
VID Name
Oper Untagged/Tagged
---- ---------------- ---- --------------------------------------------------1 vlan1
1006 vlan1006
DOWN U:eth 1-4
T:
DOWN U:dsl 1
T:
-----------------------------------------------------------------------------falcon:/#>
© 2017 Westermo Teleindustri AB
313
Westermo OS Management Guide
Version 4.21.1-0
Example
falcon:/#> configure
falcon:/config/#> no vlan 1006
falcon:/config/#> end
Port dsl 1 did not belong to any VLAN, setting as untagged in VLAN 1.
vlans: Problem activating settings.
There was some problem activating your configuration changes!
This could result in a non-functional system, continue anyway (y/N)? y
OK, accepting configuration anyway -- please review the running configuration.
Stopping DHCP Clients ...................................... [ OK ]
Configuration activated. Remember "copy run start" to save to flash (NVRAM).
falcon:/#> show vlan
Press Ctrl-C or Q(uit) to quit viewer, Space for next page, <CR> for next line.
VID Name
Oper Untagged/Tagged
---- ---------------- ---- --------------------------------------------------1 vlan1
UP
U:ALL
T:
-----------------------------------------------------------------------------falcon:/#> cp running-config startup-config
falcon:/#>
There are additional setting you may consider changing when running the Falcon
as a switch:
ˆ Limit remote management: When the xDSL port is mapped to VLAN 1, the
Falcon will be open for remote management via the xDSL port just as it is
via the Ethernet ports on VLAN 1.
This is usually no problem, as the Falcon by default is assigned the default IP
address (192.168.2.200) on interface vlan1, and that address is not routable
via the ISP. However, if limiting remote management is still a concern, you
could, e.g., remove the IP address of interface vlan1. The Falcon can then
be managed only via the console port (CLI) instead.
Example
falcon:/#> configure
falcon:/config/#> iface vlan1
falcon:/config/iface-vlan1/#> inet static
falcon:/config/iface-vlan1/#> show address
192.168.2.200/24
falcon:/config/iface-vlan1/#> no address
falcon:/config/iface-vlan1/#> leave
Configuration activated. Remember "copy run start" to save to flash (NVRAM).
falcon:/#>
ˆ As switch there is no firewall: Acting as a switch, the Falcon no longer serves
as a firewall towards the Internet. You should therefore ensure that you
314
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
protect your local network, typically by running firewall in an external router
(or in a directly attached PC), see fig. 13.2b.
ˆ Disable IGMP Snooping: VLAN 1 has IGMP snooping enabled by default. This
should be fine even when the xDSL port is on VLAN 1, however, if you have
concerns about running IGMP snooping on a port towards your ISP you can
disable IGMP snooping on VLAN 1.
Example
falcon:/#> configure
falcon:/config/#> vlan 1
falcon:/config/vlan-1/#> no igmp
falcon:/config/vlan-1/#> leave
Stopping IGMP Snooping daemon .............................. [ OK ]
Configuration activated. Remember "copy run start" to save to flash (NVRAM).
falcon:/#>
ˆ Disable IP Forwarding: As long as all ports are mapped to VLAN 1, the Falcon
will act as a switch, even though the IP forwarding configuration option is
enabled. However, if you have concerns about having IP forwarding enabled,
you can disable it.
If you use the Basic Setup Page in the Web interface (section 13.2.1) to
configure the Falcon as switch (bridge), IP forwarding will be disabled automatically.
Example
falcon:/#> configure
falcon:/config/#> ip
falcon:/config/ip/#> no forwarding
falcon:/config/ip/#> leave
Configuration activated. Remember "copy run start" to save to flash (NVRAM).
falcon:/#>
© 2017 Westermo Teleindustri AB
315
Westermo OS Management Guide
Version 4.21.1-0
13.2
Managing ADSL/VDSL ports via the web interface
The Web interface provides configuration of xDSL ports (sections 13.2.1-13.2.3)
as well as listing of xDSL port statistics.
The xDSL statistics is provided in two views – an overview with a selection of
statistics for all xDSL ports, including some status information (section 13.2.4),
and a detailed page with a larger set of statistics (section 13.2.5).
13.2.1
Basic Setup for Falcon DSL router
Menu path: Basic Setup
This feature requires a JavaScript enabled web browser. To simplify the setup of
the Falcon unit for remote access, a basic setup page is provided with the most
basic settings compiled into one view. In many cases this page may be sufficient
for setting up the Falcon for remote access.
Note
When you enter the basic setup page and make changes to the configuration
and press the apply button, some settings will be reset. See section 13.2.1.1
below for more information.
Figure 13.3: Basic Setup Profile and Mode
To set up the switch using the Basic Setup, two fundamental settings have to be
set first. These two settings control the other options displayed on the page.
Router
WAN Profile
Bridged
316
The unit will be set up as a router with a firewall
protecting the LAN side from the WAN side.
The unit will act as a plain switch.
Continued on next page
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
DHCP
Router Mode
PPPoE
Static
Continued from previous page
The WAN side will expect a DHCP-server to provide the switch with an IP address.
The WAN side will set up a PPPoE connection
with the ISP to provide the internet connection.
The IP address, netmask and gateway will be
manually entered.
The DHCP router mode (shown above) does not need any additional settings. The
Static IP and PPPoE router modes require additional settings as described below.
Irrespective of the selected router mode, you may also need to fill out ADSL/VDSL
port settings, as shown at the end of this section.
Static IP Settings
Figure 13.4: Basic Setup Static IP
If the static IP mode is selected you are asked to fill in the following entries.
Address
Netmask
The IPv4 address to assign to the interface.
The netmask for the IPv4 address. Identifies which IP addresses are located on the same subnet.
Continued on next page
© 2017 Westermo Teleindustri AB
317
Westermo OS Management Guide
Version 4.21.1-0
Default
Gateway
Continued from previous page
Statically configured default gateway of the unit. This is the
IP address of the gateway to send packages to when no more
specific route can be found in the routing table. This value
overrides any value retrieved dynamically (e.g. using DHCP).
Leave empty to enable dynamically retrieved gateway address
or if no default gateway should be available.
PPPoE Settings
Figure 13.5: Basic Setup PPPoE
If the PPPoE mode is selected you are asked to fill in the following entries.
Username
Password
318
The username provided by the PPPoE provider.
The password provided by the PPPoE provider.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
DSL Settings
Figure 13.6: Basic Setup DSL settings
In addition you may have to change the DSL settings if they do not satisfy the
requirements from your ISP, see fig. 13.6.
Mode
ATM
Encapsulation
ATM PVC
Framing
Annex
Filter
Specify whether the xDSL port should operate ADSL port
or VDSL port. Default: ADSL
ATM encapsulation. Default: LLC
Set the appropriate VPI and VCI for the ATM PVC. Default:
VPI 8, VCI 35
Annex A or B can be set for either ADSL or VDSL mode.
Annex L, M, L-M, I and J can only be set for ADSL. The
annex I and J options are extensions of ADSL annex A
and B. The annex L and M options are extensions of ADSL
annex A. The annex A-B option is only available for VDSL
mode. Default: Annex A (POTS)
External splitter or not. POTS/ISDN filter. Default: Enabled
© 2017 Westermo Teleindustri AB
319
Westermo OS Management Guide
Version 4.21.1-0
13.2.1.1
Basic Setup Behavior
As noted above, some settings will be reset when applying the basic setup page.
This is what will happen:
ˆ When applying bridged profile:
All but one VLAN is removed and its interface settings are reset.
Details:
– All VLANS are removed and VLAN 1 re-created. As a result of this, all
advanced settings on VLAN 1 and it’s associated interface will be lost.
– All ports are associated untagged to VLAN 1.
– The firewall is removed.
– IP-forwarding (routing) is turned off.
ˆ When applying a router profile:
All settings for LAN and WAN and its associated interfaces are reset. Firewall
rules are reset. All existing PPPoE configurations are removed.
Details:
– All VLANS are removed and VLAN 1 (LAN) and VLAN 1006 (WAN) recreated. As a result of this, all advanced settings on VLAN 1 and VLAN
1006 and their associated interfaces will be lost.
– The DSL port is associated untagged to VLAN 1006 (WAN).
– All remaining ports are associated untagged to VLAN 1 (LAN).
– The firewall is removed and then re-created. This will result in loss of all
current NAT, Port forwarding and Access rules.
– IP-forwarding (routing) is turned on.
In addition for the DHCP and Static modes:
– A NAT-rule for external interface VLAN 1006 (WAN) and internal VLAN 1
(LAN) is added.
– Firewall filtering rules denying inbound UDP and TCP port 53 (DNS) are
added for the external interface VLAN 1006 (WAN).
In addition for the PPPoE mode:
– A PPPoE configuration is added.
320
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
– A NAT-rule for the PPPoE interface (WAN) and internal VLAN 1 (LAN) is
added.
– Firewall filtering rules denying inbound UDP and TCP port 53 (DNS) are
added for the PPPoE interface (WAN).
Note
Firewall filtering of inbound UDP and TCP port 53 is added to prevent the unit
to become an open DNS relay on the WAN side.
Open DNS relay is considered to be a security problem and can be used for
remote attacks of the ISP’s DNS server. DNS relay is enabled on all interfaces
and should be filtered away on all interfaces facing public networks. Normal
DNS traffic originating from the inside (from the LAN) will work as expected
and is not affected by these rules.
© 2017 Westermo Teleindustri AB
321
Westermo OS Management Guide
Version 4.21.1-0
13.2.2
List and Edit ADSL/VDSL Port Settings
Menu path: Configuration ⇒ Port ⇒ DSL
When entering the DSL configuration page you will be presented to a list of all
DSL ports available on your switch, see fig. 13.7.
Figure 13.7: DSL Port configuration settings overview
Alarm
Port
Enabled
Type
Link Alarm
Enabled
There is an active link alarm associated with the port.
Only shown if link alarm is enabled and the link is down.
The port label.
A green check-mark means the xDSL port is enabled, and
a dash means it is disabled.
ADSL or VDSL
When link alarm is enabled an alarm will be generated if
port link is down. Alarms trigger an SNMP trap message
to be sent and alarms to be shown on the administration web. In the ports overview table a green check-mark
means enabled, and a dash means disabled.
Edit
Click this icon to edit a port’s settings.
Restart
Click this icon to retrain the DSL ports.
To change the settings for a specific xDSL port you will have to click the edit icon
which will take you to the DSL port setting edit page see section 13.2.3.
322
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
13.2.3
Edit xDSL Port Settings
Menu path: Configuration ⇒ Port ⇒ DSL ⇒
Figure 13.8: DSL port configuration settings edit page
On this page you can change the settings for the xDSL port.
Enabled
Mode
ATM Encapsulation
ATM PVC Framing
Enable or Disable the port
Specify whether the xDSL port should operate ADSL port
or VDSL port. Default: ADSL
ATM encapsulation. Default: LLC
Set the appropriate VPI and VCI for the ATM PVC. Default:
VPI 8, VCI 35
Continued on next page
© 2017 Westermo Teleindustri AB
323
Westermo OS Management Guide
Version 4.21.1-0
Annex
Filter
Priority Mode
Port Priority
Inbound Rate
Limit
Outbound Traffic
Shape
Link Alarm
Bandwidth
Statistics
324
Continued from previous page
Annex A or B can be set for either ADSL or VDSL mode.
Annex L, M, L-M, I and J can only be set for ADSL. The
annex I and J options are extensions of ADSL annex A
and B. The annex L and M options are extensions of ADSL
annex A. The annex A-B option is only available for VDSL
mode. Default: Annex A (POTS)
External splitter or not. POTS/ISDN filter. Default: Enabled
Here you select on what information priority will be
based:
Port Based Based on the port’s priority. See the next
item (Priority).
IP
Based on the content of the IP ToS bits
(IPv4) or the IP TC bits (IPv6).
VLAN Tag
Based on the content of the (802.1p) priority field inside the received packet’s
VLAN tag.
The port’s priority level.
Bandwidth limit for inbound traffic
Bandwidth limit for outbound traffic
When link alarm is enabled an alarm will be generated if
port link is down. Alarms trigger an SNMP trap message
to be sent and alarms to be shown on the administration
web.
Enable or disable bandwidth monitoring per port. See
Section 11.2.2 for how to view the statistics that is gathered when this function is enabled.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
13.2.4
ADSL/VDSL statistics Overview
Menu path: Status ⇒ Port ⇒ DSL
On the DSL port statistics overview page you will be presented to a selection
of static data for each port. Additional statistic numbers are presented on the
detailed view page.
Note: If only one DSL port is present in the unit, you will be redirected to the
detailed statistics and status page.
Alarm
Port
Negotiation State
State
Downstream Rate
Upstream Rate
Total Bytes In
Total Bytes Out
Details
Auto Refresh
Refresh
An alarm icon appears at the start of a line if there is a
link alarm on a port.
The port label.
Current state of the DSL-line negotiation.
Link state
Negotiated DSL downstream rate in bit/s.
Negotiated DSL upstream rate in bit/s.
Total number of bytes received on the port.
Total number of bytes sent out on the port.
Click this icon to view more detailed statistics for the
port.
Click on a value to make the page reload with updated
statistics automatically every 5, 15, 30 or 60 seconds.
Click Off to turn off auto refresh.
Click on this button to reload with updated statistics.
© 2017 Westermo Teleindustri AB
325
Westermo OS Management Guide
Version 4.21.1-0
13.2.5
Detailed ADSL/VDSL Port Statistics
Menu path: Status ⇒ Port ⇒ DSL ⇒
If only one DSL port is present in the unit, or when clicking the details-icon in the
overview page you will be presented to the detailed statistics page for the DSL
port.
326
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Link Status
Link Uptime
DSL mode
Negotiation State
Negotiations
Remote Vendor Name
Rate
SNR (dB)
Line attn (dB)
Signal attn (dB)
Output power (dBm)
Traffic Counters
Traffic Size, Inbound
Auto Refresh
<<Previous
Next>>
Refresh
Clear Port
Status of link, (Up/Down). If a link-alarm is associated with this port, an alarm icon is displayed if
the link-alarm is active.
The time since link was established.
ADSL or VDSL
Current state of the DSL-line negotiation.
Number of negotiations since unit startup.
Identifier string of DSLAM vendor.
Negotiated DSL downstream and upstream rate in
bit/s.
Upstream and Downstream Signal to Noise Ratio
(SNR) in dB on this link.
Line attenuation is the loss of signal over distance,
in dB, downstream and upstream.
Signal attenuation in dB, downstream and upstream.
Output power in dBm, downstream and upstream.
See section 11.2.2 for details.
See section 11.2.2 for details.
Click on a value to make the page reload with updated statistics automatically every 5, 15, 30 or
60 seconds. Click Off to turn off auto refresh.
Go to statistics for previous port. Only shown if
more than one DSL port available.
Go to statistics for next port. Only shown if more
than one DSL port available.
Click on this button to reload with updated statistics.
Clear all statistics counters for the port shown.
© 2017 Westermo Teleindustri AB
327
Westermo OS Management Guide
Version 4.21.1-0
13.3
Managing ADSL/VDSL ports via the CLI
The table below shows xDSL port management features available via the CLI.
Command
Configure ADSL and VDSL port settings
port [dsl|xdsl|. . . ] <PORTLIST>
[no] mode <adsl [annex <a|b|i|j|l|m|l-m>] |
vdsl [annex <a|b|a-b>]>
[no] filter
ADSL specific port settings
mode adsl
[no] encap <llc|vcmux>
[no] pvc <VPI/VCI>
Default
Section
adsl annex a
Section 13.3.1
Section 13.3.2
Enabled
Section 13.3.3
llc
8/35
Section 13.3.4
Section 13.3.5
Port settings in common with Ethernet ports (chapter 10)
[no] enable
Enabled
[no] priority <0-7>
0
[no] priority-mode <tag|ip|port>
tag
[no] link-alarm
Disabled
[no] rate-limit <70-2560>
Disabled
[no] traffic-shaping <70-2560>
Disabled
[no] bandwidth-statistics
Disabled
[no] default-vid <VLAN_ID>
Disabled
Show ADSL/VDSL related status and statistics
show dsl
show ports
show rmon
13.3.1
Section
Section
Section
Section
Section
Section
Section
Section
10.3.3
10.3.6
10.3.7
10.3.8
10.3.9
10.3.10
10.3.11
10.3.15
Section 13.3.6
Section 10.3.16
Section 11.3
Managing xDSL port settings
Syntax port [dsl|xdsl|...]
<PORTLIST>
Context Global Configuration context
328
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Usage Enter the xDSL Port Configuration context.
A ”PORTLIST” is a comma separated list of ranges of xDSL ports without
intermediate spaces, e.g., ”1/1,1/2” on a slotted product, or ”1-3,5” on a
non-slotted product.
The port qualifier keyword ”xdsl” (or ”dsl”) is not needed if the numbers
in the ”PORTLIST” are unique to DSL ports.
For a more general description of the ”port” command, see section 10.3.1.
Use ”show port dsl <PORT|PORTLIST>” or ”show port xdsl <PORT|
PORTLIST>” to list port configuration information for the given xDSL port(s).
Also available as ”show” command within the xDSL Port Configuration context.
Default values Not applicable.
Entering the xDSL configuration context on a Falcon:
Example
falcon:/#> configure
falcon:/config/#> port dsl 1
falcon:/config/port-dsl1/#>
Listing configuration information on the xDSL port on a Falcon:
Example
falcon:/config/#> show port dsl 1
xDSL -------------------------------------------- Priority ---- Limit - Default
Port
Ena Mode Filter Encap PVC
Annex Alarm Mode Level
In | Out
Vid
===============================================================================
DSL 1
YES adsl
YES
llc 8/35
A
NO tag
0
None None
Auto
===============================================================================
falcon:/config/#>
13.3.2
Setting xDSL port mode (ADSL or VDSL) and carrier type
Syntax [no] mode <adsl [annex <a|b|i|j|l|m|l-m> | vdsl [annex <a|b|a-b>>
Context xDSL Port Configuration context
Usage Specify whether the xDSL port should operate as ADSL port or VDSL port,
and
© 2017 Westermo Teleindustri AB
329
Westermo OS Management Guide
Version 4.21.1-0
ˆ ADSL:
– Use ”mode adsl annex <a|i|l|m|l-m>” to specify ADSL mode over
a POTS carrier network.
– Use ”mode adsl annex <b|j>” to specify ADSL mode over an ISDN
carrier network.
– For further information on ADSL Annex settings, see section 13.1.1.
When selecting ADSL mode, the ADSL specific settings ”encap” (section 13.3.4) and ”pvc” (section 13.3.5) are enabled.
ˆ VDSL:
– Use ”mode vdsl annex a” to specify VDSL mode over a POTS carrier network.
– Use ”mode vdsl annex b” to specify VDSL mode over an ISDN carrier network.
– Use ”mode vdsl annex a-b” to auto-detect whether your VDSL connection is over a POTS or ISDN carrier. This alternative is usually
preferable for VDSL connections due to its simplicity, but the autodetection mechanism may experience problems on long copper cables. If this is the case, please try ”mode vdsl annex a” or ”mode
vdsl annex b” depending on the carrier type of your VDSL connection.
Use ”no mode” to reset the mode setting to the default value.
Use ”show mode” to show whether the xDSL port is set to operate as ADSL or
VDSL port, and the type of carrier network used, Annex A, Annex I, Annex L
or Annex M (POTS) or Annex B or Annex J (ISDN). Annex I and J not supported
in VDSL mode.
Default values ADSL over POTS (”mode adsl annex a”)
13.3.3
Specify whether external splitter is used or not
Syntax [no] filter
Context xDSL Port Configuration context
Usage Specify whether a (external) splitter is used or not, i.e., is the Falcon unit
connected directly to the telephone jack or via a splitter.
330
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Use command ”filter” if a splitter is used, and ”no filter” if no splitter
is used.
Use ”show filter” to show the xDSL port’s filter setting.
Default values ”filter” (i.e., an external splitter is assumed by default)
13.3.4
Configure ADSL/ATM encapsulation type
Syntax [no] encap <llc|vcmux>
Context xDSL Port Configuration context (only available when ADSL mode is
used, see section 13.3.2)
Usage Specify whether bridged LLC or bridged VC-MUX ATM encapsulation is
used. What encapsulation option to use depends on your ADSL provider.
Use command ”llc” to use bridged LLC and ”vcmux” to use bridged VCMUX encapsulation.
Use ”no encap” to reset the encapsulation mode to the default setting.
Use ”show encap” to show the xDSL port’s ADSL/ATM encapsulation setting.
Default values ”llc”
13.3.5
Configure ADSL/ATM VPI and VCI
Syntax [no] pvc <VPI/VCI>
Context xDSL Port Configuration context (only available when ADSL mode is
used, see section 13.3.2)
Usage Specify the VCI and VPI used for the ATM PVC by your ADSL provider.
Some examples: ”pvc 0/38” is common in U.K., ”1/32” is common in Germany, while ”pvc 8/35” is common for many other ADSL providers inside
and outside Europe.
Use ”no pvc” to reset the PVC to use default VPI/VCI. (In future versions of
WeOS the use of ”no pvc”, as well as the default PVC setting, may change.)
Use ”show pvc” to show the ATM PVC setting, i.e., which VPI and VCI are
configured.
© 2017 Westermo Teleindustri AB
331
Westermo OS Management Guide
Version 4.21.1-0
Default values ”pvc 8/35”
13.3.6
Show xDSL port status
Syntax show dsl
Context Admin Exec context.
Usage Show the status of all xDSL ports.
Default values Not applicable.
Example
falcon:/#> show dsl
Port, DSL mode
: DSL 1, ADSL/Anx-A
Channel, role
: channel 0, role CPE
Link state, uptime
: UP, 0 Days 0 Hours 7 Mins 15 Secs
Negotiation state
: Sync state, 4 changes since boot
Remote vendor name
: GSPN
Downstream -----------------------Rate
: 8000 kbps
SNR
: 12.5 dB
Line attn
: 8.3 dB
Signal attn : 8.2 dB
Output power: N/A
Upstream -------------------------Rate
: 832 kbps
SNR
: 12.0 dB
Line attn
: 7.0 dB
Signal attn : 7.0 dB
Output power: 12.4 dB
falcon:/#>
332
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Chapter 14
Power Over Ethernet (PoE)
Some WeOS Viper products[58, 60, 62] have Ethernet ports with support for
Power Over Ethernet (PoE[20] and PoE+[21]).
This chapter gives an overview of PoE support in WeOS products (section 14.1).
Sections 14.2 and 14.3 concern PoE management support via the Web interface
and CLI. PoE related SNMP support is covered in chapter 6, while management of
PoE alarms/events is documented in chapter 24.
As of WeOS v4.21.1, PoE management via LLDP[21] is not supported.
14.1
Overview of Power over Ethernet (PoE)
Feature
Per-Port PoE Configuration
Enable/Disable
Allocation Priority
Power Limit
PoE Status
Consumed power
Allocated Power
Detected PoE Units
© 2017 Westermo Teleindustri AB
Web
CLI
General Description
X
X
X
X
X
X
Section 14.1.2
-”-
X
X
X
X
X
X
Sections 14.1.1-14.1.2
Section 14.1.1
333
Westermo OS Management Guide
Version 4.21.1-0
14.1.1
PoE Power Classes
When plugging in a PoE unit to a PoE port on the switch, the switch will detect the
class of the connected PoE unit, depending on the unit’s resistance and thereby
its maximum power consumption.
Table 14.1 lists the maximum power consumption for units of the different PoE
classes, as well as the (somewhat higher) power actually allocated by the switch,
which considers cable losses. Thus, when admitting a class 0 unit, the switch
allocates 15.4 W to ensure 12.94 W reach the PoE unit.
PoE
Class
0
1
2
3
4
Max Unit Power
Consumption (W)
(Pcss,nt )
12.94
3.84
6.49
12.95
25.50
Allocated
Power (W)
(Pcss,oc )
15.4
4.0
7.0
15.4
30.0
Table 14.1: Power allocated to and consumed by units of different PoE classes.
It is also possible to configure a maximum power limit on each individual PoE port,
see section 14.1.2. The power allocated on the port then becomes the minimum
of the (a) configured power limit, and (b) the power allocated for attached unit’s
class (as listed in table 14.1).
The following additional classification is made for the connected unit depending
on resistance:
ˆ Good: Ok. A PoE unit is connected. (Resistance within specification of PoE
class 0-4).)
ˆ Open: Ok. Port not connected. (”Infinite” resistance, i.e., open circuit).
ˆ Short: Ok, when non-PoE unit is connected. (Resistance determined as short
circuit.)
ˆ Low: The connected unit is detected as a PoE unit and served, although its
the resistance is too low to meet the PoE specification (and too high to be
determined as short circuit (non-PoE unit connected)).
ˆ High: The connected unit is detected as a PoE unit and served, although its
334
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
the resistance is too high to meet the PoE specification (and too low to be
determined as unconnected (open circuit)).
14.1.2
Allocation of PoE Power
There is maximum value for the amount of power a PoE switch can deliver (Pstch,m ),
see the User Guide of your PoE product, listed in section 1.5, for more information
on max output power. When more power is requested than available, the switch
will stop/refuse1 delivering power on the port(s) with lowest priority.
14.1.2.1
Calculating available power, and per-port power limitation
As of WeOS v4.21.1, PoE power is always allocated to handle max consumption
by all admitted PoE units. For each port, the max consumption (Pport,m ) is
calculated as the minimum value of:
ˆ (Pcss,oc ): The power allocated to units of the attached class (see right
column of table 14.1).
ˆ (Pport−mt ): The power limit configured for the port (if any).
The available power is calculated as max output power of the switch1 , minus the
sum the max power for all (admitted) ports.
X
Pbe = Pstch,m −
Pport,m
dmtted
If a new PoE unit is attached, its Pcss,oc will be compared to Pbe :
ˆ If there is enough power available to serve the new unit, it will be admitted.
ˆ If there is not enough power available to serve the new unit, the switch will
deliver power to the ports with highest priority (see below). Thus, to admit
the new unit, one (or more) of the already admitted units will be declined
power.
1 To
compensate for limited accuracy in measured power consumption, your WeOS PoE unit may
allow the measured and allocated power to raise somewhat above the stated Pstch,m of the
product, before power delivery is stopped/refused on some port. The customer should still ensure
that PoE equipment attached to the WeOS PoE switch do not use more than Pstch,m in total.
© 2017 Westermo Teleindustri AB
335
Westermo OS Management Guide
Version 4.21.1-0
Preference
Order
Port
Name
Configured
Priority
Tie-break
Priority
1 (highest)
2
3
4
5
6
7
8 (lowest)
X7
X9
X10
X5
X6
X1
X2
X3
critical
critical
critical
high
high
low
low
low
6
7
8
4
5
1
2
3
Table 14.2: Example of allocation preference order for a given PoE priority configuration on a 12-port Viper PoE unit[60].
14.1.2.2
PoE Port Priority
There are three levels of PoE priority (low, high, critical), which can be configured
per port. If there is not enough power to serve all attached PoE units, preference
will be given to ports with higher priority.
As situations can occur where the switch must chose between two ports of the
same level of configured priority, there is a need for a second level ”tie-break”
priority. This tie-break priority is allocated in ascending port order, giving the
lowest PoE port the highest tie-break priority. The ”tie-break” priority is as follows
(starting with the highest tie-break priority):
ˆ 12-port Viper PoE units[58, 60]: X1(1), X2(2), X3(3), X5(4), X6(5), X7(6),
X9(7), X10(8)
ˆ 20-port Viper PoE units[62]: X5(1), X6(2), X7(3), X11(4), X12(5), X13(6),
X17(7), X18(8)
The tie-break priority order may become configurable in future WeOS releases.
Table 14.2 illustrates the power allocation preference order in a specific configuration example on a 12-port Viper PoE unit, where ports X7-X10 have been configured with priority critical, X5-X6 with priority high, and X1-X3 have priority low.
336
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
14.2
Managing PoE via the web interface
The Web interface provides configuration of PoE ports as well as listing of global
and port specific PoE status.
14.2.1
List PoE Settings
Menu path: Configuration ⇒ PoE
When entering the PoE configuration page you will be presented to a list of all
PoE ports available on your switch, and their settings.
Port
PoE Enabled
Priority
The port label. (Only PoE capable Ethernet ports are listed.)
Shows if PoE is enabled or disabled on the port.
Shows the configured PoE priority (Low, High or Critical) for
the port.
Continued on next page
© 2017 Westermo Teleindustri AB
337
Westermo OS Management Guide
Version 4.21.1-0
Continued from previous page
Shows the configured Power Limit for the port (in Watts), or
Disabled if no port specific limit has been set.
Limit
Edit
14.2.2
Click this icon to edit a port’s PoE settings.
Edit PoE Port Settings
Menu path: Configuration ⇒ PoE ⇒
On this page you can change the PoE settings for the port.
Enabled
Priority
Power Limit
338
Enable/disable PoE on the port
PoE power allocation priority. When more power is requested than available, power will be dropped on the
ports with lowest priority. Possible values:
ˆ Low (Shut down first)
ˆ High
ˆ Critical (Shut down last)
See section 14.1.2 for more information.
Set port specific power limit. Allowed values are 1-30
(Watts), or Disabled (i.e., no port specific power limit).
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
14.2.3
PoE Status
Menu path: Status ⇒ PoE
On the PoE port status page you will be presented to global and port specific PoE
status data.
© 2017 Westermo Teleindustri AB
339
Westermo OS Management Guide
Version 4.21.1-0
Maximum Power
Allocated Power
Consumed Power
Power Usage
Port
PoE Enabled
Priority
Power Limit
Class
Consumed Power
Detection Details
Auto Refresh
Refresh
340
Global Status
The maximum power (in Watts) the switch is able to
deliver.
Allocated power (in Watts). See section 14.1.2 for
information on allocation and classes.
The total power consumed on all PoE ports.
Percentage of available power currently consumed
(i.e., ”Consumed Power”/”Maximum Power”).
Port Status
The PoE port label.
Shows if PoE is enabled or disabled on the port.
Shows the configured PoE priority (Low, High or Critical) for the port.
Shows the configured power Limit for the port (in
Watts), or Disabled if no port specific limit has been
set.
Shows the PoE class (0-4) of the connected PoE unit,
or Unknown if the class cannot be determined.
Currently consumed power (in Watts) by the connected PoE unit.
Additional details on the unit connected to the PoE
port (see also the Class column):
ˆ Unknown: Unit Resistance Unknown (e.g. PoE
disabled on port)
ˆ Short: Non-PoE unit connected
ˆ Low: PoE Unit connected. Resistance OK, but
low
ˆ Good: PoE unit connected. Resistance Good.
ˆ High: PoE Unit connected. Resistance OK, but
high
ˆ Open: Nothing Connected
Click on a value to make the page reload with updated status automatically every 5, 15, 30 or 60
seconds. Click Off to turn off auto refresh.
Click on this button to reload with updated status.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
14.3
Managing PoE via the CLI interface
Command
Configure PoE settings
poe
[no] port <PORTLIST|all>
[no] enable
[no] priority <low|high|critical>
[no] limit <1-30>
Default
Section
Enabled
Low
Disabled
Section
Section
Section
Section
Section
Show PoE status
show poe [full] [port <PORTLIST>]
14.3.1
14.3.1
14.3.2
14.3.3
14.3.4
14.3.5
Section 14.3.6
Manage PoE Settings
Syntax poe
Context Global Configuration context.
Usage Use command ”poe” to enter PoE configuration context.
Use ”show poe” to list global PoE settings and per-port PoE settings, also
available as ”show” command within the PoE configuration context.
Default values Not applicable.
14.3.2
Manage per-port PoE settings
Syntax [no] port <PORTLIST|all>
Context PoE configuration context.
Usage Enter PoE Port configuration context.
Use ”port <PORTLIST>” to configure per-port settings for the PoE ports in
the given list, e.g., ”port X2”, or ”port X1-X5,X10”. Use ”port all” to
configure per-port settings for all PoE ports.
Use ”no port <PORTLIST>” to reset PoE port settings to their default values
for the given port range.
© 2017 Westermo Teleindustri AB
341
Westermo OS Management Guide
Version 4.21.1-0
Use ”show” to list global PoE settings and per-port PoE settings. Use ”show
port <PORTLIST|all>” to show port specific settings for some or all PoE
ports (also available as ”show” command within the PoE Port configuration
context).
Default values Not applicable
14.3.3
Enable/Disable PoE on a PoE port
Syntax [no] enable
Context PoE Port configuration context.
Usage Enable/disable PoE on this port. Use ”enable” to enable, and ”no enable”
to disable PoE on this port.
Use ”show enable” to show whether PoE is enabled (or disabled) on this
port.
Default values Enabled (”enable”)
14.3.4
Set PoE allocation priority
Syntax [no] priority <low|high|critical>
Context PoE Port configuration context.
Usage Configure PoE allocation priority setting (”priority low” is the lowest
priority, while ”priority critical” is the highest .
”no priority” will reset priority to default (”priority low”).
See section 14.1.2 for information on how to select between ports of the
same configured priority.
Use ”show priority” to show PoE allocation priority setting on this port.
Default values Low (”priority low”)
14.3.5
Set PoE Power Limit
Syntax [no] limit <1-30>
342
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Context PoE Port configuration context.
Usage Configure specific PoE Power limit (in Watts) on this port, e.g., ”limit
20” to limit the delivered power to 20 Watts. The max power delivered
is the minimum value of the limit configured for this port, and the power
allocated for class of PoE unit attached (see Pcss,oc in table 14.1).
Use ”no limit” to remove port specific power limits (max power based
solely on Pcss,oc ).
Use ”show limit” to show PoE power limit setting on this port.
Default values Disabled (”no limit”)
14.3.6
Show PoE Status
Syntax show poe [full] [port <PORTLIST>]
Context Admin Exec context.
Usage Show PoE global and per port status.
Use ”show poe” (or ”show poe port <PORTLIST>”) to list global PoE status
information, and a status summary for all PoE ports (or a given subset).
Use ”show poe full” (or ”show poe full port <PORTLIST>”) to list global
PoE status information, and detailed status information for all PoE ports (or
a given subset).
Default values Not applicable
© 2017 Westermo Teleindustri AB
343
Westermo OS Management Guide
Version 4.21.1-0
Chapter 15
Virtual LAN
WeOS supports static port based VLANs and VLAN tagging according to IEEE
802.1Q[18]. In addition, WeOS supports WeOS Adaptive VLAN Trunking (AVT)
to simplify VLAN configuration in larger WeOS networks.
Section 15.1 provides general information about the VLAN properties and VLAN
management features in WeOS. This section also covers features available to
manage and inspect the MAC forwarding database on WeOS devices.
Section 15.3 covers VLAN settings via the Web interface, and section 15.4 covers
VLAN and MAC forwarding database settings via the CLI.
15.1
Overview of VLAN Properties and Management
Features
Table 15.1 summarises VLAN management features in WeOS. Section 15.1.1 provides general VLAN information and sections 15.1.2-15.1.6 contain further information on specific VLAN features.
15.1.1
Introduction to VLANs
Virtual LAN (VLAN) technology is used to create a set of separate LANs over a
single physical LAN infrastructure. Each VLAN constitutes a broadcast domain,
and traffic on one VLAN is (logically) isolated from traffic on another VLAN. WeOS
344
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Feature
General VLAN functionality
Enable/disable dynamic VLAN
Per VLAN functionality
Add/modify/delete VLAN
Enable/disable VLAN
VLAN name
Untagged/Tagged ports
VLAN priority
IGMP Snooping
VLAN CPU Channel
Forbid ports
Port-based network access control
View VLAN settings
View VLAN status
Web
CLI
X
X
Sec. 15.1.7
X
X
Secs. 15.1.1-15.1.3
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
MAC forwarding database functionality
Set MAC aging timeout
Set static MAC filters
X
X
View forwarding database settings
View forwarding database status
X
X
General Description
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
15.1.1
15.1.4
15.1.5
15.1.6
15.1.7
15.2
Sec. 15.1.8
Sec. 15.1.8
Table 15.1: Summary of VLAN management features.
supports creation of static port based VLANs and VLAN tagging as described further in this section. We start with two examples to explain the terms untagged
and tagged.
Fig. 15.1 shows a situation where three networks, the ADMIN VLAN, the OFFICE
VLAN, and the MARKETING VLAN share a single switch.
ˆ Each VLAN is assigned a VLAN identifier, a VLAN ID (VID); in this example
VIDs 1 (ADMIN), 2 (OFFICE) and 3 (MARKETING).
ˆ Each VLAN is assigned a set of ports. In this example ports 1/1-1/2 are
associated with the ADMIN VLAN, Ports 2/1-2/4 with the OFFICE VLAN, and
ports 2/5-2/8 with the MARKETING VLAN.
© 2017 Westermo Teleindustri AB
345
Westermo OS Management Guide
Version 4.21.1-0
Switch
1/1 1/2
VLAN 1
ADMIN
2/1 2/2 2/3 2/4 2/5 2/6 2/7 2/8
VLAN2
OFFICE
VLAN3
MARKETING
Figure 15.1: VLANs sharing a single switch.
In this example we have assumed that only regular hosts (PCs, servers, etc.;
not other switches) attach to the ports of the switch. Traffic sent and received
on each switch port are regular Ethernet packets (without VLAN headers), and
here we refer to this by saying that the switch ports are associated with their
respective VLAN untagged.
Note
A port associated untagged on a VLAN, will send and receive regular Ethernet packets (i.e., without VLAN header) on that port.
Consider the case where a PC attached to port 2/1 of the switch in fig. 15.1
transmits a broadcast packet. That packet will be forwarded onto all other ports
of VLAN 2 (OFFICE), i.e., ports 2/2-2/4, but not to any of the other ports.
Fig. 15.2 shows a situation where three networks, the ADMIN VLAN, the OFFICE
VLAN, and the MARKETING VLAN share two switches as well as the connection
between them.
Switch B
Switch A
1/1 1/2
VLAN 1
ADMIN
2/1 2/2 2/3 2/4 2/5 2/6 2/7 2/8
VLAN2
OFFICE
VLAN3
MARKETING
1/1 1/2
VLAN 1
ADMIN
2/1 2/2 2/3 2/4 2/5 2/6 2/7 2/8
VLAN2
OFFICE
VLAN3
MARKETING
Figure 15.2: VLANs sharing two switches and the connection between them.
346
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
ˆ As in the previous example, each VLAN is assigned a VID; in this example
VIDs 1 (ADMIN), 2 (OFFICE) and 3 (MARKETING).
ˆ Each VLAN is assigned a set of ports. (For simplicity of this example, we
have chosen to use the same port assignment on both switches.) Port 1/1
is associated (untagged) with the ADMIN VLAN, Ports 2/1-2/4 are associated (untagged) with the OFFICE VLAN, and ports 2/5-2/8 are associated
(untagged) with the MARKETING VLAN.
In addition, port 1/2, where the cable between the two switches is connected, is
associated with all three VLANs. In order for the switches to distinguish which
VLAN a packet belongs to when transmitted over a shared connection, the switch
will insert a VLAN header (VLAN tag) into the packet, which includes information
about the VLAN ID (here 1, 2 or 3). Thus, in this example port 1/2 would be
associated with VLAN 1, 2 and 3 tagged1 .
Note
A port associated tagged on a VLAN, will send and receive tagged Ethernet
packets (i.e., Ethernet packets including a VLAN header) on that port.
Consider the case where a PC attached to port 2/1 of switch A in fig. 15.2 transmits a broadcast packet. That packet will be forwarded onto ports 2/2-2/4 of
switch A untagged, and onto port 1/2 of switch A tagged with VID 2. When the
tagged packet is received on port 1/2 on switch B, that switch can determine that
the packet belongs to VLAN 2, and will forward it onto ports 2/1-2/4 untagged.
Note
A port cannot be associated with more than one VLAN untagged. A port
cannot be associated both untagged and tagged with the same VLAN.
We refer to the VLAN with VID 1 as the switch default VLAN. Ports not associated
with any VLAN (untagged or tagged) will automatically be associated with the
default VLAN. Section 15.1.3 provides more information on the default VLAN.
For each VLAN on a switch, an associated network interface will be created. The
name of a VLAN network interface is vlan<VID>, e.g., vlan1 for VLAN 1, and
vlan100 for VLAN 100. The network interface can be assigned an IP address
(IPv4), and the switch can then be managed remotely via that VLAN. It is also
1 It is recommended that a port, which is shared between several VLANs, is associated tagged
with all those VLANs, however, it is possible to configure the port untagged on one VLAN and
tagged on all other VLANs without risk for ambiguity.
© 2017 Westermo Teleindustri AB
347
Westermo OS Management Guide
Version 4.21.1-0
possible to route IP traffic between network interfaces. For more information on
network interfaces and routing, see chapter 21.
Internally, a WeOS switch can have one or more channels to the CPU, where each
channel has a capacity of 100 Mbit/s or 1000 Mbit/s. Section 15.1.6 describes how
VLANs can be mapped to different CPU channels to achieve increased routing
performance.
Layer-2 priority was described in a previous chapter, see section 10.1.4. In addition to different per port priority settings, it is possible to assign specific layer-2
priority per VLAN, see section 15.1.4.
The switch supports efficient distribution of IP multicast packets by use of IGMP
snooping. See section 15.1.5 for more information on per VLAN IGMP snooping
features.
The switch provides support for dynamic VLANs by WeOS Adaptive VLAN Trunking (AVT). AVT can be used to simplify VLAN configuration in larger WeOS LAN
infrastructures. AVT is described further in section 15.1.7.
15.1.2
Supported number of VLANs and VLAN integrity
Every VLAN needs to be associated with a unique VLAN ID (VID).
ˆ Switches support configuration of up to 64 simultaneous VLANs2 .
ˆ Valid VIDs for configuration are in range 1-4094.
ˆ Some VLAN IDs are reserved for specific use - currently this concerns a set
of VIDs in use by the FRNT protocol, see section 16.1.3.
Switches only accept packets for VLANs to which the inbound port is associated.
Additional rules for accepting a packet is described below:
ˆ When an untagged packet is received on a port, that packet will be mapped
to the port’s default VID. If the port is associated with that VLAN (tagged or
untagged), the packet will be accepted, otherwise dropped.
ˆ The port’s default VID will be the VID of the VLAN to which the port is associated untagged. If the port is not associated untagged to any VLAN, the
default VID is set to the fall-back default-VID (see also section 10.1.11) if
configured, otherwise to VID 1.
2 Special restriction on DDW-x42/DDW-x42-485: On these products the limit is 60 VLANs when
FRNT is configured on the unit, and 64 VLANs when FRNT is not configured.
348
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
ˆ Priority tagged packets, i.e., packets with VID 0, will be associated with the
port’s default VID.
ˆ Typically tagged packets (VID in range 1-4094) or priority tagged packets
(VID 0) are only accepted on ports where there is at least on VLAN associated
tagged. In addition, the packet will only be accepted if the inbound port is
associated (untagged or tagged) the VLAN of the packet.
A common MAC address database is used for all VLANs (shared VLAN learning).
15.1.3
Switch default VLAN
In WeOS the VLAN with VID 1 (VLAN 1) is denoted as the switch default VLAN.
Ports not associated with any VLAN (neither untagged nor tagged) will automatically be configured untagged to the switch default VLAN. This could happen when
a port is removed from a VLAN, or when a whole VLAN is removed.
Note
The main purpose of the switch default VLAN is to avoid loss of remote manageability of a switch due to a change in the VLAN configuration. Without
a default VLAN, you risk to lose remote access to the switch if the ports
used to connect to the switch are removed from all VLANs (unintentionally
or deliberately).
With the default VLAN feature, the switch is still manageable via those ports,
given that proper IP and firewall settings are configured for the network
interface associated with the switch default VLAN.
The switch default VLAN cannot be removed. However, it is possible to remove
all ports from the default VLAN by assigning them to other VLANs.
15.1.4
VLAN Priority
It is possible to assign an IEEE 802.1p priority to a VLAN. This feature can be
useful when an operator likes to assign a higher priority to traffic on a certain
VLAN, e.g., a VLAN dedicated for IP telephony.
When a VLAN priority is configured, all packets associated with that VLAN will
be treated according to the given VLAN priority, rather than basing the packet’s
priority on VLAN tag priority, IP ToS/DiffServ or inbound port identifier. For more
information on layer-2 priority, see section 10.1.4.
© 2017 Westermo Teleindustri AB
349
Westermo OS Management Guide
Version 4.21.1-0
15.1.5
IGMP Snooping and VLANs
Switches use IGMP snooping for efficient distribution of IP(v4) multicast over the
LAN. With IGMP snooping enabled on a VLAN, IP multicast packets will only be
forwarded onto ports leading to a receiver of that IP multicast address, or to
ports assumed to lead to an IP multicast router.
With IGMP snooping disabled on a VLAN, multicast traffic will be forwarded on all
ports of that VLAN, i.e., it is treated similar to broadcast traffic.
By default IGMP snooping is enabled on each newly created VLAN. More information on IGMP Snooping and IGMP Snooping settings is found in chapter 20.
15.1.6
Mapping VLANs to a CPU channel
A switch can have one or more channels to the switch CPU, each with a capacity of
100 Mbit/s or 1000 Mbit/s3 . By default every new VLAN (with a network interface)
is mapped to CPU channel ”0” (zero).
On devices with multiple CPU channels increased routing performance may be
achieved by assigning different VLANs to different CPU channels. Assume VLANs
1 and 2 are mapped to the same CPU channel of 100 Mbit/s capacity. Then the
maximum theoretical routing throughput between the two VLAN interfaces is 50
Mbit/s full duplex, while the maximum theoretical routing throughput would be
100 Mbit/s full duplex if these VLANs were mapped to different CPU channels.
Note
Routing performance may also be limited by CPU performance, packet size
and enabled services.
A VLAN can only be mapped to a single CPU channel.
15.1.7
Dynamic VLANs
WeOS provides dynamic VLAN support via the WeOS Adaptive VLAN Trunking
(AVT) protocol. With AVT enabled, VLAN configuration on inter-switch links is
3 WeOS products with ”Corazon” or ”Coronet” platform (see section 1.5) have 1000 Mbit/s channels to CPU, while others have 100 Mbit/s channels.
350
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
simplified - once a switch detects that it is connected to another switch, all VLANs
defined on the local switch will automatically be added to that port, see fig. 15.3.
Future versions of WeOS may include dynamic VLAN support via the standard
IEEE GVRP[18] protocol in addition to AVT.
15.1.7.1
Determining Inter-Switch Ports
To determine if a port on a switch is connected to another switch, AVT will utilise
information from the FRNT and RSTP protocols:
ˆ FRNT: If FRNT is enabled on the switch, any port configured as an FRNT port
will be classified as an inter-switch port by AVT. If FRNT is disabled, or if
the FRNT port configuration is changed, AVT will adapt its inter-switch port
classification accordingly. For more information on FRNT, see chapter 16.
ˆ RSTP: If RSTP is enabled on a port, AVT will consider the reception of an
RSTP or STP message as a sign that it is connected to another switch on
the receiving port. The port will continue to be classified as an inter-switch
port until the link goes down or until RSTP is disabled on that port. For more
information on RSTP, see chapter 18.
15.1.7.2
Dynamic addition/deletion of VLANs to Inter-Switch Ports
Once a port has been defined as an inter-switch port, that port will dynamically
be associated (tagged) with all VLANs configured on the switch. The exception is
when that port has been configured in association mode forbid on some VLAN(s)
- the port will not be associated with those VLANs.
Further details of the mechanism to associate VLANs dynamically to an interswitch port are given below:
ˆ Association mode of dynamically added VLANs: All VLANs configured on the
switch will be associated tagged by AVT. This applies even to those VLANs
configured untagged on that port. Fig. 15.3 shows an example.
© 2017 Westermo Teleindustri AB
351
Westermo OS Management Guide
Version 4.21.1-0
Switch A
Switch B
(AVT and RSTP enabled)
1/1 1/2
(AVT and RSTP enabled)
2/1 2/2 2/3 2/4 2/5 2/6 2/7 2/8
VLAN 1
(untagged)
VLAN2
(untagged)
VLAN3
(untagged)
1/1 1/2
VLAN 1
(untagged)
2/1 2/2 2/3 2/4 2/5 2/6 2/7 2/8
VLAN2
(untagged)
VLAN3
(untagged)
a) before connecting the switches
Switch A
(AVT and RSTP enabled)
1/1 1/2
Switch B
(AVT and RSTP enabled)
2/1 2/2 2/3 2/4 2/5 2/6 2/7 2/8
VLAN 1
(untagged)
VLAN2
(untagged)
VLAN3
(untagged)
1/1 1/2
VLAN 1
(untagged)
2/1 2/2 2/3 2/4 2/5 2/6 2/7 2/8
VLAN2
(untagged)
VLAN3
(untagged)
VLAN 1, 2 and 3 (all tagged)
b) after connecting the switches
Figure 15.3: Using Adaptive VLAN trunking (AVT) to dynamically add VLANs to
inter-switch ports.
Note
As AVT only considers the VLANs configured on the (local) switch when
adding VLANs to an inter-switch port, the operator of the LAN infrastructure should ensure that all switches have the same set of VLANs
defined. Otherwise the VLANs forwarded by different switches will be
inconsistent, resulting in lack of full connectivity on some VLAN(s).
ˆ Removing dynamically added VLANs: When a port loses its status as interswitch port, all VLANs dynamically added to that port will be removed. The
port will then only be associated with the VLANs it has been configured with,
and with association mode (tagged or untagged) according to the configuration.
ˆ Prohibiting that a VLAN is added to a port: It is possible to prohibit that
some VLAN(s) is dynamically added to a port even when AVT is enabled.
352
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
This feature is useful when the unit acts as a routing switch, where traffic
between some ports should be routed rather than switched.
To prohibit that a VLAN is dynamically added to a port, that port should be
configured with association mode forbid on that VLAN.
As of WeOS version v4.21.1 the forbid association mode only hinders a
port to be added to a VLAN dynamically via AVT. Ports not configured untagged/tagged with any VLAN will still be mapped to the switch default VLAN
(VLAN 1), irrespective if that port is configured as forbid on VLAN 1. For more
information about the switch default VLAN, see section 15.1.3.
15.1.7.3
Prohibit disabling of Inter-Switch Ports
A port determined as inter-switch port by AVT will not be possible to disable by
management (Web, CLI, SNMP, etc.). This feature is added in order to avoid
unintentional loss of connectivity to the switch.
15.1.8
MAC forwarding database
WeOS switches maintain a MAC forwarding database holding information about
where to forward packets for each known MAC address. As of WeOS v4.21.1 a
single MAC forwarding database is used for all VLANs, referred to as shared VLAN
learning in [18].
15.1.8.1
Managing Unicast MAC addresses
When the switch comes up, it will not know which stations are attached to its
ports. The switch inspects the destination MAC address of each incoming packet
without finding a match in the forwarding database - unknown unicast MAC addresses will be broadcasted on all ports of the associated VLAN.
The switch will automatically learn the location of stations in the LAN, by inspecting the source MAC address of each incoming packet. Once it knows on which
port a certain MAC address resides, all future packets to that station will be forwarded only onto that port.
© 2017 Westermo Teleindustri AB
353
Westermo OS Management Guide
Version 4.21.1-0
Note
Switches ”learn” the location of (unicast) MAC address by inspecting the
”source” MAC address, while they ”forward” packets based on the ”destination” MAC address.
Unicast MAC addresses learnt automatically will stay in the MAC forwarding database
until they are aged out – the aging timeout defaults to 300 seconds. The aging
timeout is configurable, and aging can be disabled.
15.1.8.2
Managing Broadcast and Multicast MAC addresses
Packets transmitted to the broadcast MAC address (”ff:ff:ff:ff:ff:ff”) will be forwarded onto all ports in the associated VLAN. Other group MAC addresses (here
referred to as multicast MAC addresses) are handled differently if IGMP Snooping
is enabled or not (see chapter 20 for detailed information on IGMP Snooping):
ˆ IGMP Snooping Disabled: With IGMP Snooping disabled on a VLAN, packets
sent to multicast MAC addresses will be handled in the same way as broadcast, i.e., such packets will be forwarded onto all ports in the associated
VLAN.
ˆ IGMP Snooping Enabled: With IGMP Snooping enabled on a VLAN, packets
sent to multicast MAC addresses will be blocked on all ports by default, and
only forwarded onto ports (1) where the switch has learnt that there is a host
interested in receiving traffic to that multicast MAC address, or (2) which the
switch believes lead to a multicast router.
WeOS also allows an operator to manually specify where to forward multicast MAC
addresses, i.e., the operator can add static multicast MAC filters. This feature is
useful for several reasons:
ˆ IGMP snooping and non-IP multicast: With IGMP snooping enabled, all MAC
multicast will be blocked, except those learnt via IGMP snooping. As IGMP
snooping only learns MAC multicast based on IP multicast, all other types of
MAC multicast will be blocked.
Adding static MAC filters enables the use of non-IP multicast on VLANs where
IGMP snooping is enabled.
ˆ IGMP Snooping and IP multicast in the 224.0.0.X range: IP multicast in the
224.0.0.X range should be forwarded onto all ports in the VLAN irrespective
if any host has indicated interest in that multicast address via IGMP or not.
354
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
In WeOS the operator has the flexibility to select which addresses in the
224.0.0.X range to forward on a LAN, by adding filters for the corresponding multicast MAC address. The factory default configuration includes MAC
filters for some of the most common multicast addresses in the 224.0.0.X
range, which are then forwarded onto all ports even if IGMP snooping is
enabled.
When specifying the destination port list in a MAC filter, one can specify both
regular Ethernet (and DSL) ports, as well as the internal CPU port(s) of the switch.
The latter is used if the multicast packet should be processed by the switch itself.
15.2
Port-based network access control
WeOS supports port-based network access control (PNAC). This security feature
is used to stop unauthorised PCs or other equipment to access the network. Authentication is required to gain access. WeOS provides two authentication methods: IEEE 802.1X and MAC based authentication.
Ports with access control enabled (i.e., controlled ports) will by default be ”blocked”
for incoming traffic. Only when a connected device has successfully authenticated itself will it be allowed/authorised to send data through the port. Packets
from unauthorised devices are still dropped, i.e., only packets with a source MAC
address of devices authorised via 802.1X or MAC authentication are allowed.
Incoming broadcast and multicast packets from unauthorised devices will also be
blocked. Outgoing broadcast and multicast packets will, however, not be blocked
and are sent out as usual on controlled ports. IGMP joining of multicast groups will
not work for unauthorised clients, as incoming IGMP join messages are dropped
until the client is granted access.
In WeOS, port-based network access control is managed per VLAN. Enabling access control on a VLAN implies that all untagged ports on that VLAN are subject
to access control by default. Often some or a few ports need to be excluded from
access control, e.g., ports connected to a server, uplink ports (towards Internet),
and VLAN trunk ports. These ports can be excluded by a special configuration
option in the CLI ”except-auth” (see section 15.4.17) or in the web GUI (see
section 15.3.5).
© 2017 Westermo Teleindustri AB
355
Westermo OS Management Guide
Version 4.21.1-0
Access controlled
ports
Authenticated
client
Switch
PC
1
5
PC
2
6
3
7
4
8
Server
Blocked
ports
INTERNET
Figure 15.4: Port-based network access control
Port-based network access control and VLAN trunk ports
As of WeOS v4.21.1, port-based network access control is only working as
expected for access ports, i.e., ports only associated with a single VLAN (untagged). VLAN trunk ports (ports associated tagged to one or more VLANs)
should be excluded from access control. Although it is possible to have access control enabled on such ports, the behaviour is neither defined nor
supported, and may change in future WeOS releases.
In order to acquire access, the connected device needs to authenticate itself to
the switch. See fig. 15.4 for a scenario. The PC on port 1 has authenticated itself, whereas the one on port 2 has not. The first PC is able to access the server
or the Internet connection on ports 6 and 8. The second PC or anything connected to ports 3 or 4 will be blocked by the switch until they have authenticated
themselves.
The two authentication mechanisms available in WeOS for port-based network
access control are described further below: IEEE 802.1X in section 15.2.1 and
MAC based authentication in section 15.2.2.
356
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
15.2.1
Authentication using IEEE 802.1X
WeOS units are able to act as IEEE 802.1X [19] authenticators. WeOS uses
the RADIUS[39] protocol with extensions for Extensible Authentication Protocol
(EAP[38]) to communicate to a backend authentication server.
WeOS neither includes a RADIUS server nor a local authentication server mechanism for 802.1X. Instead the 802.1X authentication server must be provided
externally.
As of WeOS v4.21.1, WeOS does not support Authenticator initiation as defined
by §8.4.2.1 in the IEEE 802.1X standard[19]. The 802.1X client (supplicant) must
initiate the authentication procedure to gain access4 .
Fig. 15.5 illustrates the principles of a successful authentication with IEEE 802.1X.
In reality the protocol exchanges several messages between the supplicant, the
authenticator and the RADIUS backend server (see the standard documents for
details). The WeOS unit acts as an IEEE 802.1X authenticator, relaying the EAP
messages to the RADIUS server.
When configuring the 802.1X authenticator in WeOS, the RADIUS server (or group
of RADIUS servers) must be specified. The procedure is as follows:
1. RADIUS server settings (AAA): Enter the appropriate settings for your RADIUS server(s): IP address, password, etc. See chapter 9 on Authentication,
Authorisation and Accounting (AAA) for more information.
2. Define RADIUS server group (AAA): (Optional) The RADIUS servers can be
grouped together, simplifying configuration in some cases. See chapter 9
on AAA for more information.
3. Define AAA instance(s) for 802.1X (AAA): To allow individual RADIUS servers
or server groups to be used as 802.1X authentication backends, they need
to be listed in an 802.1X AAA instance. See chapter 9 on AAA for more
information.
4. Enable 802.1X per VLAN: When 802.1X is enabled on a VLAN, the relevant
AAA instance is defined, thereby defining which RADIUS server(s) to relay
802.1X messages to from this VLAN. See sections 15.3.4 (Web) and 15.4.15
(CLI) for further details.
4 The 802.1X supplicants included with Microsoft Windows, Ubuntu Linux and most other equipment supports supplicant initiation.
© 2017 Westermo Teleindustri AB
357
Westermo OS Management Guide
Version 4.21.1-0
Authentication request with IEEE 802.1X
802.1X capable
client
Switch
RADIUS
EAP−Request
EAPOL Request
PC
Access controlled
ports
1
5
2
6
3
7
4
8
RADIUS
INTERNET
Successful authentication reply with IEEE 802.1X
Switch
RADIUS
EAP−Success
EAPOL Reply
PC
1
5
2
6
3
7
4
8
RADIUS
Unblocked
by WeOS
INTERNET
Figure 15.5: Principles of authentication with IEEE 802.1X and RADIUS
15.2.2
Authentication based on MAC addresses
Authentication can be based on the client’s MAC address. This is often combined
with IEEE 802.1X authentication to grant access to 802.1X capable devices and
legacy equipment lacking 802.1X support. When combined, MAC authentication
will have precedence over 802.1X authentication.
MAC based authentication is not as secure as IEEE 802.1X. Devices are granted
358
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
access based on the MAC address without any cryptographic authentication exchange, and it is fairly easy to modify the MAC address on a PC and most other
equipment.
MAC authentication is set up using lists of one or more MAC address patterns.
MAC patterns may contain a wild-card at the end to match a whole range of addresses. Examples: The pattern 00:11:22:33:44:55 matches exactly one address,
while the pattern 00:AA:BB:* matches all addresses beginning with 00:AA:BB.
When enabling MAC authentication on a VLAN in WeOS, the associated MAC list
(white-list) must be specified. The procedure is as follows:
1. Create MAC Authentication List (AAA): Create a MAC list, and add MAC patterns to that list. A MAC pattern by default applies to all ports on the VLAN
the MAC list will be mapped to, however, the MAC pattern may apply to a
specific port. See chapter 9 on Authentication, Authorisation and Accounting (AAA) for more information, in particular sections 9.3.26-9.3.29 (CLI),
and 9.2.20 (Web).
2. Enable MAC authentication per VLAN: When MAC authentication is enabled
on a VLAN, the relevant MAC list is specified, thereby defining which MAC
addresses to grant access. Access is granted on all ports, except for MAC
patterns limited to a specific port. See sections 15.3.4 (Web) and 15.4.15
(CLI) for further details.
The switch will listen on the controlled ports for Ethernet packets originating from
currently unknown MAC addresses. When such a packet arrives, it will use the
packet’s source MAC and search through the specified MAC list for a matching
entry. If one is found, the port will be opened for the specific MAC address.
Packets that do not match will be discarded (alternatively, such packets can be
authentication via 802.1X).
A port will remain open for an authorised MAC as long as traffic flows. If no packets is received through the port from an authorised MAC address for 5 minutes5 ,
the port will be closed again for this address, and the authentication procedure
will be re-done when new packets arrive.
As of WeOS v4.21.1 does not support MAC based authentication with a backend
authentication server (e.g, RADIUS).
5 MAC
aging time is by default 5 minutes, see sections 15.1.8.1 and 15.4.2 for more information.
© 2017 Westermo Teleindustri AB
359
Westermo OS Management Guide
Version 4.21.1-0
15.3
Managing VLAN settings via the web interface
Menu path: Configuration ⇒ VLAN ⇒ VLANs
When entering the VLAN configuration page you will be presented to a list of all
VLANs configured on your switch, see below. Here you get an overview of the
settings for all VLANs and you can create or delete VLANs. The default VLAN
(VID 1) cannot be removed (see section 15.4.6). To change the settings for a
specific VLAN, click the edit icon which will take you to the VLAN settings edit
page.
VID
Name
Enabled
Status
Prio
IGMP
Interface
360
The VLAN’s unique identifier.
The name of the VLAN. Automatically generated from VLAN
identifier when the VLAN is created using the web tool.
Used to enable or disable a VLAN. Ports on a disabled VLAN
are temporarily moved to the system default VLAN. A green
check-mark means the VLAN is enabled, and a dash means it
is disabled.
Current operational status of the VLAN, Up or Down.
VLAN priority setting. Values between 0-7 or disabled. See also
section 15.1.4. Disabled is shown using a dash.
In the VLAN overview table a green check-mark means that
IGMP snooping is enabled, and a dash means it is disabled, on
a specific VLAN. See section 15.1.5 for more information.
A list of associated interfaces.
Continued on next page
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Port(s)
New VLAN
Continued from previous page
List of ports assigned to each VLAN. Grouped as tagged and
untagged for ports configured statically to this VLAN, or as
dynamic for ports dynamically added to this VLAN by WeOS
Adaptive VLAN Trunking (AVT). (See section 15.1.7 for more
information on AVT).
1/1-1/3 means port 1/1, 1/2 and 1/3, the first and last port, and
all ports in-between.
Click this button to create a new VLAN. You will be presented
to a form where you can configure the new VLAN.
Edit
Click this icon to edit a VLAN.
Delete
Click this icon to remove a VLAN. You will be asked to acknowledge the removal before it is actually executed.
© 2017 Westermo Teleindustri AB
361
Westermo OS Management Guide
Version 4.21.1-0
15.3.1
Edit VLAN settings using the web interface
Menu path: Configuration ⇒ VLAN ⇒ VLANs ⇒
When clicking the Edit icon for a VLAN you will be presented to the VLAN edit
page.
On VLAN Edit page you can change the settings for the VLAN as described below:
VID
Enabled
Name
362
The VLAN’s unique identifier. You cannot change the VID of an
already created VLAN.
Used to enable or disable a VLAN. Ports on a disabled VLAN
are temporarily moved to the system default VLAN. To enable
the VLAN - check the box, to disable un-check the box.
The name of the VLAN. You cannot change the VLAN name
using the web tool.
Continued on next page
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Prio
IGMP
Port
Continued from previous page
VLAN priority setting. Values between 0-7 or disabled. See also
section 15.1.4. Select the desired VLAN priority in the drop
down list, or select disable to disable VLAN priority.
To enable IGMP snooping on this VLAN - check the box, to disable IGMP un-check the box. See section 15.1.5 for more information.
The ports on your switch is grouped as on the actual hardware,
in slots. To assign a port to the VLAN, check the Tagged or
Untagged check-box located underneath the port label. In the
picture above you see all ports but 2/3 associated untagged to
VLAN 1.
A port may not be associated tagged and untagged to the
same VLAN at the same time. It may not be associated untagged to more than one VLAN at a time. If you associate a
port untagged to a VLAN any existing untagged association to
another VLAN on that port will automatically be removed. You
will be notified if this happens. For more information on the
tagged and untagged association modes, see section 15.1.1.
The Forbidden check-box is used to specify that this port can
not be dynamically assigned to this VLAN (see section 15.1.7
for more information on dynamic VLANs).
© 2017 Westermo Teleindustri AB
363
Westermo OS Management Guide
Version 4.21.1-0
15.3.2
Create a new VLAN using the web interface
Menu path: Configuration ⇒ VLAN ⇒ VLANs ⇒ New VLAN
When clicking the New VLAN button you will be presented to the new VLAN page.
The New VLAN and the Edit VLAN pages differ only by the possibility to change
the VID (VLAN ID). See section 15.3.1 for additional attribute descriptions.
VID
Name
364
The VLAN’s unique identifier.
The VLAN name will be automatically generated when using
the web management tool. The name is shown directly when
you change and leave the VID field if your browser is JavaScript
enabled, otherwise it will be generated when you click the
Apply button.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
15.3.3
Managing Dynamic VLAN using the web interface
This enables WeOS Adaptive Dynamic Trunking (AVT) on the switch. For more
information on AVT in section 15.1.7.
Menu path: Configuration ⇒ VLAN ⇒ Dynamic
© 2017 Westermo Teleindustri AB
365
Westermo OS Management Guide
Version 4.21.1-0
15.3.4
Managing port-based network access control using the
web interface
Menu path: Configuration ⇒ VLAN ⇒ Port Access
The VLAN Port Access page shows an overview of the currently configured VLANs
with the port-based network access control settings.
VID
Name
802.1X
MAC auth
Excluded Ports
Edit
366
The VLAN’s unique identifier.
The name of the VLAN.
The description of the referenced 802.1X configuration, a dash means it is disabled. See section 9.2.17
for configuration of 802.1X.
The description of the referenced MAC authentication configuration, a dash means it is disabled. See
section 9.2.20 for configuration of MAC authentication
List of ports on this VLAN that are excluded from
port access control.
Click this icon to edit the port access configuration
for this VLAN.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
15.3.5
Edit port-based network access control settings
Menu path: Configuration ⇒ VLAN ⇒ Port Access ⇒
When clicking the Edit icon for a VLAN you will be presented to the VLAN Port
Access edit page.
VID
Name
802.1X settings
The VLAN’s unique identifier.
The name of the VLAN.
Enable IEEE 802.1X authentication for ports on this
VLAN by selecting a 802.1X configuration. See section 9.2.17 for how to create and edit the 802.1X
configurations.
Continued on next page
© 2017 Westermo Teleindustri AB
367
Westermo OS Management Guide
Version 4.21.1-0
MAC Auth settings
Excluded Ports
368
Continued from previous page
Enable MAC based authentication by selecting a
configuration. See section 9.2.20 for managing MAC
authentication configurations.
The ports on your switch is grouped as on the actual
hardware, in slots. Check the box underneath the
port label to exclude that port from access control.
An excluded port will be open and does not require
authentication. This is suited for uplink ports, trunk
ports and for connecting servers. The default for
ports is unchecked, thus enabling port access control/authentication. Check-boxes can be shown as
disabled, like port 1 and 2 in the above picture. This
means that the current VLAN does not have this port
as a member and is therefore not relevant for exclusion. See section 15.3.1 for managing the relations
between ports and VLANs.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
15.3.6
Port-based network access control statistics
Menu path: Status ⇒ Port Access
Here you can see an overview over port access status on a per-port basis. The
802.1X column shows if IEEE 802.1X is enabled for a port or not. The MAC auth
column shows if MAC based authentication is enabled.
You can also see the current number of authenticated hosts. This value is only
showing hosts that have authenticated recently. There may be more hosts on
the network that can be authenticated via MAC based authentication but are
inactive on the network for the moment. See section 15.2.2 for information about
inactivity and MAC based authentication.
A detailed view of the authenticated hosts is shown if you click on the magnifier
icon for a port. This view shows all authenticated host by their MAC address. This
list shows hosts that are authenticated with both IEEE 802.1X and MAC based
authenticated together.
© 2017 Westermo Teleindustri AB
369
Westermo OS Management Guide
Version 4.21.1-0
15.4
Managing VLAN settings via the CLI
Command
MAC Forwarding Database Configuration
fdb
[no] aging-timeout <0|1-3825>
[no] mac <MACADDR> port <PORTLIST>
General VLAN Configuration
[no] vlans
[no] dynamic <adaptive>
Per VLAN Configuration
[no] vlan <VID>
[no] enable
name <VLANNAME>
[no] untagged <PORTLIST>
[no] tagged <PORTLIST>
[no] forbid <PORTLIST>
[no] priority <0-7>
[no] igmp
channel <CHANNELID>
[no] dot1x-auth <ID>
[no] mac-auth <ID>
[no] except-auth <PORTLIST>
Default
Section
300
Section 15.4.1
Section 15.4.2
Section 15.4.3
Disabled
Enabled
vlan<VID>
Disabled
Enabled
0
Disabled
Disabled
Disabled
Section 15.4.4
Section 15.4.5
Section
Section
Section
Section
Section
Section
Section
Section
Section
Section
Section
Section
15.4.6
15.4.7
15.4.8
15.4.9
15.4.10
15.4.11
15.4.12
15.4.13
15.4.14
15.4.15
15.4.16
15.4.17
Show VLAN Status and MAC Forwarding Database Status
show vlans
show fdb
Section 15.4.18
Section 15.4.19
Show Port-based Network Access Control Status
show dot1x-auth
show mac-auth
Section 15.4.20
Section 15.4.21
370
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
15.4.1
Managing MAC Forwarding Database Settings
Syntax fdb
Context Global Configuration context
Usage Use the ”fdb” command to enter the MAC Forwarding Database context
(fdb).
Use ”show fdb” to show current FDB settings (list of configured MAC address filters, and the configured aging timeout). Also available as ”show”
command within the MAC Forwarding Database.
Default values Not applicable.
15.4.2
Configure MAC Address Aging Timeout
Syntax [no] aging-timeout <0|1-3825>
Context MAC Forwarding Database context (fdb)
Usage Set the aging timeout (in seconds) for unicast MAC addresses learnt dynamically. The configured aging timeout will only be an approximation of
the actual aging timeout. The value is first rounded upwards in steps of
15 seconds. The MAC entries will be purged from the forwarding database
within 1/7th of the resulting aging timeout.
Use ”no aging-timeout” or ”aging-timeout 0” to disable aging entirely.
Use ”show aging-timeout” to view the current setting.
Default values 300 (seconds)
15.4.3
Configure Static MAC Filter Entries
Syntax [no] mac <MACADDRESS> port <[PORTS] [ALL] [CPU] | [NONE]>
Context MAC Forwarding Database context (fdb)
Usage Add or delete a static MAC address filter. The ”MACADDRESS” is written as
a colon separated hexadecimal value, e.g., ”01:23:45:56:89:AB”.
The ”PORTLIST” states the port(s) where packets with the given (destination) MAC address are to be forwarded. As of WeOS v4.21.1, the static MAC
© 2017 Westermo Teleindustri AB
371
Westermo OS Management Guide
Version 4.21.1-0
filters are only intended to be used for multicast MAC addresses (not unicast
MAC or the broadcast MAC addresses).
The ”PORTLIST” can include both visual ports (e.g., ”eth 2/1-2/4” on a
slotted WeOS unit) as well as the internal CPU port(s):
ˆ PORT(S): Port, set of or range of ports, e.g. eth 1,3-5
ˆ ALL: All visible ports, excluding internal CPU port(s)
ˆ NONE: No ports, filter this MAC address
ˆ CPU: The internal CPU port(s)
Use ”no MAC <MACADDRESS>” to remove a specific static MAC filter, or ”no
MAC” to remove all static MAC filters.
Use ”show mac” to list configured MAC address entries.
Default values (The factory default configuration includes a set of static MAC
filters.)
15.4.4
Managing general VLAN settings
Syntax [no] vlans
Context Global Configuration context
Usage Enter the General VLAN Configuration context (vlans). The General VLAN
Configuration context can be used to configure VLAN settings applicable to
all VLANs.
Use ”no vlans” to remove all VLANs except the switch default VLAN (VLAN 1).
All ports will be configured untagged on VLAN 1.
Use ”show vlans” to list all configured VLANs and general VLAN settings.
Default values Not applicable.
15.4.5
Enable dynamic VLAN
Syntax [no] dynamic <adaptive>
Context General VLAN Configuration context (vlans)
372
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Usage Use the ”dynamic adaptive” command to enable WeOS Adaptive Dynamic Trunking (AVT) on the switch. For more information on AVT in section 15.1.7.
Future versions of WeOS may include support for dynamic VLAN via GVRP in
addition to AVT, but currently only AVT is supported.
Use ”no dynamic” to disable dynamic VLAN support.
Use ”show dynamic” to see the dynamic VLAN setting.
Default values Disabled
15.4.6
Managing individual VLANs
Syntax [no] vlan <VID>
Context Global Configuration context
Usage Enter VLAN Configuration context of the given VID. If this is a new VLAN,
the VLAN will be created first upon leaving the VLAN context with end or
leave.
Use ”no vlan <VID>” to remove an existing VLAN. The default VLAN (VLAN 1)
cannot be removed. Removal of a VLAN may imply that some ports will no
longer be associated with any VLAN - such ports will be configured to the
default VLAN (VLAN 1) untagged.
Use ”show vlan” (or ”show vlans”) to list all configured VLANs and general
VLAN settings. Use ”show vlan VID” to list detailed configuration information for a specific VLAN (also available as ”show” command within the VLAN
Configuration context of the given VID.
Default values Not applicable.
© 2017 Westermo Teleindustri AB
373
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/config/#>
VLAN ID
:
Status
:
Name
:
Channel
:
Priority
:
Untagged
:
Tagged
:
Forbid
:
IGMP
:
Learning
:
802.1Q VLAN
:
802.1X Auth
:
MAC Auth
:
Except Port Auth :
example:/config/#>
15.4.7
show vlan 1
1
Enabled
vlan1
0
Disabled
U:eth 1-4
T:
F:
Enabled
Enabled
Enabled
Disabled
Disabled
Enable/disable a VLAN
Syntax [no] enable
Context VLAN Configuration context
Usage Enable or disable a VLAN. A disabled VLAN is similar to a deleted VLAN,
except that its configuration is stored, and will be activated when the VLAN
is enabled. That is, when a VLAN is disabled, its ports may be moved onto
the default VLAN (unless they are associated with another VLAN), and any
network interface associated with the VLAN will be disabled.
Use ”show enable” to view the current configuration.
Default values enable
15.4.8
VLAN name
Syntax name <ID>
Context VLAN Configuration context
Usage Specify VLAN name, i.e., VLAN description. Max 15 characters, only
alpha-numerical characters ([a-z,A-Z,0-9]) allowed.
Use ”show name” to view the VLAN name setting.
374
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Default values If no VLAN ”name” command is given, the VLAN name defaults
to vlanVID, e.g., vlan100 for VID 100.
15.4.9
Manage untagged ports
Syntax [no] untagged <PORT|PORTLIST>
Context VLAN Configuration context
Usage Associate port(s) with this VLAN VID in untagged mode. Only a single
VLAN VID can be associated untagged with each port. Ports associated with
a VLAN VID untagged will have that VID as default VID - this will have precedence over any (fall-back) default VID configuration set in port context.
Use ”no untagged <PORTLIST>” to remove untagged ports from a VLAN. If
removal of an untagged port implies that the port is no longer associated
with any VLAN, that port will be configured to VLAN 1 untagged.
Use ”show untagged” to view ports associated untagged with this VLAN.
Default values Factory default lets all ports be associated with the default VLAN
(VLAN 1) untagged. For new VLANs, ports must explicitly be added.
Error messages
ˆ A notification message is given in case the addition of port
as untagged on one VLAN implies that the same port will be removed
as untagged on another VLAN.
ˆ A notification message is given in case the addition of port as untagged
on one VLAN implies that the same port will be removed as tagged on
the same VLAN (a port cannot be associated both tagged and untagged
with the same VLAN).
A ”PORTLIST” is a comma separated list of port ranges without intermediate
spaces, e.g., ”1/1-1/3,2/3”.
15.4.10
Manage tagged ports
Syntax [no] tagged <PORT|PORTLIST>
Context VLAN Configuration context
Usage Associate port(s) with this VLAN VID in tagged mode.
© 2017 Westermo Teleindustri AB
375
Westermo OS Management Guide
Version 4.21.1-0
Use ”no tagged <PORTLIST>” to remove tagged ports from a VLAN. If removal of a tagged port implies that the port is no longer associated with any
VLAN, that port will be configured to VLAN 1 untagged.
Use ”show tagged” to view ports associated tagged with this VLAN.
Default values Not applicable.
Error messages A notification message is given in case the addition of port as
tagged on one VLAN implies that the same port will be removed as untagged
on the same VLAN (a port cannot be associated both tagged and untagged
with the same VLAN).
A ”PORTLIST” is a comma separated list of port ranges without intermediate
spaces, e.g., ”1/1-1/3,2/3”.
15.4.11
Manage forbidden ports
Syntax [no] forbid <PORT|PORTLIST>
Context VLAN Configuration context
Usage Prohibit that ports are dynamically added (AVT) to this VLAN ID, see also
sections 15.1.7 and 15.4.5.
Use ”no forbid <PORTLIST>” to remove ports from the list of ports forbidden to be associated with this VLAN.
Use ”show forbidden” to view ports associated forbidden with this VLAN.
Default values Not applicable.
A ”PORTLIST” is a comma separated list of port ranges without intermediate
spaces, e.g., ”1/1-1/3,2/3”.
15.4.12
VLAN priority setting
Syntax [no] priority <0-7>
Context VLAN Configuration context.
Usage Set the (IEEE 802.1p) priority associated with this VLAN. Incoming packets associated with this VLAN will receive this priority.
376
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
”no priority” will disable VLAN priority for this VLAN. Priority for packets
associated with this VLAN will then be based on port priority settings.
Use ”show priority” to view the priority setting for this VLAN.
Default values Disabled (”no priority”).
15.4.13
VLAN IGMP Snooping
Syntax [no] igmp
Context VLAN Configuration context.
Usage Enable, or disable IGMP Snooping for this VLAN.
Use ”show igmp” to view the IGMP snooping setting for this VLAN.
Default values IGMP snooping enabled.
15.4.14
CPU channel mapping
Syntax channel <CHANNELID>
Context VLAN Configuration context.
Usage Specify CPU channel to use for this VLAN. The channel identifier can take
values in the range <0-CHANNELIDMAX>. The purpose of this command is
to improve routing performance by mapping VLANs to different CPU channels, see section 15.1.6.
Hint
Use the ”show system-information” command (see section 7.3.2) to
find out the number of channels.
ˆ Look for the line ”Channel interfaces” in the information of the CPU
card to see the number of channels.
ˆ CHANNELIDMAX equals ”number of channels”-1.
Use ”show channel” to view the CPU channel setting for this VLAN.
Default values 0 (zero), i.e., by default all VLANs will use channel 0.
© 2017 Westermo Teleindustri AB
377
Westermo OS Management Guide
Version 4.21.1-0
15.4.15
IEEE 802.1X authentication
Syntax [no] dot1x-auth <ID>
Context VLAN Configuration context.
Usage Specify the IEEE 802.1X configuration to be used for this VLAN. Setting this enables port-based network access control for all ports untagged
in this VLAN, except for the ports defined with ”except-auth” (see section 15.4.17). The ID value references the 802.1X configuration. This configuration is managed in the AAA subsystem, see chapter 9. Use ”no dot1x-auth”
to disable IEEE 802.1X authentication for this VLAN.
Use ”show dot1x-auth” to view the IEEE 802.1X authentication setting for
this VLAN.
Default values Disabled, i.e. IEEE 802.1X is not used.
15.4.16
MAC based authentication
Syntax [no] mac-auth <ID>
Context VLAN Configuration context.
Usage Specify the MAC authentication configuration to be used for this VLAN.
Setting this enables port-based network access control for all ports untagged
in this VLAN, except for the ports defined with ”except-auth” (see section 15.4.17). The ID value references the MAC authentication configuration. This configuration is managed in the AAA subsystem, see chapter 9.
Use ”no mac-auth” to disable MAC based authentication for this VLAN.
Use ”show mac-auth” to view the MAC based authentication setting for this
VLAN.
Default values Disabled, i.e. MAC based authentication is not used.
15.4.17
Except ports from authentication
Syntax [no] except-auth <PORT|PORTLIST>
Context VLAN Configuration context.
378
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Usage Disables port-based network access controls for specific ports. This is
used together with ”dot1x-auth” and ”mac-auth” to exclude specific ports
from needing authentication. This is suitable for uplinks, trunks and ports
with servers connected. Use ”no except-auth” to remove all port exceptions, thus enabling access control on all untagged ports in this VLAN.
Use ”show mac-auth” to view ports configured to be excluded from portbased network access control for this VLAN.
Default values Disabled, no ports excluded.
15.4.18
Show VLAN status (all VLANs)
Syntax show vlans
Context Admin Exec context
Usage Show VLAN status information for all VLANs.
Default values Not applicable.
15.4.19
Show Current MAC Forwarding Database
Syntax show fdb
Context Admin Exec context
Usage Show the current state of the MAC forwarding database. This includes
the list of MAC addresses known to the switch, and the port(s) to forward
packets to each MAC address. The ageing timeout for automatically learnt
unicast MAC addresses is also shown.
Default values Not applicable.
© 2017 Westermo Teleindustri AB
379
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/#> show fdb
MAC
VLAN State Portvec
Port(s)
===============================================================================
00:07:7c:81:de:1a
ANY
0x0f 0x0
CPU
00:07:7c:81:de:1d
ANY
0x01 0x0
CPU
00:0d:88:cd:3a:9c
ANY
0x01 0x1
ETH 1/1
01:00:5e:00:00:01
ANY
0x07 0x3fff
ALL
01:00:5e:00:00:02
ANY
0x07 0x3fff
ALL
01:00:5e:00:00:04
ANY
0x07 0x3fff
ALL
01:00:5e:00:00:05
ANY
0x07 0x3fff
ALL
01:00:5e:00:00:06
ANY
0x07 0x3fff
ALL
01:00:5e:00:00:09
ANY
0x07 0x3fff
ALL
01:00:5e:00:00:0a
ANY
0x07 0x3fff
ALL
01:00:5e:00:00:0d
ANY
0x07 0x3fff
ALL
01:00:5e:00:00:0e
ANY
0x07 0x3fff
ALL
01:00:5e:00:00:12
ANY
0x07 0x3fff
ALL
01:00:5e:00:00:18
ANY
0x07 0x3fff
ALL
01:00:5e:00:00:66
ANY
0x07 0x3fff
ALL
01:00:5e:00:00:6b
ANY
0x07 0x3fff
ALL
01:00:5e:00:00:fb
ANY
0x07 0x3fff
ALL
01:80:c2:00:00:0e
ANY
0x07 0x3f
ETH 1/1-ETH 2/4
FDB Aging time: 300 sec.
example:/#>
15.4.20
Show IEEE 802.1X authentication status
Syntax show dot1x-auth
Context Admin Exec context
Usage Show hosts that are currently authenticated with IEEE 802.1X.
Default values Not applicable.
15.4.21
Show MAC based authentication status
Syntax show mac-auth
Context Admin Exec context
Usage Show hosts that are currently authenticated with MAC based access control.
380
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Note
There may be hosts on the network that matches the MAC authentication filters, but are inactive for the moment. Inactive hosts are flushed
out of this list and will be re-authenticated again on resumed activity.
See section 15.2.2 for details.
Default values Not applicable.
© 2017 Westermo Teleindustri AB
381
Westermo OS Management Guide
Version 4.21.1-0
Chapter 16
FRNT
The Fast Reconfiguration of Network Topology (FRNT) protocol handles fast reconfiguration in switched ring topologies. When rapid convergence in case of link or
switch failure is required, FRNT becomes the protocol of choice when it comes to
layer-2 resilience and robustness.
In addition to FRNT, WeOS supports the standard RSTP protocol. Management of
RSTP is described in chapter 18.
16.1
Overview of the FRNT protocol and its features
The table below summarises FRNT features available via the Web and CLI interfaces. A general description of the FRNT protocol and its features is presented in
sections 16.1.1 and 16.2. If you are only interested in knowing how to manage
the FRNT features via the Web or CLI, please visit sections 16.3 or 16.4 directly.
Feature
Enable FRNT
Set FRNT mode (focal-point
or member switch)
Set FRNT ring ports
View FRNT Status
FRNT bus topology
382
Web
X
X
CLI
X
X
X
X
X
X
X
X
General Description
Section 16.1.1
-”-”-”-”-
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
16.1.1
FRNT introduction
The FRNT protocol handles fast reconfiguration in switched ring topologies. One
of the switches has the role of FRNT focal point while the other switches are
referred to as FRNT members. When the switches are connected in a ring, it is
the responsibility of the focal point to break the loop by putting one of its ports
(port ”M”) in blocking mode, see fig. 16.1.
Note
In an FRNT ring, only one of the switches can be configured as focal point.
The other switches should be configured as member switches (i.e., non”focal-point”).
Focal
Point
Member
M
N
M
N
Member
Member
M
M
N
N
Figure 16.1: FRNT network operating in ring mode. Port ”M” on the Focal Point is
in BLOCKING state.
Once a link failure is detected somewhere along the ring, the focal point will
put its blocked port (port ”M”) in forwarding mode to establish full connectivity
between the switches (see fig. 16.2). FRNT is event based: switches detecting
a link down event will immediately send a link down FRNT message towards the
focal point. Intermediate switches will forward the FRNT messages with highest
priority, and the focal point will open its BLOCKED port (port ”M”) upon receiving
the link down message.
Focal point opens
redundant path
Member
M
N
Focal
Point
M
N
Link
break
Member
Member
M
M
N
N
Figure 16.2: FRNT network operating in bus mode due to broken link.
Similarly, when a broken link comes back up again and the ring is fully connected,
© 2017 Westermo Teleindustri AB
383
Westermo OS Management Guide
Version 4.21.1-0
the focal point will react and put its port ”M” back to blocking state.
16.1.2
Guidelines when selecting FRNT ports
When enabling FRNT on a switch, you need to select two ports to use as FRNT
ports – FRNT port ”M” and FRNT port ”N”1 . Below are some recommendations
and rules when selecting and configuring the FRNT ports.
ˆ Fixed speed, full duplex: When using Ethernet ports as FRNT ports, fixed
speed (and full duplex) is recommended over auto-negotiation of speed and
duplex mode on the FRNT ports. Use 100 Mbit/s speed rather than 10 Mbit/s
speed for best performance.
ˆ Some fixed Gbit copper ports has bad failover performance: For fixed Gbit
copper ports, the failover performance may be significantly higher. The
reason behind this relates to restrictions in the Gbit Ethernet standard with
respect to link down behaviour, which in turn affect the link-down detection
performance.
For some switchcores, work-arounds exist to achieve fast link-down detection performance for Gbit copper ports. For example, fixed Gbit copper
ports of MV88E6352 switchcore have good FRNT performance, while fixed
Gbit copper ports of MV88E6185 have bad FRNT performance. See Detailed System Overview page in the Web (section 4.4.2) or use the ”show
system-information” in the CLI (section 7.3.2) to find information about
what switchcore(s) is used in your product.
ˆ Avoid using copper SFPs as FRNT ports: When using Ethernet ports as FRNT
ports, choose fixed Ethernet ports or fiber SFPs. Copper SFPs may be used as
FRNT ports, but will generally imply non-negligible degradation of fail-over
performance.
ˆ SHDSL ports as FRNT ports: It is possible to use SHDSL ports as FRNT ports,
but failover performance is degraded as compared to (fixed) Ethernet ports.
FRNT will not work correctly on SHDSL links with speed below 64 kbit/s.
1 In
384
earlier WeOS versions, port ”M” and ”N” have been denoted port ”1” and ”2” respectively.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
16.1.3
VLANs used by FRNT
FRNT uses VLAN IDs 4020-4022 and 4032-4033 for its signalling. Thus, when
FRNT is enabled on a switch, these VLANs are implicitly reserved and cannot be
configured by the user.
Warning
Note on using intermediate active equipment For FRNT to operate properly, there should not be any ”non-FRNT- enabled” switches (or other active equipment) in the FRNT ring. However, if two FRNT nodes are interconnected via a non-FRNT switch for testing purposes, that intermediate
switch must be configured to let VLANs 4020-4022 and 4032-4033 through.
16.1.4
FRNT Bus
Member
M
Member
Member
Member
M
M
M
N
FRNT
Bus Edge
N
N
Focal
Point
M
FRNT
Bus Edge
Figure 16.3: Configuring FRNT in bus topology. Edge nodes have only one FRNT
port. One of the bus edges must be configured as focal point.
It is possible to use the FRNT protocol also in a bus topology. When setting up
an FRNT bus, you will lose the redundancy properties as compared to running a
regular FRNT ring. The major reason for setting up an FRNT bus is to achieve a
”horseshoe” topology, by combining FRNT bus with Ring Coupling (chapter 17)
and hook the horseshoe into an FRNT super-ring. The redundancy is then provided by the Ring Coupling uplinks.
Note
The main use case for configuring an FRNT bus is to achieve a horseshoe
topology, by using the FRNT bus together with Ring Coupling. See section 17.1.1.3 for more information on the horseshoe use case.
In an FRNT bus, two of the FRNT nodes will be configured as FRNT bus edge
nodes, i.e., these nodes will only be configured with one FRNT port each.
© 2017 Westermo Teleindustri AB
385
Westermo OS Management Guide
Version 4.21.1-0
ˆ One of the bus edge nodes must be configured as FRNT focal point, with
only one FRNT port (port ”M”).
ˆ The other bus edge node is configured as FRNT member switch, with only
one FRNT port (port ”M”).
ˆ Intermediate FRNT nodes are configured as regular FRNT members with two
FRNT ports.
When running FRNT as a bus, the focal point (located in one edge) will detect if
the bus is intact or if it is broken. Thus, the FRNT status on the focal point will
indicate the status of the configured bus, and you can determine the status of
the FRNT bus (OK or Broken) by inspecting the FRNT status on the focal point. As
opposed to when running FRNT as a ring, a focal point configured as bus edge
will not put its FRNT port ”M” in blocking mode when the topology is intact.
16.2
FRNT and RSTP coexistence
With WeOS it is possible to run FRNT and RSTP on the same switch, be it with
some topology restrictions. Fig. 16.4 shows an example of such a configuration,
where two of the switches in the FRNT ring (thick lines) are running RSTP on the
”non-FRNT” ports.
FRNT
Switch
FRNT
Switch
Loop
handled
by FRNT
FRNT/
RSTP
Switch
RSTP
Switch
FRNT/
RSTP
Switch
Loop
handled
by RSTP
RSTP
Switch
RSTP
Switch
Loop
handled
by RSTP
RSTP
Switch
RSTP
Switch
RSTP
Switch
Figure 16.4: Example of coexistence of FRNT and RSTP.
As both RSTP and FRNT want to control a port’s state (FORWARDING/BLOCKING),
only one of the protocols may be activated on each port to avoid protocol con-
386
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
flicts. Therefore, if both FRNT and RSTP are configured to operate on a certain
port, FRNT will have precedence to control the port’s state.
Warning
FRNT and RSTP are each able to handle loops within their respective domains, however, if a physical loop is created including some links controlled
by RSTP and others by FRNT, a broadcast storm is likely to occur, since neither RSTP nor FRNT is able to discover the loop, see fig. 16.5. Thus, if RSTP
and FRNT is mixed in the same layer-2 network, the operator must ensure
that loops across RSTP and FRNT links never occur.
FRNT
Switch
FRNT
Switch
Loop
handled
by FRNT
FRNT/
RSTP
Switch
RSTP
Switch
Loop
Causing
Storm!
Loop
handled
by RSTP
RSTP
Switch
FRNT/
RSTP
Switch
RSTP
Switch
RSTP
Switch
Loop
handled
by RSTP
RSTP
Switch
RSTP
Switch
Figure 16.5: Example of loop spanning FRNT and RSTP links - a broadcast storm
is likely to occur.
© 2017 Westermo Teleindustri AB
387
Westermo OS Management Guide
Version 4.21.1-0
16.3
Managing FRNT settings via the web interface
16.3.1
Managing FRNT settings
Menu path: Configuration ⇒ L2 Redundancy ⇒ FRNT
On the FRNT configuration page you will be presented to the current settings for
FRNT on your switch, see below.
Ring ID
Focal Point
Port M/Port N
Couplings
A unique identifier for the FRNT-ring. Currently only one
ring is available.
The focal point is the unit in the ring which is responsible
for making decisions on topology change. A green checkmark indicates this unit will take the role as focal point
in the FRNT ring. A dash indicates the unit will act as a
member unit.
FRNT requires two ports to be assigned FRNT-ports.
These are connected to peer units participating in the
FRNT ring. The two ports connected to other units in the
FRNT ring.
Note: Ports with copper SFPs should not be used as FRNT
ports, due to slow link down indication on copper SFPs.
See section 16.1.2 for further guidelines on FRNT port selection.
Lists the currently configured FRNT Ring-Couplings associated with this FRNT-ring, and the coupling uplink ports.
Edit
Click this icon to edit an FRNT instance.
Delete
Click this icon to remove an FRNT instance.
If no FRNT instance is configured you may create one by clicking the New button.
388
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
When editing a new or existing instance the page below is displayed.
The FRNT settings are described in the table above. The lower part contains a
section, Couplings, where the FRNT Ring-Couplings, associated with this FRNT
instance, is listed. This section will appear after clicking the Apply when a new
FRNT instance is created.
To create a new Coupling instance, click the New Coupling button (visible until
MAX_RING_COUPLING_INSTANCES (section 17.4) has been reached). New and
existing Ring-Couplings are edited on the page below:
© 2017 Westermo Teleindustri AB
389
Westermo OS Management Guide
Version 4.21.1-0
Enabled
Hello Time
Uplinks
Port
Priority
Adjustment
Echo Interval
Path Cost
390
A green checkbox if the coupling instance is enabled, a
minus sign if not. On edit page, check/uncheck box to
enable/disable coupling instance.
The interval between two hello messages in milliseconds.
The uplink port.
The uplinks priority. Used for calculating active uplink.
Priority adjustment delta for this uplink. Makes the uplink
sticky by adjusting the effective priority with this value
when uplink becomes active.
The interval (ms) between packets sent to the uplink
neighbour to verify uplink connectivity.
The uplinks path cost. Used for calculating active uplink.
Auto (check-box checked) indicates path-cost is automatically calculated (based on link speed).
Delete
Click this icon to remove a coupling instance.
Add
Click this icon to add a new coupling instance.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
16.3.2
FRNT Status and Statistics
Menu path: Status ⇒ L2 Redundancy ⇒ FRNT
On this page FRNT status and statistics are presented.
Figure 16.6: FRNT status and statistics in web
FRNT Status and
Ring
Enabled
Mode
Status
Port M
Port N
Topology
Change Count
Time Since
Last Change
Statistics
Instance ID for the FRNT ring.
Indication if the ring is enabled or not.
Focal point or member.
Ring status, OK or BROKEN.
Status of port operating1 as FRNT port M.
Status of port operating1 as FRNT port N.
Number of FRNT topology changes.
Time since last FRNT topology change.
Continued on next page
© 2017 Westermo Teleindustri AB
391
Westermo OS Management Guide
Version 4.21.1-0
Continued from previous page
Ring Coupling
Local/Global
Port
Active
MAC
Effective Priority
Path-Cost
Speed/Duplex
Hello Time
Synchronized
Link Changes
Auto Refresh
Refresh
Local - uplinks located on this switch.
Global - uplinks reported from other switches in the FRNT
ring.
The uplink port name, if any available on the distributing
unit. Otherwise an information message stating that no
uplinks are available.
A green check-mark indicates this is the active uplink for
the ring coupling instance.
The MAC address of the unit distributing this piece of uplink information.
The actual priority value used in uplink selection. When
configuring an adjustment delta this may differ from the
configured priority for an active uplink. Used to minimise
uplink changes when an active uplink goes down and up
again.
The current path-cost. If auto configuration selected, this
value is calculated based on port speed.
Speed duplex on the uplink port. Only applicable for local
uplinks.
The configured and effective (negotiated) hello-time on
each unit.
A green check-box indicates this uplink has been synchronised with its neighbour at the remote end of the uplink.
Only applicable for local uplinks.
Number of link changes. Only applicable for local uplinks.
Click on a value to make the page reload with updated
statistics automatically every 5, 15, 30 or 60 seconds.
Click Off to turn off auto refresh.
Click on this button to reload with updated statistics.
1 If
the port referred to as FRNT port ”M” and FRNT port ”N” in the FRNT statistics page (operational FRNT ”M” and ”N”) does not match the administratively configured FRNT ”M” and ”N” ports
(see the FRNT configuration page in section 16.3.1), the ports are logically swapped/aligned with
the ”M” and ”N” ports of the focal-point.
392
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
16.4
Managing FRNT settings via the CLI
Command
Configure FRNT settings
[no] frnt [ID]
[no] focal-point
ring-ports <PORT-M [, PORT-N]>
Default
Section
focal-point
Section 16.4.1
Section 16.4.2
Section 16.4.3
Show FRNT status
show rings
show ipconfig
16.4.1
Section 16.4.4
Section 7.3.35
Managing FRNT
Syntax [no] frnt [<ID>]
Context Global Configuration context
Usage Enter FRNT context of the given FRNT instance ID. Currently only a single
FRNT instance is supported, thus the value of the FRNT ID is ignored.
The FRNT instance is only activated upon the selection of valid FRNT ring
ports, see section 16.4.3.
Use ”no frnt [ID]” to remove an existing FRNT instance.
Use ”show frnt” to list configured FRNT settings (also available as ”show”
command within the FRNT Configuration context).
Default values Default ID is 1
16.4.2
FRNT focal point and member switch
Syntax [no] focal-point
Context FRNT Configuration context
Usage Configure device to act as FRNT focal point for this FRNT instance. Use
”focal-point” to configure the device to act as an FRNT focal-point, and
”[no] focal-point” to configure the device as an FRNT member switch.
© 2017 Westermo Teleindustri AB
393
Westermo OS Management Guide
Version 4.21.1-0
Use ”show focal-point” to show whether the unit is configured as focalpoint or member switch
Default values focal-point
16.4.3
FRNT Ring Ports
Syntax ring-ports <PORT-M [,PORT-N]>
Context FRNT Configuration context
Usage Set the physical ports (Ethernet ports or SHDSL ports) to use as FRNT
ports ”M” and ”N”.
For each FRNT instance, there is normally two FRNT ports named Port ”M”
and Port ”N”, configured by use the ”ring-ports <PORT-M ,PORT-N>” command.
On a member switch Port ”M” and ”N” have similar roles, however, on a
focal point their roles differ - when the ring is fully connected the focal point
will put its Port ”M” in BLOCKING state.
Note
For restrictions on how to select FRNT ports, see section 16.1.2.
In the special case when you wish to configure a node as FRNT Bus Edge,
you only configure port ”M” (”ring-ports <PORT-M>”).
Use ”show ring-ports” to show configured FRNT ring port(s).
Default values Not applicable
16.4.4
Show FRNT ring status
Syntax show rings
Context Admin Exec context.
Usage Show status of configured FRNT rings. This will provide information
ˆ whether the ring is up (ring mode) or if the ring is broken (bus mode).
Note: A focal point switch will detect ring failures located anywhere in
394
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
the ring, while a member switch can only detect local failures (local
FRNT port is down, or if a neighbour is down).
ˆ If the FRNT ports on this switch are connected in-line with the M/N ports
of the focal-point, or if they are logically swapped (i.e., if the FRNT ports’
administrative M/N state equals the operational M/N state, or if ports are
swapped).
ˆ The status of the local FRNT ports (UP/DOWN, FORWARDING/BLOCKING).
Default values Not applicable.
© 2017 Westermo Teleindustri AB
395
Westermo OS Management Guide
Version 4.21.1-0
Chapter 17
FRNT Ring Coupling and
Multi-Link Dual Homing
This chapter describes WeOS FRNT Ring Coupling and Multi-Link Dual Homing,
two similar layer-2 (switching) fail-over functions.
FRNT Ring Coupling enables bridging of two or more FRNT rings via multiple
layer-2 uplinks. Only one uplink is active at a time, while others are hot stand-by
backups, providing redundancy and loop-free connectivity. It is also possible to
use FRNT Ring Coupling to bridge an ”FRNT bus” to an FRNT ring, thereby forming
a horseshoe topology.
Multi-Link Dual-Homing (or simply ”Dual-Homing”) lets you connect a WeOS switch
to a layer-2 topology via multiple uplinks. Optimal fail-over performance is achieved
when connecting the dual-homing switch uplinks to a single FRNT ring, or to two
adjacent FRNT rings (connected with Ring Coupling), but Dual-Homing can also
be used in other layer-2 topologies.
Both FRNT Ring Coupling and Multi-Link Dual-Homing provide fine-grained control
of which uplink is to be preferred as active. By default, the link with highest
speed/duplex mode is elected. To avoid shifting between active uplinks when a
new uplink becomes available, a feature referred to as sticky uplink is provided.
Enabling sticky uplink gives ”zero” fail-over time on link-up and mitigates possible
problems with flapping links.
Section 17.1 presents further information on FRNT Ring Coupling and the MultiLink Dual-Homing functionality. Web and CLI support for these features are covered in sections 17.2 and 17.3 respectively.
396
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
17.1
Overview
Feature
FRNT Ring Coupling
Enable
Ring Hello Interval
Define Uplink(s)
Uplink Path-Cost
Uplink Priority
Uplink Echo Interval
Ring Coupling Status
Web
X
X
X
X
X
X
X
X
CLI
X
X
X
X
X
X
X
X
Multi-Link Dual-Homing
Enable
Synchronized
Multiple Instances
Define Uplink(s)
Uplink Path-Cost
Uplink Priority
Uplink Echo Interval
Dual-Homing Status
X
X
X
X
X
X
X
X
X
X
X
17.1.1
X
X
X
X
X
General Description
Section 17.1.1
Section 17.1.1.2
Sections 17.1.1 and 17.1.3
Section 17.1.3
-”Section 17.1.3
Section 17.1.2
Section 17.1.2.1
Section 17.1.2.2
Sections 17.1.2 and 17.1.3
Section 17.1.3
-”-”-
FRNT Ring Coupling
FRNT Ring Coupling (RiCo) enables redundant bridging between two or more
FRNT rings. Fig. 17.1a shows a simple example where two RiCo nodes in an FRNT
sub-ring are connected with one uplink each to the FRNT super-ring. A super-ring
is a ring without RiCo nodes. It is possible to use more than two RiCo nodes in
the sub-ring, and each RiCo node can have more than one uplink, as shown in
fig. 17.1b. (Up to MAX_RING_COUPLING_UPLINKS (section 17.4) can be created.)
Only one of the uplinks is forwarding data – the active uplink, ”solid” in fig. 17.1,
while the other uplink(s) are hot-standby backups, ”dashed” in fig. 17.1. To prevent traffic to flow over backup uplinks the RiCo nodes put all backup uplinks in
BLOCKING state.
© 2017 Westermo Teleindustri AB
397
Westermo OS Management Guide
Version 4.21.1-0
Super−ring
Super−ring
FRNT
RiCo
RiCo
Sub−ring
FRNT
a)
FRNT
RiCo
RiCo
RiCo
Sub−ring
FRNT
b)
Figure 17.1: Ring Coupling with two FRNT rings: (a) single uplinks, and (b) multiple uplinks per Ring Coupling node.
In the CLI example below the leftmost Ring Coupling node in fig. 17.1a is a WeOS
unit configured as an FRNT member switch (see chapter 16) with ring-ports ’1’
and ’2’, and port ’3’ as uplink.
Example
example:/#> configure
example:/config/#> frnt
Activating FRNT0 with default settings, remember to change the
Invalid settings: No ring ports defined
example:/config/frnt-1/#> ring-ports 1,2
example:/config/frnt-1/#> no focal-point
example:/config/frnt-1/#> coupling
Creating new instance 1
example:/config/frnt-1/coupling-1/#> uplink 3
example:/config/frnt-1/coupling-1/uplink-eth3/#> priority 100
example:/config/frnt-1/coupling-1/uplink-eth3/#> leave
Starting Fast Redundant Network Topology v0 daemon ......... [
Starting Ring bridging/dual-homing daemon .................. [
Configuration activated. Remember "copy run start" to save to
example:/#>
ring ports!
OK ]
OK ]
flash (NVRAM).
Here, the uplink priority was given the value ”100” to make it the preferred active
uplink. The default is 128, for further details see section 17.1.3.
398
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
It is of course possible to connect multiple sub-rings to one super-ring. The uplinks from the sub-rings can be connected to individual nodes in the super-ring
(see fig. 17.2a) or the nodes in the super-ring can be shared, see fig. 17.2b.
Sub−ring
FRNT
RiCo
RiCo
Super−ring
Super−ring
FRNT
FRNT
RiCo
RiCo
RiCo
RiCo
RiCo
RiCo
Sub−ring
Sub−ring
Sub−ring
FRNT
FRNT
FRNT
a)
b)
Figure 17.2: Two sub-rings connecting to (a) individual nodes in the super-ring, or
(b) shared nodes in the super-ring.
The topology can be extended even further by connecting sub-rings to sub-rings
in a tree structure with a super-ring as root. Fig. 17.3 shows two examples, a
ladder topology (a) and a tree topology (b).
© 2017 Westermo Teleindustri AB
399
Westermo OS Management Guide
Version 4.21.1-0
Super−ring
FRNT
RiCo
RiCo
Sub−ring
Sub−ring
FRNT
FRNT
Super−ring
RiCo
FRNT
RiCo
RiCo
RiCo
RiCo
Sub−ring
Sub−ring
FRNT
FRNT
RiCo
Sub−ring
RiCo
FRNT
RiCo
RiCo
RiCo
RiCo
RiCo
RiCo
RiCo
Sub−ring
Sub−ring
Sub−ring
FRNT
FRNT
FRNT
a)
b)
Figure 17.3: Examples of tree topologies: (a) shows a ”ladder” (a tree without
branches), and (b) a more generic tree of FRNT rings.
400
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
17.1.1.1
Ring Coupling and Routing
FRNT Ring Coupling is a function to connect FRNT rings at layer-2 (switching).
This improves capacity and failover performance compared to layer-3 (routing)
mechanisms such as OSPF (chapter 27). Nevertheless, routing techniques have
good scalability characteristics as the network is segmented into different broadcast domains.
Although technically feasible, it is strongly recommend to separate ring coupling
and routing, localising the distinct functions in dedicated WeOS units, i.e., do not
use RiCo nodes also as routers.
Note
In some cases using RiCo nodes as routers makes sense. To ensure correct
operation of the RiCo node, the CPU bandwidth is reduced by default, i.e.,
when ”cpu-bandwidth-limit” is set to ”auto” (section 8.3.6) on a WeOS
unit configured for FRNT Ring Coupling or Multi-link Dual-Homing. This in
turn reduces routing performance.
This automatic reduction of CPU bandwidth can be overridden by changing
the CPU bandwidth limit setting (section 8.3.6).
17.1.1.2
Ring Coupling Hello Interval
RiCo nodes in the same FRNT ring exchange Hello messages to discover each
other as part of the active uplink election process, see also section 17.1.3. These
Hello messages are transmitted every 100 ms by default on the FRNT ring ports.
The hello interval can be fine tuned – a lower value gives faster failover, but may
have an adverse effect on the CPU usage. When the CPU usage increases RiCo
nodes may not be able to send Hello messages and will time out. This can lead
to unpredictable performance and loss of connectivity.
It is recommended that all RiCo nodes within an FRNT ring are configured with
the same Hello interval. If there are RiCo nodes with different Hello interval in an
FRNT ring, the protocol will default to the highest interval announced by any RiCo
node, i.e., a RiCo node’s effective hello interval may differ from its configured
hello interval. E.g., if you wish to transition from using ”hello-time 100” to
”hello-time 80”, all RiCo nodes will use interval 100 ms until all RiCo node’s in
the FRNT ring has been configured with interval 80 ms.
© 2017 Westermo Teleindustri AB
401
Westermo OS Management Guide
Version 4.21.1-0
Note
Configuring RiCo nodes with the same RiCo Hello interval is particularly important when forming a Horseshoe topology (section 17.1.1.3) by connecting
RiCo uplinks via an FRNT bus.
17.1.1.3
Horseshoe topologies by use of Ring Coupling and FRNT Bus
In some case you may wish to extend the FRNT super-ring with a ”horseshoe”
network rather than connecting an FRNT sub-ring to the super-ring. Such a topology is shown in fig. 17.4 by use of Ring Coupling together with an ”FRNT bus”
(section 16.1.4).
Super−ring
FRNT
RiCo
Uplink
Horse shoe
RiCo
FRNT
Bus Edge &
Focal point
RiCo
Uplink
RiCo
FRNT bus
FRNT
Bus Edge
Figure 17.4: Adding ”horseshoe” to FRNT super-ring by use of Ring Coupling and
FRNT bus.
The use of an FRNT bus rather than an FRNT ring in the ”sub-network” can be
motivated when:
ˆ When you do not have the cable needed or spare ports available to connect
the nodes in the ”sub-network” together in a ring.
ˆ If the additional robustness and failover performance, achieved by forming
the ”sub-network” as a ring, is of less importance.
402
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
17.1.2
Multi-Link Dual-Homing
Multi-Link Dual-Homing makes it possible for a WeOS dual-homing node to have
redundant connections to an FRNT ring. Fig. 17.5 shows an example where two
dual-homing nodes are connected to the FRNT ring with two connections each.
Dual
Hom.
FRNT
Dual
Hom.
FRNT
Dual
Hom.
Dual
Hom.
a)
b)
Figure 17.5: Dual-Homing with single FRNT ring (super-ring). (a) and (b) show the
same topology in different ways.
Consider one of the dual-homing nodes in fig. 17.5a, assuming it is a WeOS switch
with Ethernet ports ’1’ and ’2’ as uplinks, and where port ’1’ is to be active by
default. A possible configuration is given below:
Example
example:/#> configure
example:/config/#> dual-homing
Creating new instance 1
example:/config/dual-homing-1/#> uplink 1
example:/config/dual-homing-1/uplink-eth1/#> priority 100
example:/config/dual-homing-1/uplink-eth1/#> end
example:/config/dual-homing-1/#> uplink 2
example:/config/dual-homing-1/uplink-eth2/#> leave
Starting Ring bridging/dual-homing daemon .................. [ OK ]
Configuration activated. Remember "copy run start" to save to flash (NVRAM).
example:/#>
© 2017 Westermo Teleindustri AB
403
Westermo OS Management Guide
Version 4.21.1-0
Note
It is possible to connect uplinks of a WeOS dual-homing switch to any layer-2
topology, but failover performance is optimised for FRNT rings.
When connecting uplinks to LAN topologies other than FRNT the synchronised option in dual-homing must be disabled, see section 17.1.2.1, otherwise the uplinks will not come up.
Section 17.1.2.1 describes the synchronised dual-homing function, illustrates
how you can combine Multi-Link Dual-Homing and FRNT Ring Coupling, supporting topologies where you can connect the dual-homing uplinks to two adjacent
FRNT rings.
Section 17.1.2.2 presents the possibility of using multiple instances of dual-homing.
17.1.2.1
Synchronized Dual-Homing
WeOS Multi-Link Dual-Homing provides a mechanism referred to as synchronised
dual-homing. Synchronised dual-homing has two purposes:
ˆ Integrity of the uplink: A dual-homing switch monitors uplink connectivity
by exchanging specific echo packets with the remote switch. This ensures
that a link break is detected even in cases where intermediate transceivers
do not propagate link down.
ˆ Define preferred uplink when connecting two adjacent FRNT rings: It is possible to connect the uplinks of a dual-homing node to two adjacent FRNT
rings, which in turn are connected by ring coupling. The synchronised dualhoming feature will give preference to the uplink connected to a ring coupling sub-ring, i.e., the ring containing the RiCo nodes. More details later in
this section.
If you wish to connect a dual-homing switch to topologies other than FRNT you
need to disable the synchronised dual-homing feature in the dual-homing node.
An example is given below where ports 1 and 2 are configured as uplinks to nonFRNT nodes.
404
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/#> configure
example:/config/#> dual-homing
Creating new instance 1
example:/config/dual-homing-1/#> uplink 1
example:/config/dual-homing-1/uplink-eth1/#> priority 100
example:/config/dual-homing-1/uplink-eth1/#> end
example:/config/dual-homing-1/#> uplink 2
example:/config/dual-homing-1/uplink-eth2/#> end
example:/config/dual-homing-1/#> no synchronized
example:/config/dual-homing-1/#> leave
Starting Ring Coupling/Dual-Homing daemon .................. [ OK ]
Configuration activated. Remember "copy run start" to save to flash (NVRAM).
example:/#>
Additional remarks if you intend to use dual-homing to connect the uplinks to
other nodes than FRNT nodes.
ˆ Fail-over performance: Fail-over performance is optimised when connecting
the dual-homing node to a single FRNT ring, or to two FRNT rings. (Which
in turn may be connected by FRNT ring coupling). In other topologies the
switches may temporarily have stale MAC entries in their learning caches
for a short period of time (unicast traffic). Furthermore, if IGMP snooping is
used multicast traffic will also be disrupted until switches receive new IGMP
Reports via the new uplink.
ˆ Integrity of the uplink: With synchronised dual-homing disabled, the uplink
status is determined based on its physical status (up/down). If you wish additional control in this case, you could consider running LACP on the uplink,
i.e., you could create a link-aggregate with the uplink as the only member
link. See chapter 19 for further information on LACP and link aggregation.
It is possible to combine the use of Multi-Link Dual-Homing and FRNT Ring Coupling. Fig. 17.6 shows how a dual-homing node can be connected to two adjacent
FRNT rings, and fig. 17.7 illustrates an example with several dual homing nodes.
The synchronised dual-homing feature will give preference to the uplink leading
to the FRNT ring containing RiCo nodes (the ring coupling ”sub-ring”). This means
that the ’left’ dual-homing uplink (port ’1’) in fig. 17.6 will be active as long as a
RiCo node in that ring is reachable, and in turn has an active uplink.
To ensure that the dual-homing node fail-over to the other uplink (the ’right’
uplink (port ’2’) in fig. 17.6) if no RiCo node is reachable via the sub-ring, port
’2’ should be configured with better uplink priority, see also section 17.1.3. A
configuration example is given below.
© 2017 Westermo Teleindustri AB
405
Westermo OS Management Guide
Version 4.21.1-0
Sub−ring
Super−ring
RiCo
FRNT
FRNT
RiCo
1
2
Dual
Hom.
Figure 17.6: Dual-Homing used in an FRNT Ring Coupling Topology.
Sub−ring
Super−ring
Dual
Hom.
Dual
Hom.
FRNT
Dual
Hom.
FRNT
RiCo
RiCo
Figure 17.7: Multiple Dual-Homing nodes in an FRNT Ring Coupling Topology.
Example
example:/#> configure
example:/config/#> dual-homing
Creating new instance 1
example:/config/dual-homing-1/#> uplink 1
example:/config/dual-homing-1/uplink-eth1/#> end
example:/config/dual-homing-1/#> uplink 2
example:/config/dual-homing-1/uplink-eth2/#> priority 100
example:/config/dual-homing-1/uplink-eth2/#> leave
Starting Ring Coupling/Dual-Homing daemon .................. [ OK ]
Configuration activated. Remember "copy run start" to save to flash (NVRAM).
example:/#>
406
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
17.1.2.2
Multiple instances of Dual-Homing
It is possible to create multiple dual-homing instances on a WeOS switch. Each
instance has its own set of uplinks, referred to as an uplink domain – one of the
uplinks in the domain will be active, while the others are backups. A sample
topology is shown in fig. 17.8.
Sub−ring
FRNT
Super−ring
RiCo
FRNT
RiCo
Uplink
Domain A
(Instance 1)
Dual
Hom.
(Instance 2)
Uplink
Domain B
Sub−ring
Super−ring
RiCo
FRNT
FRNT
RiCo
Figure 17.8: Possibility to setup multiple uplink domains for dual-homing.
Warning
The upper and lower LANs in fig. 17.8 must not have additional interconnections. Otherwise a layer-2 loop would be created via the dual-homing
node, unless the dual-homing node has VLAN barriers between uplinks of
the different instances.
© 2017 Westermo Teleindustri AB
407
Westermo OS Management Guide
Version 4.21.1-0
17.1.3
Active uplink election
Both FRNT Ring Coupling and Multi-Link Dual Homing makes use of an uplink election mechanism to determine which of the available uplinks that should become
active and which should be backups. For Dual-Homing, the election is handled
within the dual-homing node itself, but for FRNT ring coupling the election process the RiCo nodes in the FRNT ring negotiate which uplink should be active.
Only uplinks which are up are considered when electing the active uplink. To ensure that an uplink is up, RiCo and Dual-Homing nodes send Uplink Echo packets
which are returned by the FRNT node at the other end of the uplink. Thus, a
RiCo and Dual-Homing node will only consider an uplink to have status up if it
is physically up, and if it receives responses to its Uplink Echo packets (loss of
3 packets packets indicates link down). These Hello messages are transmitted
every 200 ms by default on the uplink ports. The uplink echo interval can be fine
tuned – a lower value gives faster failover, but may have an adverse effect on
the CPU usage. For Dual-Homing, the Uplink Echo message exchange can be disabled, thereby enabling the dual-homing node to be connected other networks
than FRNT, see section 17.1.2.1.
To determine which of the available uplinks that is preferred, an cost vector is
formed for every uplink and compared. The uplink with the lowest cost vector is
elected as active uplink1 . The cost vector consists of the following fields.
ˆ Link speed/duplex Cost: The most significant component of the cost vector depends on the link’s speed and duplex setting. This link speed/duplex
component is calculated as shown in table 17.2, similar to link cost calculation in RSTP (see section 18.1.3). It is also possible to configure the link
speed/duplex cost manually. Default: Auto (see table 17.2)
ˆ Priority: The next component of the cost vector is the uplink priority, which
is used when two or more uplinks have the same link speed/duplex cost.
Then the uplink with the lowest priority value is elected as active uplink.
Default: 128
ˆ Base MAC address: (Only for Ring Coupling) If link speed/duplex cost and
uplink priority are equal for two RiCo nodes, the node with the lowest baseMAC address will win.
1 The exception is dual-homing with synchronised dual-homing enabled. Then uplinks to FRNT
rings with reachable Ring Coupling nodes have precedence over other uplinks, see also section 17.1.2.1.
408
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
ˆ Link/port identity: Finally, if all other components are equal, the port with
the lowest port number is elected as active.
Bandwidth
Full Duplex
Half Duplex
Two Aggregated Links
10 Mbps
100 Mbps
1 Gbps
2,000,000
200,000
20,000
4,000,000
400,000
40,000
1,000,000
100,000
10,000
Table 17.2: Link speed & duplex to link cost component translation table. For
aggregated links (see chapter 19) the link speed/duplex cost is half the cost of
a single link for the given link speed and duplex mode. This is shown in the
right-most column.
To mitigate issues with flapping uplinks, e.g., caused by bad cables, dual-homing
nodes and ring coupling nodes can be configured to use a sticky uplink, as opposed to the deterministic uplink election described above. With sticky uplink enabled, the priority component of an uplink’s cost vector is reduced with a given
value (the adjustment value) once that link is elected as active. That is, with
sticky uplink configured, the effective priority of an uplink can differ from the
configured priority.
Example
Consider three uplinks with same speed and duplex. Link A has ”priority
100”, link B has ”priority 110” with ”adjustment 20”, and link C has
”priority 120” with ”adjustment 40”. All nodes keep information about
each-others announced link cost (100, 110 and 120). If Link A goes
down, link B will take over as it has lower (i.e., better) priority than link
C (110<120), and link B will decrease its effective priority to 90 in its announcements.
If link A comes up again, link B will continue to be active as ”90<100”. The
mechanism works in the same way for dual-homing, even though priority is
never ”announced” to any other node.
17.1.4
Handling Multicast
To provide fast fail-over of multicast traffic, FRNT Ring Coupling and Multi-Link
Dual-Homing uplinks are added to the list of multicast router ports, see section 20.1.1. This is both done at the Ring Coupling nodes and Dual-Homing
© 2017 Westermo Teleindustri AB
409
Westermo OS Management Guide
Version 4.21.1-0
nodes, as well as on switches on the remote side of the uplink2 . This means that
all layer-2 multicast traffic is always sent over the uplinks, even if IGMP snooping
is enabled.
2 An exception is when connecting a Dual-Homing uplink to a non-FRNT switch, the fail-over of
multicast traffic will instead occur on the next reception of an IGMP Report (if IGMP snooping is
enabled). See also section 17.1.2.1.
410
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
17.2
Managing via the Web
17.2.1
Managing FRNT Ring Coupling Settings
FRNT ring couplings are set up in the FRNT context, see section 16.3.1 for further
information.
17.2.2
Managing Dual-Homing Settings
Menu path: Configuration ⇒ L2 Redundancy ⇒ Dual-Homing
Here the list of currently configured Dual-Homing instances is found.
ID
Enabled
Uplinks
Synchronized
A unique identifier for the dual-homing instance.
A green checkbox if the dual-homing instance is enabled,
a minus sign if not.
A list of the uplinks configured for this dual-homing instance.
A green check-box indicates this uplink has been synchronised with its neighbour.
Edit
Click this icon to edit a dual-homing instance.
Delete
Click this icon to remove a dual-homing instance.
Use the New button to create a new Dual-Homing instance.
Up to MAX_DUAL_HOMING_INSTANCES (section 17.4) can be created.
© 2017 Westermo Teleindustri AB
411
Westermo OS Management Guide
Version 4.21.1-0
Enabled
Synchronized
Uplinks
Port
Priority
Adjustment
Echo Interval
Path Cost
412
Check/uncheck box to enable/disable dual-homing instance.
Check/uncheck box to enable/disable Synchronized mode
which requires synchronisation with its neighbour.
The uplink port.
The uplinks priority. Used for calculating active uplink.
Priority adjustment delta for this uplink. Makes the uplink
sticky by adjusting the effective priority with this value
when uplink becomes active.
The interval (ms) between packets sent to the uplink
neighbour to verify uplink connectivity.
The uplinks path cost. Used for calculating active uplink.
Auto (check-box checked) indicates path-cost is automatically calculated (based on link speed).
Delete
Click this icon to remove a dual-homing instance.
Add
Click this icon to add a new dual-homing instance.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
17.2.3
Dual-Homing Status and Statistics
Menu path: Status ⇒ L2 Redundancy ⇒ Dual-Homing
On this page dual-homing status and statistics is presented.
Figure 17.9: Dual-homing status and statistics in web
Port
Active
Effective Priority
Path-Cost
Speed/Duplex
Synchronized
The uplink port name.
A green check-mark indicates this is the active uplink for
the dual-homing instance.
The actual priority value used in uplink selection. When
configuring an adjustment delta this may differ from the
configured priority for an active uplink. Used to minimise
uplink changes when an active uplink goes down and up
again.
The current path-cost. If auto configuration selected, this
value is calculated based on port speed.
Speed duplex on the uplink port.
A green check-box indicates this uplink has been synchronised with its neighbour at the remote end of the uplink.
Only applicable for local uplinks.
Continued on next page
© 2017 Westermo Teleindustri AB
413
Westermo OS Management Guide
Version 4.21.1-0
Preferred
Auto Refresh
Refresh
414
Continued from previous page
A green check-box indicates this uplink is preferred.
Click on a value to make the page reload with updated
statistics automatically every 5, 15, 30 or 60 seconds.
Click Off to turn off auto refresh.
Click on this button to reload with updated statistics.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
17.3
Managing via CLI
Command
Configure Ring Coupling Settings
frnt
[no] coupling [ID]
[no] enable
[no] hello-interval <50..10000>
[no] uplink <PORT>
[no] path-cost <auto|<COST>>
[no] priority <1..65535> [adjust <DELTA>]
[no] echo-time <20..1000>
Configure Multi-Link Dual-Homing Settings
[no] dual-homing [ID]
[no] enable
[no] synchronized
[no] uplink <PORT>
[no] path-cost <auto|<COST>>
[no] priority <1..65535> [adjust <DELTA>]
[no] echo-time <20..1000>
Default
1
Enabled
100 (msec)
Auto
128
200 (msec)
1
Enabled
Enabled
Auto
128
200 (msec)
FRNT Ring Coupling and Multi-Link Dual-Homing Status
show coupling
show dual-homing [ID]
17.3.1
Section
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
16.4.1
17.3.1
17.3.2
17.3.3
17.3.4
17.3.5
17.3.6
17.3.7
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
17.3.8
17.3.9
17.3.10
17.3.11
17.3.12
17.3.13
17.3.14
Sec. 17.3.15
Sec. 17.3.16
Managing FRNT Ring Coupling
Syntax [no] coupling [ID]
Context FRNT Configuration context
Usage Use ”coupling ID” to enter FRNT Ring Coupling Configuration context of
the given Ring Coupling instance ID. Currently only a single Ring Coupling in-
© 2017 Westermo Teleindustri AB
415
Westermo OS Management Guide
Version 4.21.1-0
stance is supported, thus the value of the coupling ID is ignored. ”coupling
ID” creates an FRNT Ring Coupling instance unless it already exists.
Use ”no coupling” to remove the ring coupling instance.
Use ”show coupling” to show configuration information for the ring coupling instance. (Also available as ”show” command within the FRNT Ring
Coupling Configuration context.)
Default values Default ID is 1
17.3.2
Enable/Disable FRNT Ring Coupling
Syntax [no] enable
Context FRNT Ring Coupling Configuration context
Usage Enable or disable an FRNT Ring Coupling instance. Use ”enable” to
enable the coupling instance, and ”no enable” to disable the coupling instance (without losing configuration settings for this instance).
Use ”show enable” to show whether the coupling instance is enabled or
disabled.
Default values Enabled
17.3.3
Set FRNT Ring Coupling Hello Interval
Syntax [no] hello-interval <50..10000>
Context FRNT Ring Coupling Configuration context
Usage Use ”hello-interval VALUE” to set the hello interval (in milliseconds)
to be announced by his ring coupling node.
Note
The effective hello-interval used will be the highest interval announced
by any ring coupling node in the FRNT ring.
”no hello-interval” resets the configured hello interval to the default setting (100 milliseconds).
Use ”show hello-interval” to show the configured hello interval.
416
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Default values 100 (msec)
17.3.4
Managing FRNT Ring Coupling Uplink Ports
Syntax [no] uplink [PORT]
Context FRNT Ring Coupling Configuration context
Usage Use ”uplink PORT” to define the given port as uplink for this ring coupling node, and enter the Ring Coupling Uplink Configuration context for the
port. A port can be an Ethernet port (chapter 10), a DSL port (chapter 12
and chapter 13), or a link aggregate (chapter 19).
Up to MAX_RING_COUPLING_UPLINKS (section 17.4) can be created.
Use ”no uplink PORT” to remove the give port as uplink for this ring coupling node, or use ”no uplink” to remove all uplinks for the node.
Use ”show uplink” to list configuration information for all uplinks, and ”show
uplink PORT” to list uplink configuration settings for the given port (also
available as ”show” command within the Ring Coupling Uplink Configuration context for the port.)
Default values Not applicable
17.3.5
Set Ring Coupling Uplink Path-Cost
Syntax [no] path-cost <auto|COST>
Context Ring Coupling Uplink Configuration context
Usage Configure uplink path-cost. By default, the path-cost depends on the link
speed and duplex mode (higher speed gives lower cost). It is also possible
to set a cost manually in range 1..232 -1 (1..4294967295).
The path-cost is used when electing the active uplink – the link with the
lowest cost will be the active uplink. If the costs of two uplinks are equal,
their uplink priority (section 17.3.6) is considered. For more details, see
section 17.1.3.
Use ”path-cost auto” to have the uplink’s path-cost depend on its link
speed and duplex mode. Use ”path-cost COST” to set a static path-cost
© 2017 Westermo Teleindustri AB
417
Westermo OS Management Guide
Version 4.21.1-0
for the uplink. ”no path-cost” will reset the path cost to the default setting
(auto).
”show path-cost” will show the configured uplink path-cost.
Default values Auto
17.3.6
Set Ring Coupling Uplink Priority
Syntax [no] priority <1..65535> [adjust <DELTA>]
Context Ring Coupling Uplink Configuration context
Usage Configure uplink priority, and optionally enable sticky uplink election by
setting adjust value.
ˆ Use ”priority VALUE” to set priority value. A lower value increases
the chance for this uplink to be elected as active uplink (lower is better).
With equal path-cost (section 17.3.5), an uplink with ”priority 100”
is preferred as uplink over an uplink with ”priority 110”.
ˆ Use the optional ”adjust DELTA” setting to improve its priority (i.e.,
lower its priority with the specified ”DELTA”) once the uplink is elected
as active uplink. This gives a sticky uplink behaviour where shifting
active uplink will be less common.
Consider the following example with uplink A (”priority 100”), uplink
B (”priority 110 adjust 20”), and uplink C (”priority 120 adjust
40”), and where uplink A came up first and is the active uplink. If uplink
A goes down, uplink B takes over as it has lower priority than uplink
C (110 < 120). Uplink B will then apply its adjustment and announce
priority to 90 (110 − 20) in its hello messages. Uplink B will stay as
active uplink even if uplink A comes up again (90 < 100).
”show priority” will show the configured uplink priority.
Default values priority 128 (no adjustment)
17.3.7
Set Ring Coupling Uplink Echo Interval
Syntax [no] echo-time <20..1000>
Context Ring Coupling Uplink Configuration context
418
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Usage Use ”echo-time VALUE” to set the uplink echo interval (in milliseconds)
to check the integrity of the uplink.
”no echo-time” resets the configured echo interval to the default setting
(200 milliseconds).
Use ”show echo-time” to show the configured echo interval.
Default values 200 (msec)
17.3.8
Managing Multi-Link Dual-Homing
Syntax [no] dual-homing [ID]
Context Global Configuration context
Usage Use ”dual-homing ID” to enter Dual-Homing Configuration context of
the given Dual-Homing instance ID. Default instance ID is ”1”, thus command ”dual-homing” will enter the context of dual-homing instance 1.
”dual-homing ID” creates a dual-homing instance with given ID, unless it
already exists.
Up to MAX_DUAL_HOMING_INSTANCES (section 17.4) can be created.
Use ”no dual-homing ID” to remove a specific dual-homing instance, or
”no dual-homing” to remove all dual-homing instances.
Use ”show dual-homing” to list configuration information on all dual-homing
instances, and ”show dual-homing ID” for configuration information on a
specific dual-homing instance (also available as ”show” command within the
Dual-Homing Configuration context).
Default values Default ID is 1
17.3.9
Enable/Disable Multi-Link Dual-Homing
Syntax [no] enable
Context Dual-Homing Configuration context
Usage Enable or disable a dual-homing instance. Use ”enable” to enable the
dual-homing instance, and ”no enable” to disable the dual-homing instance
(without losing configuration settings for this instance).
© 2017 Westermo Teleindustri AB
419
Westermo OS Management Guide
Version 4.21.1-0
Use ”show enable” to show whether the dual-homing instance is enabled
or disabled.
Default values Enabled
17.3.10
Synchronized Multi-Link Dual-Homing
Syntax [no] synchronized
Context Dual-Homing Configuration context
Usage Enable or disable the dual-homing synchronisation feature. When enabled, preference when selecting active uplink will be given to uplinks where
the uplink peer announces that it has connectivity to ring-coupling node with
active uplink. See section 17.1.2.1 for more information.
Use ”synchronized” to enable and ”no synchronized” to disable synchronised dual-homing.
Use ”show synchronized” to show whether synchronised dual-homing is
enabled or disabled.
Default values Enabled
17.3.11
Managing Multi-Link Dual-Homing Uplink Ports
Syntax [no] uplink [PORT]
Context Dual-Homing Configuration context
Usage Use ”uplink PORT” to define the given port as uplink for this dual-homing
node, and enter the Dual-Homing Uplink Configuration context for the port.
A port can be an Ethernet port (chapter 10), a DSL port (chapter 12 and
chapter 13), or a link aggregate (chapter 19).
Up to MAX_DUAL_HOMING_UPLINKS (section 17.4) can be created.
Use ”no uplink PORT” to remove the give port as uplink for this dualhoming node, or use ”no uplink” to remove all uplinks for the node.
Use ”show uplink” to list configuration information for all uplinks, and ”show
uplink PORT” to list uplink configuration settings for the given port (also
available as ”show” command within the Dual-Homing Uplink Configuration
context for the port.)
420
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Default values Not applicable
17.3.12
Set Multi-Link Dual-Homing Uplink Path-Cost
Syntax [no] path-cost <auto|COST>
Context Dual-Homing Uplink Configuration context
Usage Configure uplink path-cost. By default, the path-cost depends on the link
speed and duplex mode (higher speed gives lower cost). It is also possible
to set a cost manually in range 1..232 -1 (1..4294967295).
The path-cost is used when electing the active uplink – the link with the
lowest cost will be the active uplink. If the costs of two uplinks are equal,
their uplink priority (section 17.3.13) is considered. For more details, see
section 17.1.3.
Use ”path-cost auto” to have the uplink’s path-cost depend on its link
speed and duplex mode. Use ”path-cost COST” to set a static path-cost
for the uplink. ”no path-cost” will reset the path cost to the default setting
(auto).
”show path-cost” will show the configured uplink path-cost.
Default values Auto
17.3.13
Set Multi-Link Dual-Homing Uplink Priority
Syntax [no] priority <1..65535> [adjust <DELTA>]
Context Dual-Homing Uplink Configuration context
Usage Configure uplink priority, and optionally enable sticky uplink election by
setting adjust value.
ˆ Use ”priority VALUE” to set priority value. A lower value increases
the chance for this uplink to be elected as active uplink (lower is better).
With equal path-cost (section 17.3.12), an uplink with ”priority 100”
is preferred as uplink over an uplink with ”priority 110”.
ˆ Use the optional ”adjust DELTA” setting to improve its priority (i.e.,
lower its priority with the specified ”DELTA”) once the uplink is elected
© 2017 Westermo Teleindustri AB
421
Westermo OS Management Guide
Version 4.21.1-0
as active uplink. This gives a sticky uplink behaviour where shifting
active uplink will be less common.
Consider the following example with uplink A (”priority 100”), uplink
B (”priority 110 adjust 20”), and uplink C (”priority 120 adjust
40”), and where uplink A came up first and is the active uplink. If uplink
A goes down, uplink B takes over as it has lower priority than uplink
C (110 < 120). Uplink B will then apply its adjustment and announce
priority to 90 (110 − 20) in its hello messages. Uplink B will stay as
active uplink even if uplink A comes up again (90 < 100).
”show priority” will show the configured uplink priority.
Default values priority 128 (no adjustment)
17.3.14
Set Multi-Link Dual-Homing Uplink Echo Interval
Syntax [no] echo-time <20..1000>
Context Dual-Homing Uplink Configuration context
Usage Use ”echo-time VALUE” to set the uplink echo interval (in milliseconds)
to check the integrity of the uplink.
”no echo-time” resets the configured echo interval to the default setting
(200 milliseconds).
Use ”show echo-time” to show the configured echo interval.
Default values 200 (msec)
422
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
17.3.15
Show FRNT Ring Coupling Status
Syntax show coupling
Context Admin Exec context
Usage Use ”show coupling” to show status of FRNT Ring Coupling.
Default values Not applicable
Example
example:/#> show coupling
==================================================================== ID 0x0101
Local uplink(s) -------------------------------------------------------------ID MAC
Uplink
Prio/Delta Cost
Speed
Hello
----------------------------------------------------------------------------->> 7 00:07:7c:84:90:44 eth 7
128/0
200000
100-Full
100(100)ms
Global uplink(s) ------------------------------------------------------------MAC
Uplink
Prio
Cost
Hello
-----------------------------------------------------------------------------00:07:7c:87:85:62 eth 7
128
200000
100(100)ms
example:/#>
The active uplink is marked with >>. In this case, lowest MAC address was used
as tie-breaker to elect active uplink.
17.3.16
Show Multi-Link Dual-Homing Status
Syntax show dual-homing
Context Admin Exec context
Usage Use ”show dual-homing” to show status of Multi-Link Dual-Homing instances.
Default values Not applicable
© 2017 Westermo Teleindustri AB
423
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/#> show dual-homing
==============================================================================
Instance ID: 0x0001
Synchronized mode
: Enabled
Local uplink(s) -------------------------------------------------------------ID MAC
Uplink
Prio/Delta Cost
Speed
Sync
Pref
-----------------------------------------------------------------------------4 00:07:7c:10:df:00 eth 4
100/0
------- -------- No
No
>> 3 00:07:7c:10:df:00 eth 3
128/0
200000
100-Full No
No
==============================================================================
Instance ID: 0x0002
Synchronized mode
: Enabled
Local uplink(s) -------------------------------------------------------------ID MAC
Uplink
Prio/Delta Cost
Speed
Sync
Pref
-----------------------------------------------------------------------------6 00:07:7c:10:df:00 eth 6
128/0
------- -------- No
No
>> 5 00:07:7c:10:df:00 eth 5
128/0
200000
100-Full No
No
example:/#>
The active uplink is marked with >>. In this case, only one uplink was up in each
of the dual-homing instances.
424
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
17.4
Feature Parameters
MAX_RING_COUPLING_INSTANCES
MAX_RING_COUPLING_UPLINKS
MAX_DUAL_HOMING_INSTANCES
MAX_DUAL_HOMING_UPLINKS
© 2017 Westermo Teleindustri AB
1
4
8
4
425
Westermo OS Management Guide
Version 4.21.1-0
Chapter 18
Spanning Tree Protocol - RSTP
and STP
The spanning tree protocol (STP) and its successor rapid spanning tree protocol
(RSTP) are the standard protocols to support redundancy while avoiding broadcast storms in switched networks. WeOS supports RSTP with fall-back to STP
when connecting the switch to another device only capable of STP.
STP/RSTP does not provide the same convergence performance as FRNT, however, STP/RSTP can handle arbitrary switched topologies, while FRNT operates in
a ring structure. For information on FRNT, and coexistence between FRNT and
RSTP, see chapter 16 .
RSTP is disabled at factory default.
18.1
Overview of RSTP/STP features
Table 18.1 provides a summary of available RSTP/STP features in WeOS. Further
descriptions of the spanning tree protocol and the available features are provided
in sections 18.1.1-18.1.3.
18.1.1
Spanning Tree Introduction
Loops in switched networks are dangerous, since packets can loop around forever
and jam the network - as opposed to IP and routed networks, Ethernet frames do
426
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Feature
Enable STP
Bridge priority
Max age
Hello time
Forward delay
View general RSTP/STP settings
Per Port settings
Enable STP
Admin Edge
Path Cost
View per port RSTP/STP settings
View RSTP/STP status
Web
X
X
X
X
X
X
CLI
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
General Description
Section
Section
Section
Section
18.1.2
18.1.1
18.1.1
18.1.1
Section 18.1.1
Section 18.1.3
Table 18.1: Summary of RSTP/STP features.
not include a hop count by which the switches could decide to drop a packet circulating around. Since a switched network may contain multiple loops, broadcast
packets (or other packets flooded by the switches), leads to packet proliferation;
this situation is generally referred to as a broadcast storm. On the other hand,
loops in switched networks are desirable from a redundancy perspective.
Note
The purpose of the spanning tree protocol is to ensure that an arbitrary
physical LAN topology is turned into a logical tree topology (i.e., loop free) in
such a way that all links in the network are still connected (i.e., a spanning
tree). This is accomplished by having the switches put some of their ports
in blocking state.
Since loops in switched networks are so dangerous, layer-2 redundancy protocols
such as STP and RSTP are very restrictive before putting a link in forwarding
state. The main difference between STP and RSTP is that RSTP is able to react
quicker to topology changes, thus can open an alternative path if a link in the
active tree is broken, i.e., RSTP has shorter convergence time than STP. (FRNT
has even faster convergence, see chapter 16.)
In RSTP/STP terminology, a switch is referred to as a bridge. Spanning tree is a
© 2017 Westermo Teleindustri AB
427
Westermo OS Management Guide
Version 4.21.1-0
Host
RSTP
Switch
RSTP
Switch
RSTP
Switch
RSTP
Switch
RSTP
Switch
RSTP
Switch
RSTP
Switch
RSTP
Switch
RSTP
Switch
RSTP
Switch
Host
Host
Host
Host
Host
Figure 18.1: Example of RSTP creating a spanning tree. Dashed links have logically been ”cut off” from the active topology by RSTP, eliminating the loops.
plug-and-play protocol - bridges can use RSTP/STP to form a tree without need
for any configuration. However, the protocol provides a set of parameters which
the operator can use to fine-tune the network setup. Below is a list of those
parameters of specific interest for the WeOS RSTP/STP implementation:
ˆ Bridge priority: Used for root bridge and designated bridge election. See
section 18.1.2.
ˆ Port/Path cost: Each port is assigned a ”cost”. This is used by each bridge to
find the least cost path to the root bridge as part of the tree establishment.
See section 18.1.3.
ˆ Max age/Hello time: Used to detect that a STP/RSTP neighbour is down. The
max age also puts a protocol limit to the size of the network1 .
ˆ Forward Delay: Used when operating in STP mode (i.e., not RSTP). Defines
the time period by which the protocol can be sure that STP information on a
topology change has propagated from one side of the network to the other.
The STP convergence time is limited by twice the forwarding delay (plus the
time it takes to detect the topology change).
ˆ Admin Edge: Ports where only end nodes connect are referred to as edge
ports. If a port is only used for connecting hosts (i.e., no risk for loops), it
1 In RSTP the Message Age field in the Hello Messages effectively acts as a hop count, counting
the distance from the Root. If the Message Age exceeds the Max Age the packet is dropped. Thus,
the setting of the Max Age parameter restricts the size of the RSTP LAN.
428
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
can be configured as an admin edge port.
Access ports and inter-switch ports
It is recommended that all ”inter-switch ports” (ports connecting
switches) are configured as ”non-edge ports” (admin edge disabled),
and that all ”access ports” (ports where hosts connect) are configured
as ”edge ports” (admin edge enabled).
For robustness purposes, all ports are set to ”no admin-edge” when
spanning tree is enabled. To improve performance on ”access ports”
(leading to hosts), these ports should be configured as ”admin-edge”.
When configured as admin edge the port will:
– be put in FORWARDING state quickly after system boot, and
– be kept in FORWARDING state during periods when the spanning tree
topology is changing.
An admin edge assumes the port leads to a host or a router (i.e., not another
bridge), and the port is therefore put in FORWARDING state without first verifying that the LAN is still loop free. The bridge will still send Hello Messages
on admin edge ports, and will react on any incoming Hello Messages as it
would on regular (non-”admin edge”) ports. Thus, even if loops may occur
via an admin edge port, the bridge will generally be able to receive the highpriority RSTP messages, and cut the loop by putting the appropriate port in
BLOCKING.
The IEEE std 802.1D-2004 specifies restrictions on the Max age parameter with
respect to the Hello time and the Forward delay as shown below. This affects how
these parameters can be configured.
ˆ M ge ≥ 2 ∗ (Heo tme + 1)
ˆ M ge ≤ 2 ∗ (Forrd Dey − 1)
Note
Some of the RSTP/STP parameters (Max age, Hello time, and Forward Delay)
need to be set consistently throughout all bridges with the LAN infrastructure. Therefore, bridges inherit these parameter values from the current root
bridge, irrespective of the corresponding parameter setting in the bridge itself.
© 2017 Westermo Teleindustri AB
429
Westermo OS Management Guide
Version 4.21.1-0
18.1.2
Bridge Identity
Each bridge is assigned an 8 byte bridge identifier (bridge ID) as shown in fig. 18.2.
System ID
Extension
4 bits
12 bits
Priority
Unique Bridge Address (MAC)
6 bytes
System ID
Figure 18.2: Structure of bridge ID.
The bridge ID is divided into a priority part (4 bits) and a system ID (60 bits). The
bridge with the lowest bridge ID within the LAN will become the root bridge, i.e.,
lower priority means greater chance to become root bridge. The bridge ID is also
used to select a designated bridge on a link, when multiple bridges on the link
have the same ”least cost path” to the root bridge.
The format of the bridge ID follows IEEE std. 802.1D-2004 (RSTP). It differs from
the structure specified in IEEE std. 802.1D-1998 (STP), where the priority field
was 2 bytes and the system ID field was 6 bytes. The change in structure was
made with respect to the multiple spanning tree protocol (MSTP) defined in IEEE
std. 802.1Q-2005 (WeOS currently does not support MSTP).
ˆ Priority (4 bits): Can take values in range 0-15, where 8 is default. 0 (zero)
means highest priority and 15 lowest priority. Compared to the ”old” 2 byte
priority field of STP, this is rather a priority factor field, which can be multiplied by 4096 to get the ”old” STP priority.
ˆ System ID Extension (12 bits): Set to all zeros in WeOS.
ˆ Unique Bridge Address: Tie-breaker ensuring the bridge ID will be unique.
WeOS uses the base MAC address assigned to the switch for this field.
18.1.3
Path Cost
Each port is associated with a cost referred to as a path cost. Low-speed links are
generally given a high cost, which increases the probability of the port ending up
in blocking state (and vice versa), in case spanning tree discovers a loop.
430
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
By default, the path cost of a port is assigned dynamically with values related
to the port speed (in-line with the recommendations of IEEE std 802.1D-2004).
The same path costs are used irrespective if the port is operating in RSTP or STP
mode.
Port Speed (Mbit/s)
RSTP path cost
10
100
1000
2000000
200000
20000
It is also possible to configure the path cost manually. That may be useful to get
more fine grain control of which port in the LAN should be put in blocking state.
Setting path costs manually may be desirable when operating a LAN including a
mix of RSTP and STP capable, since STP uses a different set of default path costs.
18.1.4
RSTP and STP coexistence
WeOS supports both RSTP and STP, but WeOS always attempts to run RSTP on
every spanning-tree enabled port. WeOS automatically shifts to STP mode on
a port, if it detects a bridge running STP on that port. Other ports continue
operating in RSTP mode. When operating a network including a mix of RSTP and
STP bridges, it may be necessary to configure path costs manually to get the
intended spanning tree behaviour, see also section 18.1.3.
© 2017 Westermo Teleindustri AB
431
Westermo OS Management Guide
Version 4.21.1-0
18.2
18.2.1
Managing RSTP via the web interface
Managing RSTP Settings
Menu path: Configuration ⇒ L2 Redundancy ⇒ RSTP
On the RSTP configuration page you will be presented to the current settings for
RSTP on your switch, see below. You may change the settings by editing the
page.
Enabled
Bridge Priority
Maximum Age Timeout
432
Check the box to enable RSTP. If you have a
JavaScript enabled browser the other settings
will not be displayed unless you check this box.
A priority level used in root bridge selection.
A lower value increases the probability for this
switch to be elected as root bridge.
The time the unit will wait before considering a
neighbour designated bridge is down after the
last Hello message was heard from the neighbour.
Continued on next page
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Hello Time Interval
Forward Delay Timeout
Edge Port
Continued from previous page
The time between two consecutive transmissions of hello messages.
The time an interface takes to change from
blocking to forwarding state. Only used when
operating in STP mode.
Ports connected to end hosts and routers (i.e.,
not to another switch) can be set as adminedge ports. This avoids unnecessary BLOCKING of such ports at system startup or when a
topology change occurs.
It is recommended that this box is checked for
every port where it is certain that only end
hosts and routers connect. Ports which (may)
connect to another switch should un-check this
box.
© 2017 Westermo Teleindustri AB
433
Westermo OS Management Guide
Version 4.21.1-0
18.2.2
RSTP Status and Statistics
Menu path: Status ⇒ L2 Redundancy ⇒ RSTP
Version
Topology Change
Count
Time Since Last
Topology Change
ID
434
Always RSTP, with fallback to STP.
Number of RSTP topology changes since switch startup.
Time since last topology change.
The
local
and
elected
root
bridge
ID,
used
for
root
bridge
and
designated
bridge
election;
consists
of
two
parts:
MAC Address The local MAC-address that is used
for bridge ID. If local and root values
are equal, this switch is root.
Priority
Priority value configured on the unit.
Continued on next page
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Root Port
Root Path Cost
Max Age
Hello Time
Forward Delay
Auto Refresh
Refresh
Continued from previous page
The port with the open path to the root switch. If
this switch is root, the text Unit is root will be displayed.
Calculated cost to designated root switch.
Used to detect that a STP/RSTP neighbor is down. Current value learnt from BPDUs.
The time between two consecutive transmissions of
hello messages. Current value learnt from BPDUs.
Used when operating in STP mode (i.e., not RSTP). Defines the time period by which the protocol can be sure
that STP information on a topology change has propagated from one side of the network to the other. Current value learnt from BPDUs.
Click on a value to make the page reload with updated
statistics automatically every 5, 15, 30 or 60 seconds.
Click Off to turn off auto refresh.
Click on this button to reload with updated statistics.
Port Status
Label
Type
Path Cost
State
Edge
Designated Bridge
Port label, identifying the port.
Type of port, e.g. Eth for ethernet.
Path cost associated with the port.
FORWARDING Unit forwards packets. Normal operation.
LEARNING
The port is preparing itself for entering FORWARDING state.
BLOCKING
Unit does not forward any packets.
DISABLED
Port does not participate in operation.
If TRUE the port is in admin edge mode and assumes
the port leads to a host or a router (i.e., not another
bridge), and the port is therefore put in FORWARDING
state without first verifying that the LAN is loop free.
If FRNT, the port is controlled by FRNT protocol.
The designated bridge MAC-address.
© 2017 Westermo Teleindustri AB
435
Westermo OS Management Guide
Version 4.21.1-0
18.3
Managing RSTP via the CLI
Command
Spanning Tree Configuration
[no] spanning-tree
priority <0-15|0-65536>
max-age-time <6-40>
hello-time <1-10>
forward-delay <4-30>
stp-port <PORTLIST|all>
[no] enable
[no] admin-edge
[no] path-cost <0-20000000>
Default
Section
Disabled
8 (32768)
20
2
15
Section
Section
Section
Section
Section
18.3.1
18.3.2
18.3.3
18.3.4
18.3.5
Enabled
Disabled
0 (Auto)
Section
Section
Section
Section
18.3.6
18.3.7
18.3.8
18.3.9
Spanning Tree Status
show spanning-tree
18.3.1
Section 18.3.10
Manage RSTP
Syntax [no] spanning-tree
Context Global Configuration context
Usage Enter Spanning Tree Configuration context, and activate spanning-tree (if
not already activated). Use ”no spanning-tree” to disable spanning-tree
and to remove spanning-tree configurations.
Use ”show spanning-tree” to view general spanning-tree settings, given
that spanning-tree is enabled (also available as ”show” command within the
Spanning Tree Configuration context.
Default values Disabled
18.3.2
Bridge Priority Setting
Syntax priority <0-15|0-65535>
Context Spanning Tree Configuration context
436
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Usage Set bridge priority, where a low value means high priority, which increase
the probability of being elected as root bridge. Values can be entered in
two ways, either in range 0-15, which corresponds to the 4-bit priority field
specified in IEEE std 802.1D-2004, or in range 16-65535 which corresponds
to the traditional 2 byte priority field defined in IEEE 802.1D-1998. In the
latter case, the value is divided by 4096, and stored as a value 0-15. See
section 18.1.2 for more information.
”no priority” resets the bridge priority to the default setting.
Use ”show priority” to view the current bridge priority setting.
Default values 8 (32768)
18.3.3
Max Age Setting
Syntax max-age-time <6-40>
Context Spanning Tree Configuration context
Usage Set spanning-tree max age timeout. Since bridges use the max age configured at the root bridge, this parameter setting only matters if this bridge
becomes the root bridge.
”no max-age-time” resets the max age timeout to the default setting.
Use ”show max-age-timeout” to view the current max age timeout setting.
Default values 20
Error messages An error message is given if the ”max-age-time” is not given
a valid value with respect to ”hello-time” or ”forward-delay”, see section 18.1.1.
18.3.4
Hello Interval
Syntax hello-time <1-10>
Context Spanning Tree Configuration context
Usage Set spanning-tree hello time interval. Since bridges use the hello time
configured at the root bridge, this parameter setting only matters if this
bridge becomes the root bridge.
© 2017 Westermo Teleindustri AB
437
Westermo OS Management Guide
Version 4.21.1-0
”no hello-time” resets the hello time to the default setting.
Use ”show hello-time” to view the current hello time setting.
Default values 2 (seconds)
Error messages An error message is given if the ”hello-time” is not given a
valid value with respect to ”max-age-time”, see section 18.1.1.
18.3.5
Forward Delay
Syntax forward-delay <4-30>
Context Spanning Tree Configuration context
Usage Set spanning-tree forward delay. Since bridges use the forward delay
configured at the root bridge, this parameter setting only matters if this
bridge becomes the root bridge.
”no forward-delay” resets the forward delay to the default setting.
Use ”show forward-delay” to view the current forward delay setting.
Default values 15 (seconds)
Error messages An error message is given if the ”forward-delay” is not given
a valid value with respect to ”max-age-time”, see section 18.1.1.
18.3.6
Manage RSTP Ports
Syntax [no] stp-port <PORTLIST|all>
Context Spanning Tree Configuration context
Usage Enter Spanning Tree Port Configuration context to manage per port spanningtree settings for one or more ports.
”no stp-port <PORTLIST|all>” (e.g., ”no stp-port all”) will disable spanning tree for the specified ports.
Use ”show stp-port <PORTLIST|all>” to view the spanning tree settings
for the specified port(s).
Default values Not applicable.
438
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/config/spanning-tree/#> show stp-port all
Port
Enabled Admin-Edge Path-cost
==============================================================================
Eth 1
YES
NO
AUTO
Eth 2
YES
NO
AUTO
Eth 3
YES
YES
AUTO
Eth 4
YES
YES
AUTO
Agg A1
YES
YES
20000
Agg A2
NO
NO
AUTO
example:/config/spanning-tree/#>
18.3.7
Enable Spanning Tree on a Port
Syntax [no] enable
Context Spanning Tree Port Configuration context
Usage Enable the spanning tree protocol on a port. Use ”no enable” to disable
spanning tree protocol on a port.
Default values Enabled
18.3.8
Admin Edge Setting
Syntax [no] admin-edge
Context Spanning Tree Port Configuration context
Usage Configure the port as an access port (”admin-edge”), or as an interswitch port (”no admin-edge”).
Note
It is recommended that every port where it is certain that only end
hosts and routers connect (but not switches/bridges) are configured as
”admin-edge”. Port which (may) connect to another switch/bridge
should be configured as ”no admin-edge”.
Use ”show admin-edge” to view the admin edge setting for this port.
Default values Disabled (”no admin-edge”)
© 2017 Westermo Teleindustri AB
439
Westermo OS Management Guide
Version 4.21.1-0
18.3.9
Path Cost Setting
Syntax [no] path-cost <0-20000000>
Context Spanning Tree Port Configuration context
Usage Configure the spanning tree path cost for a port. A low speed link should
get a higher cost, a high speed link a lower cost. Use ”path-cost 0” (or
”no path-cost”) to have the path-cost assigned automatically depending
on the port speed (see section 18.1.3).
Values in range 1-20000000 means a statically configured path cost of the
given value.
Use ”show path-cost” to view the path cost setting for this port.
Default values Automatic (”path-cost 0”)
18.3.10
Show RSTP Status
Syntax show spanning-tree
Context Admin Exec context.
Usage Show spanning-tree status information, including current port states, root
bridge ID, etc..
Default values Not applicable.
440
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Chapter 19
Link Aggregation
This chapter describes WeOS support for link aggregation (IEEE 802.3ad/802.1AX[17]).
With link aggregation, two or more Ethernet links can be bundled and treated as
a single MAC entity by the upper layer protocols. The primary use is to achieve
redundancy in layer-2 bus topologies. A coarse form of load balancing is also
provided, but only if different traffic flows are mapped to different aggregate
member links.
WeOS supports the standard Link Aggregation Control Protocol (LACP[17]) for
aggregation control, but also static aggregation control, where the active set of
member links is solely determined based on their link up/down state.
19.1
Link Aggregation Support in WeOS
Feature
Enable/Disable Aggregate
Define Member Ports
Static Aggregation Control
LACP Aggregation Control
Timeout (Short/Long)
Active/Passive
Show Link Aggregate Status
© 2017 Westermo Teleindustri AB
Web
X
X
X
X
X
X
CLI
X
X
X
X
X
X
General Description
Section 19.1.1
-”Section 19.1.2
Section 19.1.3
-”-”-
441
Westermo OS Management Guide
Version 4.21.1-0
19.1.1
Introduction to Link Aggregation
Link aggregation enables physical links to be bundled together to form a single
logical link, an aggregated link, see fig. 19.1. Upper layer protocols will treat the
aggregate as a single MAC entity, i.e., as one Ethernet port with its own label, a
MAC address assigned, and so on. In WeOS, aggregates are named ”a0”, ”a1”,
etc., and inherit their MAC address from one of their member ports.
Physical view
Switch
Member link
Switch
Aggregated link
Logical view
Switch
Switch
Aggregated link
Figure 19.1: Example of link aggregation with four member links
All member ports in an aggregate are able to forward data. However, the IEEE802.1AX
standard[17] mandates the aggregate to deliver packets in order per data flow
to avoid problems for upper layer protocols. This means the switch will send all
traffic of an individual data flow through the same member link. Other flows may
be sent through other member links. The effectiveness of this load balancing
depends on several factors:
ˆ The granularity by which the switch can distinguish between different traffic
flows: WeOS units determine packet flow based on the combination of the
442
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
source and destination MAC address of the packet1 (done in hardware).
ˆ The distribution of traffic flows:. If there are many flows (and if they are of
equal load) the ability to load balance improves. This depends on the traffic
patterns in your network. Avoiding patterns where all traffic end up with the
same source and destination MAC over the aggregate improves the ability
to load balance2 .
ˆ The mapping of traffic flows to different member links: WeOS units map
traffic flows to different (active) member links in a static way. This mapping
aims to equalise the number of flows mapped to each member link, but its
effectiveness is limited when the number of flows are low.
Note
To summarise, link aggregation should generally be used as a means to
achieve redundancy in bus topologies. It may be used to increase data
capacity, however, the ability to load balance between the member links is
limited and depends on the use case.
When an aggregate is configured in WeOS, the following restrictions apply:
ˆ Ethernet as member ports: Only aggregation of Ethernet ports is supported.
ˆ Member ports explicitly associated with aggregate: For a port to be part of
an aggregate, it must explicitly be associated with that aggregate.
ˆ Maximum 8 aggregates: At most 8 aggregates can be configured on a WeOS
unit.
ˆ Maximum 8 member ports per aggregate: Each aggregate can have at most
8 member ports.
ˆ Member ports in same slot: In slot based WeOS products (see section 10.1.1)
all member ports must reside in the same slot as of WeOS v4.21.1. Similar
restrictions apply to WeOS Viper, RedFox Rail (RFR) and RedFox Industrial
Rack (RFIR) products.
A aggregate has state Down when all its member ports have state Down, and the
aggregate is Up when at least one of its member ports has state Up.
1 The
algorithm to determine flow uses a hash function applied to the packet’s source and destination MAC address.
2 Switching traffic over the link aggregate may improve load balancing as opposed to routing
(routers typically use the same source and destination MAC for all unicast traffic). Multicast flows
commonly utilise different destination MACs irrespective if the WeOS units are switching or routing,
thus has good load balancing properties.
© 2017 Westermo Teleindustri AB
443
Westermo OS Management Guide
Version 4.21.1-0
The next subsections provide additional information on WeOS support for link
aggregates: sections 19.1.2 (static) and 19.1.3 (LACP) contain information on
the methods to control link aggregates in WeOS, while section 19.1.4 include
more details on using link aggregates in various low-layer features in WeOS.
19.1.2
Static Link Aggregates
For static link aggregates the including member ports are the only settings that
have to be specified in the configuration. The members in an aggregate do not
need to have the same speed settings, although that is the preferred setting
(otherwise the capacity of the aggregate will be unbalanced).
Ports that are included in an aggregate and have link up will be qualified as active
ports, and the network traffic will be sent on those links. If a link goes down or up
in the aggregate the network traffic will be distributed over the new set of active
links. Because an active link in an aggregate is qualified on the link status no
media converters are allowed between statically aggregated ports. Below is a
CLI configuration example where the static link aggregate a1 is configured with
member ports 3 and 7 on a WeOS switch.
Example
example:/#> configure
example:/config/#> aggregate a1
example:/config/aggregate-a1/#> ports 3,7
example:/config/aggregate-a1/#> type static
example:/config/aggregate-a1/#> show
Name
: a1
Status
: Enabled
Type
: static
Ports
: 3,7
example:/config/aggregate-a1/#> end
example:/config/#>
19.1.3
LACP Controlled Link Aggregates
The Link Aggregation Control Protocol (IEEE 802.3ad/802.1AX [17]) is a standard
method for aggregating member links that have the same speed and duplex
mode. The primary advantage over static link aggregation is the ability to confirm
that the remote partner can handle aggregation. It is also possible to handle
failover when media converters are present.
444
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
LACP relies upon periodic transmission of information and state between the
switches. The protocol messages (LACP-PDUs) are sent by the first party (the
Actor) to the second party (the Actor’s protocol Partner) with information about
what the Actor knows, both about its own state and that of the Partner.
Switches can be configured to active or passive participation in LACP. Passive
LACP indicates the preference for not transmitting LACP-PDUs unless its Partner
is Active LACP, i.e. it does not generate any LACP traffic by its own. Active LACP
indicates the preference to participate in the protocol regardless of the Partner
setting, i.e. it always generates LACP traffic.
LACP-PDUs are transmitted periodically when either the Actor or the Partner is
configured with Active LACP. These transmissions will occur at either a fast or slow
transmission rate depending upon the timeout setting (short or long timeout) of
the Partner system.
The LACP state is determined by the contents of the LACP-PDUs and can be in
any of the following states:
Detached The port is being detached from the aggregator.
Waiting The port is being attached to the aggregator.
Attached The port is attached to the selected aggregator.
Collecting Indicates that the receive function of this link is enabled.
Distributing Indicates that the transmit function of this link is enabled.
The switch will set a member port in forwarding state when LACP state is Distributing. For all other LACP states the port state will be blocking3 . The aggregate
is in forwarding state as long as at least one member port is in forwarding state.
Also, the aggregate will be up as long as at least one member port is up.
WeOS assumes that the configured aggregate connects two switches. If the aggregate member ports on one switch is connected to several other switches LACP
will only include member ports to one of the neighbours in the active port set:
ˆ Ports to the neighbour with the highest total bandwidth will be selected.
ˆ If several aggregates share the same bandwidth, then the aggregate is selected based on LACP system priority, system identifier, port priority, and
operational key.
3 If RSTP or FRNT are run over the aggregate, those protocols may also decide to set the ports in
blocking state.
© 2017 Westermo Teleindustri AB
445
Westermo OS Management Guide
Version 4.21.1-0
In WeOS v4.21.1, the LACP system priority is set to 0x8000 (hex), system identifier is set to the MAC address of the first member port of the aggregate, the
port priority is set to 0x8000 (hex), and the operational key is set to the configured aggregate identifier (see sections 19.2 and 19.3). More information about
aggregate selection can be found in IEEE 802.3ad/802.1AX [17].
19.1.4
19.1.4.1
Link Aggregates and Low layer protocols
Link Aggregation and VLAN
Ethernet and DSL ports on WeOS units are associated (untagged or tagged) with
one or more VLANs as described in chapter 15. Link aggregates can not be
mapped directly to VLANs. Instead the user must add each of the aggregate
member ports to the intended VLAN(s).
For the setup in fig. 19.2, the physical ports 1-4 are mapped tagged (”tagged
1-4”) to VLANs 1&2 rather than the aggregates (i.e., ”tagged a1,a2” is not
possible as of WeOS v4.21.1). An extract of the configuration file is shown below.
Example
vlan 1
name vlan1
untagged 5-7
tagged 1-4
end
vlan 2
name vlan2
untagged 8-10
tagged 1-4
end
19.1.4.2
Link Aggregation and Link Alarms
As described in section 24.1 the operational state (Up/Down) of Ethernet and DSL
ports can be used as alarm triggers, i.e., link alarms. When a port is a member of
a link aggregate, it is still possible to define link alarms for the individual member
ports. It is also possible to create link alarms for the aggregates.
Below is a CLI configuration example where a link alarm is configured for aggregate a1. The aggregate has state Down when all its member ports has state
446
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Aggregate "a1"
VLAN 1&2
(tagged)
1
3
2
4
5
6 7
8
9 10
Aggregate "a2"
VLAN 1&2
(tagged)
VLAN 1 VLAN 2
(untagged) (untagged)
Figure 19.2: The physical ports 1-4 rather than the logical aggregates (a1 and
a2) are associated with the VLANs (VLAN 1 and 2).
Down, and the aggregate is Up when at least one of its member ports has state
Up.
Example
example:/#> configure
example:/config/#> alarm
example:/config/alarm/#> trigger link-alarm
example:/config/alarm/trigger-2/#> port a1
example:/config/alarm/trigger-2/#> end
example:/config/alarm/#>
19.1.4.3
Link Aggregation and unicast/multicast MAC learning
The MAC forwarding database (FDB, see section 15.1.8) holds information on
where to forward known MAC addresses. Unicast addresses are learnt dynamically by looking at the source MAC of incoming packets, while multicast addresses
are typically learnt dynamically via IGMP snooping (chapter 20), or entered manually4 by the operator.
When a (unicast/multicast) MAC address is learnt dynamically on a member port
of a link aggregate, all ports of the aggregate are added to the MAC address’ FDB
entry, since the link aggregation flow distribution mechanism can map traffic to
the MAC address on any member port.
In the example below, aggregate a1 consists of member ports 5 and 6, and IGMP
snooping is enabled on the VLAN the ports are associated with. An IGMP report
4 See
section 15.4.3 for CLI command to enter MAC forwarding database entries manually.
© 2017 Westermo Teleindustri AB
447
Westermo OS Management Guide
Version 4.21.1-0
has been received for IP multicast address 225.1.2.3 (MAC 01:00:5e:01:02:03) on
one of the member ports and both ports are added to the forwarding database
for that MAC address.
Example
example:/#> sh ip igmp
VID Querier IP
Querier MAC
Port Interval Timeout
------------------------------------------------------------------------------1 192.168.2.200
LOCAL
VID Multicast Group Filtered MAC Addr Active ports
------------------------------------------------------------------------------1 225.1.2.3
01:00:5E:01:02:03 a1
------------------------------------------------------------------------------Total: 1 filters, max 1200, in 1 VLAN.
example:/#> sh fdb
MAC
VLAN State
Port(s)
===============================================================================
...
01:00:5e:01:02:03
ANY IGMP
5-6
...
===============================================================================
FDB Aging time: 300 sec.
example:/#>
Similarly, traffic from unicast address 00:07:7c:00:02:61 has come in on one
member port, thus both member ports are automatically added to the MAC’s
FDB entry.
448
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/#> sh fdb
MAC
VLAN State
Port(s)
===============================================================================
...
00:07:7c:00:02:61
ANY 294 s
5-6
...
===============================================================================
FDB Aging time: 300 sec.
example:/#>
When adding (multicast) MAC addresses statically to the MAC FDB, each of the
individual member ports needs to be specified. Thus, in the example below, with
ports 5 and 6 belonging to aggregate a1, the command ”mac 01:00:5e:00:11:22
port 5,6” is used (while ”mac 01:00:5e:00:11:22 port a1” would not work as
of WeOS v4.21.1).
Example
example:/#>
example:/#> configure
example:/config/#> fdb
example:/config/fdb/#> mac 01:00:5e:00:11:22 port 5,6
example:/config/fdb/#> end
© 2017 Westermo Teleindustri AB
449
Westermo OS Management Guide
Version 4.21.1-0
19.1.4.4
Running FRNT or RSTP over Link Aggregates
It is possible to run FRNT (chapter 16) or RSTP (chapter 18) over a link aggregate.
Fig. 19.3 shows an example of using FRNT together with link aggregation.
Link Aggregates
Member
1
2
3
4
Member
1
2
Focal
Member
1 Point 3
2
4
3
4
M
1
2
3
4
N
FRNT Ring
Figure 19.3: FRNT can run over aggregated links
Additional information on running RSTP over a link aggregate:
ˆ Failover performance: RSTP failover performance may be degraded when
running RSTP over a link aggregate as opposed to using regular links.
ˆ Forwarding/Blocking state: An aggregate is forwarding data packets only
if both RSTP and the link aggregate itself determine that it should be in
forwarding state.
ˆ RSTP link cost: The RSTP link cost can be configured manually. If ”auto” is
used for cost calculation, WeOS determines the aggregate link cost based
the aggregated bandwidth of the member ports (higher aggregated capacity
gives lower RSTP cost).
ˆ Link Up/Down: An aggregate is up if at least one of its member ports are
considered up. An aggregate is down if all its member ports are down.
Example
example:/#> configure
example:/config/#> spanning-tree
example:/config/spanning-tree/#> stp-port A1
example:/config/spanning-tree/stp-port-A1/#> no admin-edge
example:/config/spanning-tree/stp-port-A1/#> show
Port
Enabled Admin-Edge Path-cost
=======================================================
Agg A1
YES
NO
AUTO
450
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Additional information on running FRNT over a link aggregate:
ˆ Failover performance: FRNT failover performance may be degraded when
running FRNT over a link aggregate as opposed to using regular links.
ˆ Forwarding/Blocking state: An aggregate is forwarding data packets only
if both FRNT and the link aggregate itself determine that it should be in
forwarding state.
ˆ Link Up/Down: An aggregate is up if at least one of its member ports are
considered up. An aggregate is down if all its member ports are down.
ˆ Mixing aggregated and regular links: The topology in fig. 19.3 uses link
aggregation throughout the whole FRNT ring. It is possible to run link aggregation on a subset of the links in the FRNT ring.
Example
example:/#> configure
example:/config/#> frnt 1
example:/config/frnt-1/#>
example:/config/frnt-1/#>
example:/#> show frnt
Ring#
Version
Enabled
Ring Status
Topology Change Count
Time Since Last Change
Mode
Ring ports swapped
Port M
Port N
19.1.4.5
ring-ports A1,A2
leave
:
:
:
:
:
:
:
:
:
:
1
FRNTv0
Yes, running as PID 1585
OK
2
0 Days 0 Hours 0 Mins 14 Secs
Focal Point
No
Agg A1
UP Blocking
Agg A2
UP Forwarding
Link Aggregation and other Low-level WeOS features
Use of link aggregation with other low-level features, e.g., port monitoring (section 7.1.9), port access control (section 15.2 and chapter 9), etc. is not supported
as of WeOS v4.21.1. To use those features together with link aggregation it may
be possible to specify the individual member ports in the configuration, however,
the behaviour is undefined and its use is unsupported.
© 2017 Westermo Teleindustri AB
451
Westermo OS Management Guide
Version 4.21.1-0
19.2
Link Aggregation Settings and Status via the Web
Interface
19.2.1
Configuring Link Aggregation Settings via the Web Interface
Menu path: Configuration ⇒ Port ⇒ Aggregate
On the Link Aggregate overview page all configured link aggregates will be presented in a list, see below.
When first accessing this page link aggregates can be created by pressing the
New button.
Name
Ports
Type
New
452
The link aggregate name.
The set of ports defined for this aggregate.
The type of the aggregate, Static or LACP.
Edit
Click this icon to edit an existing aggregate.
Delete
Click this icon to remove an aggregate. You will be asked to
acknowledge the removal before it is actually executed.
Click the New button to create a new link aggregate.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
19.2.2
Create new link aggregate using the web interface
Menu path: Configuration ⇒ Port ⇒ Aggregate ⇒ New
When clicking the New button, you will be presented to the aggregate new page.
Name
Ports
Type
LACP Mode
LACP Timeout
The link aggregate name. Valid values are A{n} or a{n},
where n is an integer.
The set of ports to be included in this aggregate. Only ports
in the same slot may be aggregated together.
The type of the aggregate, Static or LACP.
Only available for type LACP. Modes:
Active
Always send frames (LACP-PDUs) along
the configured links.
Passive Only send frames (LACP-PDUs) along
the configured links if any LACP-PDU
frames have been received.
Only available for type LACP. The type of the aggregate:
Short 3 seconds
Long 90 seconds
For more information, see section 19.1.
© 2017 Westermo Teleindustri AB
453
Westermo OS Management Guide
Version 4.21.1-0
19.2.3
Edit link aggregate settings using the web interface
Menu path: Configuration ⇒ Port ⇒ Aggregate ⇒
When clicking the Edit icon for an aggregate you will be presented to the aggregate edit page, which is identical to the new page. See section 19.2.2 for
description of fields.
19.2.4
Link Aggregation Status via the Web Interface
Menu path: Status ⇒ Port ⇒ Aggregate
This page display status information for the currently configured link aggregates.
Name
Link
MAC
Type
Port Label
Port Link
454
The link aggregate name.
The aggregate link status. Up/Down.
The aggregate MAC address.
The type of the aggregate, Static or LACP.
The port label for the ports included in the aggregate.
Up/Down.
Continued on next page
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Port Active
Port
Link State
Port
LACP State
Continued from previous page
Indicates if this port is an active member of this aggregate.
The port state for this port.
FORWARDING Unit forwards packets. Normal operation.
LEARNING
The port is preparing itself for entering
FORWARDING state. (Only applicable if
RSTP/STP is used on the aggregate.)
BLOCKING
Unit does not forward any packets. The
port is put in blocking state by LACP, or
by STP/RSTP or FRNT if used on the aggregate.
DISABLED
Port does not participate in operation.
The LACP negotiation state for this port: DETACHED, WAITING, ATTACHED, COLLECTING, or DISTRIBUTING. In the
DISTRIBUTING state, the port is ready to send and receive
data as part of the aggregate. See section 19.1.3 or [17] for
more information.
© 2017 Westermo Teleindustri AB
455
Westermo OS Management Guide
Version 4.21.1-0
19.3
Managing Link Aggregation via CLI
Command
Configure Link Aggregate
[no] aggregate <AGGREGATE_ID>
[no] enable
[no] ports <PORTLIST>
[no] type <static|flhp|lacp>
LACP Specific Settings
[no] active
[no] timeout <short|long>
Aggregate Status
show aggregate
19.3.1
Default
Section
N/A
Enabled
N/A
lacp
Section
Section
Section
Section
active
short
Section 19.3.5
Section 19.3.6
19.3.1
19.3.2
19.3.3
19.3.4
Section 19.3.7
Manage a Link Aggregate
Syntax [no] aggregate <AGGREGATE_ID>
Context Global Configuration context
Usage Create, modify or remove a link aggregate.
Enter the Link Aggregate Configuration context of the given aggregate identifier (a0-aN), where N is a number (up to 8 aggregates can be created). If
this is a new link aggregate, the aggregate is created.
Use ”no aggregate <AGGREGATE_ID>” to remove an existing link aggregate, or ”no aggregate” to remove all link aggregates.
Use ”show aggregate” to list configured aggregates. To list details of a
configured aggregate, enter its configuration context and run ”show” from
there.
Default values When using the ”no aggregate” form (without providing a specific aggregate ID), all link aggregates are removed.
Example Listing configured aggregates, and listing details for a LACP aggregate.
456
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/config/#> show aggregate
a1
static 1-2
a2
lacp 5-6
example:/config/#> aggregate a2
example:/config/aggregate-a2/#> show
Name
: a2
Status
: Enabled
Type
: lacp
Ports
: 5-6
LACP mode
: active
LACP timeout
: short
example:/config/aggregate-a2/#>
19.3.2
Enable/disable a Link Aggregate
Syntax [no] enable
Context Link Aggregate Configuration context
Usage Enable/disable this aggregate instance. Use ”enable” to enable and ”no
enable” to disable this aggregate. When disabled, the configured member
ports will not be part of this aggregate, i.e., they will operate as regular
(non-aggregate) ports.
Use ”show enable” to view the currently configured setting.
Default values Enabled (”enable”)
19.3.3
Configure Link Aggregation Member Ports
Syntax [no] ports <PORTLIST>
Context Link Aggregate Configuration context
Usage Add/remove a list of ports to/from the port member set of this link aggregate. Use ”no ports” (without providing a port list) to remove all ports
from the member set.
Use ”show ports” to view the currently configured list of ports.
Default values When using the ”no ports” form (without providing a specific
PORTLIST), all ports are removed.
© 2017 Westermo Teleindustri AB
457
Westermo OS Management Guide
Version 4.21.1-0
”PORTLIST” is a comma separated list of port ranges without intermediate spaces,
e.g., ”X1-X2,X4”.
19.3.4
Configure Link Aggregate Control Mode
Syntax [no] type <static|flhp|lacp>
Context Link Aggregate Configuration context
Usage Set mode/operation for this aggregate. Use ”no type” (without providing
a mode) to reset to default value.
Warning
As of WeOS version v4.21.1, the use of FLHP for link aggregation control
is provided as a technology preview feature. All use of the FLHP link
aggregation control feature except for testing is discouraged.
Use ”show type” to view the currently configured mode.
Default values lacp (”no type”)
19.3.5
Configure LACP Active/Passive Mode
Syntax [no] active
Context Link Aggregate Configuration context (only available when aggregate
control mode is lacp)
Usage Select LACP mode, i.e. active or passive participation in LACP (see section 19.1.3). Use ”active” to select active mode and ”no active” to select
passive mode.
Use ”show active” to view the currently configured setting.
Default values Active (”active”)
19.3.6
Configure LACP Timeout
Syntax [no] timeout <short|long>
458
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Context Link Aggregate Configuration context (only available when aggregate
control mode is lacp)
Usage Select LACP timeout, i.e. the number of seconds before invalidating received LACP information (see section 19.1.3). Use ”timeout short” to set
the timeout to 3 seconds and ”timeout long” to set the timeout to 90 seconds.
Use ”show timeout” to view the currently configured setting.
Default values Short, i.e. 3 seconds (”no timeout”)
19.3.7
Show Status of Link Aggregates
Syntax show aggregates
Context Admin Exec context
Usage Display status information for all configured aggregates. The header line
displays the aggregate information including the name, its MAC address,
and the aggregate control mode.
Each member link is listed with link status, whether or not the link is currently an active member of the aggregate, and the link state.
Aggregates using LACP also displays the LACP state (see section 19.1.3)
and partner information. Partner ID is the system id of the peer, port is the
remote port, and key is the operational key. In WeOS, the operational key is
equal to the aggregate id.
Default values Not applicable
Example In this example an aggregate (a1) is configured. Both member ports
are up, but port ’Eth 5’ is unused, since no LACP partner has been discovered
on that link.
Example
example:/#> show aggregates
Aggregate a1 MAC: 00:07:7c:00:30:b5 Type: lacp
------------------------------------------------------------------------------Port
Link Active
Link State LACP State
Partner ID
Port Key
------------------------------------------------------------------------------Eth 5
UP No
Blocking
ATTACHED
00:00:00:00:00:00
0
0
Eth 6
UP Yes
Forwarding DISTRIBUTING 00:07:7c:00:02:61
2
1
example:/#>
© 2017 Westermo Teleindustri AB
459
Westermo OS Management Guide
Version 4.21.1-0
Example In this example a static aggregate (a2) is configured. Two member
ports are up and ’Eth 9’ is down.
Example
example:/#> show aggregates
Aggregate a2 MAC: 00:07:7c:84:91:6b Type: static
------------------------------------------------------------------------------Port
Link Active
Link State
------------------------------------------------------------------------------Eth 7
UP Yes
Forwarding
Eth 8
UP Yes
Forwarding
Eth 9
DOWN No
N/A
example:/#>
460
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Chapter 20
Multicast in Switched Networks
This chapter gives a brief overview of multicast, with a focus on IP multicast, and
how it can be controlled in a WeOS device using IGMP snooping. The chapter also
covers non-IGMP capable devices and how they can be integrated into a network
with IGMP enabled.
20.1
Overview
Feature
IGMP Querier Mode
IGMP Query Interval
IGMP Fast Leave
Low Bandwidth Networks
Multicast Router Ports
Multicast Router Timeout
View IGMP Snooping Settings
Web
X
X
X
X
X
X
CLI
X
X
X
X
X
X
X
General Description
Section 20.1.1
-”Section 20.1.3
Section 20.1.4
Section 20.1.2
Section 20.1.1
Multicast, as opposed to unicast, is a very efficient means of communicating
information to more than one receiver. The main difference between multicast
and broadcast is that multicast can be controlled. When disabling its control
mechanisms, like IGMP, multicast behaves like broadcast.
Thus, when distributing IP multicast data in a switched network, switches within
the LAN can:
© 2017 Westermo Teleindustri AB
461
Westermo OS Management Guide
Version 4.21.1-0
ˆ treat multicast traffic as broadcast, i.e., forward it on all ports (in the same
VLAN), or
ˆ limit forwarding of multicast only to subscribers
The latter method requires switches to inspect Internet Group Management Protocol (IGMP) control messages exchanged by hosts and routers to learn which
ports lead to subscribers – this mechanism is referred to as IGMP snooping[4].
With IGMP Snooping enabled, WeOS switches dynamically keep track of up to
2048 multicast addresses1 .
As part of the IGMP snooping support, WeOS also enables a switch to act as IGMP
querier – a role which is usually handled by a multicast router. Having switches
with IGMP querier capabilities enables efficient distribution of IP multicast in networks without multicast routers.
Warning
WeOS devices can only limit the broadcast effects of multicast on a Layer2 basis, it is therefore important to design IPv4 multicast networks so
that groups do not overlap. For example, 225.1.2.3 and 226.1.2.3 map
to the same multicast MAC address and will effectively be treated as the
same group. This means that both groups will be forwarded by the device and potentially overloading the intended receiver. See RFC 1112,
http://tools.ietf.org/html/rfc1112, for details on how IP multicast
groups map to MAC multicast addresses.
20.1.1
IGMP Snooping
The switch is capable of efficiently distributing IP(v4) multicast traffic on LAN
interfaces by means of IGMP snooping. IGMP Snooping is enabled by default per
VLAN, see section 15.1.5.
ˆ With IGMP snooping enabled on a VLAN, IP multicast packets are only forwarded to ports leading to a subscriber of that IP multicast group, and to
ports leading to an IP multicast router
ˆ With IGMP snooping disabled on a VLAN, multicast traffic is forwarded on all
ports in that VLAN, i.e., like broadcast traffic
1 Special restriction for DDW-x42 and DDW-x42-485: On these products the MAC address
database can hold at most 1000 addresses in total (unicast and multicast MAC). Thus, the upper limit for multicast addresses possible to keep track of is roughly 1000.
462
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
ˆ Ports shared between multiple VLANs may have different IGMP snooping
settings on different VLANs, i.e., one VLAN may have IGMP snooping enabled
and another may have it disabled. The disabled mode takes precedence on
such ports, i.e., multicast will be flooded on ports where at least one VLAN
has IGMP Snooping disabled
As part of the IGMP snooping functionality, the switch can also act as an IGMP
Querier, and settings for querier mode, and query interval are provided.
Querier mode: By default the switch has auto mode enabled. It relies on the
standard IGMP protocol to elect a designated IGMP querier on each LAN2 .
With auto mode unknown multicast is flooded to the elected querier, which
acts as a distribution point to the rest of the network. Keep this in mind
when designing your network, for many use-cases it is valuable information.
The forced querier mode is a non-standard setting specific to WeOS that in
some cases can be more fault tolerant. When all switches on a LAN are set
to this mode they will all discard the election mechanism of the protocol and
always send queries. This not only incurs a notable penalty on the network
due to the overhead of IGMP messages flooding the LAN, but it also causes
all unknown multicast to be flooded to all switches. This makes the network
less vulnerable to the loss of one querier and all multicast is always available
to end devices. It is however not recommended due to the broadcast like
effects it causes.
In proxy mode, the switch normally only acts as a silent forwarder of IGMP
queries (and reports) between the IGMP querier and end devices. However,
to prevent loss of multicast traffic in the case when there exist no elected
IGMP querier on a LAN, the switch will initiate queries with the source IP
address 0.0.0.03 . This feature of proxy mode can be used to optimise lowbandwidth setups, see section 20.1.4 for more information.
On VLANs where the network interface is not assigned an IP address, the
switch will automatically fall back to proxy mode, regardless of the querier
mode setting.
Query interval: The switch can be configured to send out queries on intervals
12, 30, 70 and 150 seconds, default 12 sec. This interval is also used when
timing out multicast to end devices that for some reason stop answering the
2 The querier with the lowest IP address on each LAN is elected. Usually the gateway or multicast
router.
3 Address 0.0.0.0 is a special case and is never part of the IGMP querier election process, as
clearly stated in the standard.
© 2017 Westermo Teleindustri AB
463
Westermo OS Management Guide
Version 4.21.1-0
queries.
Multicast router timeout: When a multicast router, or a switch acting as IGMP
querier, goes down, the lack of IGMP Query messages will cause a reelection
to establish a new IGMP querier. This timeout can be configured via the CLI
”multicast-router-timeout” setting. Default: 300 sec.
When a multicast receiver attached to a switch port leaves a multicast group (i.e.,
stops subscribing to an IP multicast address or is simply disconnected from the
port), the IGMP snooping leave latency (the time until the switch stops forwarding
the associated multicast data) is within 2-3 times the configured Query Interval.
20.1.2
Multicast Router Ports
When IGMP snooping is enabled, the switch will learn on which ports there are
interested receivers of a certain multicast group. It accomplishes this by listening
to IGMP Report messages sent by all subscribers. Thus, the switch only forwards
multicast on ports leading to members of each specific multicast group.
The switch also forwards all multicast traffic, both subscribed (known) and unknown, on ports leading to multicast routers. The following ports are considered
as multicast router ports:
ˆ Ports configured as multicast router ports
ˆ Ports where IGMP Queries are received, usually queries are sent by multicast
routers, but also by IGMP snooping aware switches like WeOS
ˆ FRNT Ring Coupling ports and Multi-link Dual-Homing ports: To provide fast
fail-over of multicast traffic, FRNT Ring Coupling and Multi-link Dual-Homing
uplinks (see chapter 17) are added to the list of multicast router ports. This
is both done at the Ring Coupling nodes and Dual-Homing nodes, as well as
on switches on the remote side of the uplink4 .
FRNT ring ports are no longer considered multicast router ports. The Fast Reconnect feature of FRNT is instead handled per multicast group: if a multicast
receiver is located on a ring port, the other ring port is automatically added to
the ATU MAC filter5 . In case of ring breakage this practice ensures an extremely
low reconfiguration time for multicast over FRNT.
4 An exception is when connecting a Dual-Homing uplink to a non-FRNT switch, the fail-over of
multicast traffic will instead occur on the next reception of an IGMP Report (if IGMP snooping is
enabled). See also section 17.1.2.1.
5 This can be seen using the CLI command ”show fdb”
464
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
20.1.3
IGMP Fast Leave
WeOS IGMP snooping supports IGMP Leave by default and Fast Leave can be
enabled on a per-port basis. The CLI ”igmp-fast-leave-ports” setting allows
using the keyword ”all”, but Fast Leave is recommended only for access ports.
Example
example:/#> configure
example:/config/#> ip
example:/config/ip/#> no igmp-fast-leave-ports
example:/config/ip/#> igmp-fast-leave-ports eth 3,6
example:/config/ip/#> leave
Configuration activated. Remember "copy run start" to save to flash (NVRAM).
example:/#> copy run start
example:/#> show ip igmp
Static Multicast ports
------------------------------------------------------------------------------Static router ports
: --Dual homing/Coupling ports : --FRNT ports
: --VID Querier IP
Querier MAC
Port Interval Timeout
------------------------------------------------------------------------------1 0.0.0.0
LOCAL
VID Multicast Group Filtered MAC Addr Active ports
------------------------------------------------------------------------------1 239.255.255.250 01:00:5E:7F:FF:FA 6
1 224.0.0.251
01:00:5E:00:00:FB 3, 6
1 225.1.2.3
01:00:5E:01:02:03 6
------------------------------------------------------------------------------Total: 3 filters, max 2048, in 1 VLAN.
example:/#>
When an IGMP Leave is received on a port configured with Fast Leave it will issue
a group specific query for the group being left and then immediately cut the
multicast stream for that (multicast MAC) group. With Fast Leave disabled WeOS
honors a grace period of, at most, two query intervals for the benefit of multicast
receivers attached on downstream port splitters (hubs or unmanaged switches).
When no membership report/reply is received the multicast group will time-out
within three query intervals.
© 2017 Westermo Teleindustri AB
465
Westermo OS Management Guide
Version 4.21.1-0
20.1.4
Low Bandwidth Networks
In low-bandwidth topologies, like FRNT over an SHDSL ring, you typically cannot
afford wasting bandwidth on unwanted traffic. With the IGMP Proxy Mode and
Fast Leave settings for IGMP snooping this can be avoided.
In the standard auto mode of IGMP all unknown multicast must be forwarded to
the elected querier. But if there is no elected querier, or if all switches instead
have proxy mode enabled, unknown multicast will be stopped before entering
the low-bandwidth ring.
Only when a subscriber appears will the traffic be classified as known and forwarded on the ring to the receiver. By also enabling Fast Leave, on the access
port towards the receiver, the multicast overhead can be kept to a near minimum.
466
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
20.2
Managing IGMP in the Web Interface
Global Configuration
Menu path: Configuration ⇒ IGMP
When entering the IGMP configuration page you will be presented with the global
settings for IGMP snooping. Enabling or disabling IGMP is done per VLAN, see
Section 15.
Querier Mode
Query Interval
The IGMP querier mode should have the same
setting across all devices in the same LAN:
Automatic: Automatic querier election. Recommended default
Querier: Forced Querier mode, the device always sends IGMP queries, every Query Interval seconds
Proxy: Fallback mode in which the switch normally does not initiate queries by itself,
only forwards queries and reports.
For more information on the modes, see Section 20.1.1
Number of seconds between each query.
© 2017 Westermo Teleindustri AB
467
Westermo OS Management Guide
Version 4.21.1-0
Fast Leave Ports
Multicast Router Ports
Ports where multicast should not linger when
receiving IGMP Leave.
Ports on which to forward all multicast. Useful if the switch fails to automatically detect a
multicast router, or when you have a non-IGMP
aware end devices.
Click Apply to save and apply the changes.
IGMP Status
Menu path: Configuration ⇒ IGMP Status
468
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
20.3
Managing IGMP in the CLI
The available general IP settings and monitoring commands are shown below.
Command
Configure General IGMP Snooping settings
ip
[no] igmp-mode <auto|querier|proxy>
[no] igmp-interval <12|30|70|150>
[no] igmp-fast-leave-ports <PORTLIST>
[no] multicast-router-ports <PORTLIST>
[no] multicast-router-timeout <1-2147483647>
Per VLAN IGMP Snooping settings
vlan <VID>
[no] igmp
Show IGMP Snooping Status
show ip igmp
20.3.1
Default
Section
auto
12 sec
Disabled
Disabled
300
Section
Section
Section
Section
Section
Section
Enabled
Section 15.4.6
Section 15.4.13
21.7.1
20.3.1
20.3.2
20.3.3
20.3.4
20.3.5
Section 20.3.6
IGMP Querier Mode
Syntax [no] igmp-mode <auto|querier|proxy>
Context IP Configuration context
Usage Set IGMP Querier mode. In auto mode the device will participate in the
querier election process (querier with lowest IP becomes querier). In forced
querier mode the device will send IGMP queries even if there are other
querier present with lower IP address. In proxy mode the device will act
as an IGMP proxy, only initiating queries when no other eligible querier is
available.
Note: if there is no IP address configured for an interface, the device will
fall back to proxy mode regardless of the mode setting.
”no igmp-mode” resets the IGMP Querier mode to the default setting (”auto”).
Use ”show igmp-mode” to view configured IGMP Querier mode (”auto”,
”querier” or ”proxy”).
© 2017 Westermo Teleindustri AB
469
Westermo OS Management Guide
Version 4.21.1-0
Default auto
20.3.2
IGMP Querier Interval
Syntax [no] igmp-interval <12|30|70|150>
Context IP Configuration context
Usage Set IGMP Querier interval (seconds). The same interval is used for all
interfaces.
”no igmp-interval” resets the IGMP Querier interval to the default setting,
”12” sec.
Use ”show igmp-interval” to view configured IGMP Querier interval.
Default 12 (sec)
20.3.3
IGMP Fast Leave
Syntax [no] igmp-fast-leave-ports <PORTLIST>
Context IP Configuration context
Usage Add or remove IGMP Fast Leave ports. For details, see section 20.1.3
”no igmp-fast-leave-ports <PORTLIST>” removes the specified port(s)
and ”no igmp-fast-leave-ports” all ports from the list of IGMP Fast Leave
ports.
Use ”show igmp-fast-leave-ports” to view configured multicast router
ports.
Default Disabled
A ”PORTLIST” is a comma separated list of port ranges without intermediate
spaces, e.g., ”1/1-1/3,2/3”.
20.3.4
Static Multicast Router Port Settings
Syntax [no] multicast-router-ports <PORTLIST>
Context IP Configuration context
470
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Usage Add or remove multicast router ports. All (layer-2) multicast traffic will be
forwarded on multicast router ports, see section 20.1.1.
”no multicast-router-port <PORTLIST>” removes the specified port(s)
and ”no multicast-router-port” all ports from the list of multicast router
ports.
Use ”show multicast-router-port” to view configured multicast router
ports.
Default Disabled
A ”PORTLIST” is a comma separated list of port ranges without intermediate
spaces, e.g., ”1/1-1/3,2/3”.
20.3.5
Multicast Router Timeout
Syntax [no] multicast-router-timeout <1-2147483647>
Context IP Configuration context
Usage Set the ”other IGMP Querier present” timeout (sec). The same interval is
used for all interfaces.
Timeout for learned multicast router ports. With IGMP, and IGMP Snooping
for switches, the elected querier is a critical component for successful operation. If it is lost, or suddenly gets a new IP address, another device must
take over. This timeout adjusts the timeout before this device can take over.
”no multicast-router-timeout” resets the ”other IGMP Querier present”
timeout to the default setting (”300”).
Use ”show multicast-router-timeout” to view configured ”other IGMP
Querier present” timeout.
The timeout should never be set lower than the IGMP Query Interval!
Default 300 (sec)
20.3.6
Show IGMP Snooping Status Information
Syntax show ip igmp
Context Admin Exec context
© 2017 Westermo Teleindustri AB
471
Westermo OS Management Guide
Version 4.21.1-0
Usage Show IGMP snooping status information.
Default N/A
472
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Chapter 21
General Interface and Network
Settings
This chapter presents WeOS network interface settings, such as the interface IP
address and common IP network settings, e.g., default gateway, DNS server and
NTP server. Topics specific to various routing protocols and services, e.g., RIP,
OSPF, VRRP, etc. are left to chapters 26-31.
Section 21.1 presents the general concepts of network interfaces in WeOS. It also
covers the notion of interface admin distance and management interface, as well
as IP related settings for DNS, NTP, etc. Section 21.4 and section 21.5 cover
management of interfaces and general network settings via the Web interface.
The corresponding CLI settings are divided into section 21.6, interface settings,
and section 21.7, general network settings.
21.1
Overview
The table below summarises general interface and network features. Sections 21.221.3 contain further information on specific interface and network features.
Feature
Interface settings
Enable/disable interface
MAC address
© 2017 Westermo Teleindustri AB
Web
X
CLI
Description
X
Section 21.2.1
X
Section 21.2.4
Continued on next page
473
Westermo OS Management Guide
Version 4.21.1-0
Feature
Primary IP address
Secondary IP addresses
Netmask (Prefix Length)
MTU
Interface admin distance
Management interface
ICMP Redirect (sending)
DHCP client: client ID
DHCP client: vendor class ID
DHCP client: link-local address
View interface configuration
View interface status
General network settings
Default gateway
Enable/disable unicast routing
DNS client support
Set DNS server
Dynamic DNS
DNS search path
DNS proxy server support
NTP (NTP client)
View general network config.
View general network status
21.2
Continued from previous page
Web CLI Description
X
X
Section 21.2.5
X
X
Section 21.2.5
X
X
Section 21.2.5
X
X
X
X
Section 21.2.6
X
X
Section 21.2.7
X
Section 21.2.8
X
X
Section 21.2.9
X
X
Section 21.2.10
X
Section 21.2.5
X
X
X
X
X
X
X
X
Section 21.3.1
”
X
X
X
X
X
X
X
X
X
Section 21.3.3
”
”
Section 21.3.4
Section 21.3.2
X
X
X
Network interfaces
WeOS supports several kinds of network interfaces:
ˆ LAN/VLAN network interfaces: A network interface is created for every VLAN
configured on the switch (chapter 15).
474
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
ˆ PPP network interfaces: (only for WeOS Extended) A network interface is
created for every PPP instance configured on the switch (chapter 33). As of
WeOS v4.21.1, PPP support is available over Ethernet/DSL ports using PPP
over Ethernet (PPPoE), and over serial ports with or without external modem.
ˆ Loopback network interface: The loopback interface lo is a logical network
interface, which is always present. Its primary IP address cannot be changed,
but it is possible to add secondary IP addresses, which can be useful in some
situations,e.g., for OSPF (chapter 27).
ˆ GRE interfaces: (only for WeOS Extended) For every configured GRE tunnel
(chapter 34), an associated GRE network interface is created.
ˆ SSL interfaces: (only for WeOS Extended) For every configured SSL VPN
tunnel (chapter 36), an associated SSL network interface is created.
ˆ Blackhole interface:
WeOS has a hidden blackhole interface (”null0”),
which can be used to avoid routing loops in case of incomplete subnetting,
or to avoid that VPN traffic is forwarded towards the default gateway when
the VPN tunnel is down. See section 26.1.4.3.
Fig. 21.1 shows how VLAN interfaces (vlan1-vlan4) are mapped to VLANs and
ports, i.e., Ethernet and DSL ports. When using PPPoE, a PPP interface is created on top of a VLAN interface (see pppoe0 and vlan4 in fig. 21.1). modem0
represents the network interface when running PPP over a serial port. The GRE
and loopback interfaces are logical interfaces not directly associated with any
physical port.
Every network interface can be assigned an IP(v4) address and netmask. By
assigning an IP address to an interface, the operator is able to remotely manage
the switch via that interface. Furthermore, if routing (IP forwarding) is enabled,
the switch is able to forward packets between network interfaces. Section 21.3
gives a brief overview of WeOS routing features. chapter 26 gives a more detailed
introduction to WeOS routing support, while chapters 27 and 28 covers dynamic
routing with OSPF and RIP.
Note
IP forwarding is not available for products running software level WeOS Standard. However, it is possible to configure static (unicast) routes in WeOS
Standard products as described in sections 26.2.1 (Web) and 21.7.3 (CLI).
© 2017 Westermo Teleindustri AB
475
Westermo OS Management Guide
Version 4.21.1-0
Router
WeOS Layer 2/3 Switch
Routing
modem0
vlan1
1
1
2
Network
Interfaces
(Routing)
VLAN 3
VLAN 4
VLANs
(Switching)
6
9
Ethernet/DSL
Ports (1−10)
vlan2
VLAN 1
Serial
Port(1)
pppoe0
(vlan4)
vlan3
VLAN 2
3
4
5
7
8
10
modem0
pppoe0
vlan1 vlan2 vlan3 (vlan4)
Switch
VLAN 1
1 2 3
Serial
a)
Switch
VLAN 2
4 5 6
Switch
VLAN 3
6 7 8 9
Network
Interfaces
Switch
VLAN 4
10
Ethernet/DSL
b)
Figure 21.1: A network interface is associated with each VLAN, and VLANs are in
turn associated with Ethernet (or DSL) ports as shown in figure a). Furthermore,
when using PPPoE, a PPP network interface will be created and mapped on top
of an associated VLAN interface, see pppoe0 and vlan4. The routing switch can
conceptually be seen as a router connecting a set of switches, as shown in figure
b). In this sample setup, port 6 is shared by VLANs 2 and 3 (by use of VLAN
tagging).
21.2.1
Interface Operational Status (up/down)
For a network interface to get operational status up, it must be enabled in the
configuration. But for some types of interfaces there may be additional criteria
to reach interface (operational) status up, as shown in the list below:
ˆ Loopback network interface: The loopback interface lo is always up.
ˆ LAN/VLAN network interfaces: For a VLAN interface to get status up, the
interface must be enabled and its associated VLAN must also be up. In
turn, the associated VLAN is up when that VLAN is enabled, and any of its
associated ports have link up status. See chapter 15 for more information
on VLANs.
Note
It is possible to circumvent the link status propagation property by
configuring a LAN/VLAN network interface as always up (”enable
always”, see section 21.6.2). Disabling link status propagation may
significantly impact layer-3 protocols such as RIP, OSPF, VRRP, and
more -– the protocols will have to fall-back to other methods to detect link-down, e.g. hello message timeout and similar. Do not use the
”enable always” setting unless you really know what you are doing.
476
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
ˆ PPP network interfaces: (only for WeOS Extended) For a PPP interface to
get status up, the PPP interface (and the associated PPP instance) must be
enabled and successfully have carried out the PPP handshaking, including
PPP authentication and IP address negotiation. For PPPoE, this implies that
the underlying VLAN interface must also be up. See chapter 33 for more
information on PPP.
ˆ GRE interfaces: (only for WeOS Extended) For a GRE interface to get status
up, the GRE interface (and the associated GRE tunnel instance) must be
enabled.
21.2.2
Interface Settings at Factory Default
WeOS products typically have all Ethernet and DSL ports mapped to VLAN 1 by
factory default, and the network interface associated with VLAN 1 is named vlan1.
The exception is Falcon, as described later in this section.hus by factory default,
a WeOS unit has network interfaces vlan1 and lo (logical ”loopback” interface).
The factory default settings for interfaces vlan1 and lo are presented below. Most
of the loopback settings are permanent (non-configurable).
Interface parameters
Administrative Mode
IP address
Netmask
Secondary IP addresses
Secondary Netmask
MAC address
MTU
TCP-MSS
Admin Distance
Management Interface
Factory Default Setting (General)
vlan1
lo
Enabled
Dynamic
(DHCP)
(Dynamic)
192.168.2.200
255.255.255.0
Auto
Auto (1500)
Disabled
1
Enabled1
Enabled
Static
127.0.0.1
255.0.0.0
Disabled
N/A
N/A
16436
Disabled
16
Disabled
The interface administrative distance and management interface concepts are
described in sections 21.2.6 and 21.2.7.
As stated earlier, Falcon has a different factory default settings than other WeOS
products. The Ethernet ports are all mapped to VLAN 1 and interface vlan1 as
© 2017 Westermo Teleindustri AB
477
Westermo OS Management Guide
Version 4.21.1-0
usual, but the Falcon xDSL port resides on a separate VLAN (VLAN 1006) and
interface (vlan1006). The factory default settings for the associated interfaces
are shown below. Most of the loopback interface (lo) settings are permanent
(non-configurable).
Interface parameters
Administrative Mode
IP address
Netmask
Secondary IP addresses
MAC address
MTU
TCP-MSS
Admin Distance
Management Interface
Factory Default Setting (Falcon)
vlan1
vlan1006
lo
Enabled
Static
192.168.2.200
255.255.255.0
Disabled
Auto
Auto (1500)
Disabled
16
Enabled1
Enabled
Dynamic
(DHCP)
N/A
Disabled
Auto
Auto (1500)
Disabled
1
Disabled
Enabled
Static
127.0.0.1
255.0.0.0
Disabled
N/A
16436
Disabled
16
Disabled
Note
On Falcon, the xDSL port associated with VLAN 1006 is intended to be used
as the upstream ”WAN” port for Internet access. Interface vlan1006 inherits
its admin distance from the base interface, which by default is 1. For security
reasons, management services are filtered out on vlan1006 by default.
21.2.3
Creating Additional Network Interfaces
As shown in fig. 21.1 the switch will have one network interface for every VLAN
defined on the switch. Thus, additional VLAN network interfaces can be created
by creating new VLANs (see chapter 15). Similarly, a PPP network interface is
created for every configured PPP instance, a GRE network interface is created for
every configured GRE instance, etc.
The default settings for new VLAN and PPP (PPPoE and PPP over serial/modem)
interfaces are shown in the table below, followed by a table presenting default
settings for GRE and SSL VPN interfaces (PPP, GRE and SSL VPN interfaces are
available for products running software level WeOS Extended).
1 At
478
factory default, all management services except Telnet are allowed on interface vlan1.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
It is not possible to create additional loopback interfaces. To have additional
loopback IP addresses you can instead configure secondary IP addresses on the
lo interface.
Interface
Parameters
Administrative Mode
IP address
Netmask
MAC address
MTU
Admin Distance
TCP-MSS
Management Interface
vlan<VID>
Enabled
Static1
Disabled
Disabled
Auto
Auto (1500)
16
Disabled
Enabled5
Interface
Parameters
Administrative Mode
IP address
Netmask
MAC address
MTU
Admin Distance
TCP-MSS
Management Interface
Default Setting
pppoe<ID> modem<ID>2
Enabled
Dynamic3
(IPCP)
N/A
N/A
14924
”Inherited”
1412
”Inherited”
Enabled
Dynamic3
(IPCP)
N/A
N/A
Auto (1500)
16
Disabled
Enabled4
Default Setting
gre<ID> ssl<ID>
Enabled
Static
Disabled
Disabled
N/A
1476
16
Disabled
Enabled5
Enabled
Static
Disabled
Disabled
Auto6
Auto (1500)
16
Disabled
Enabled5
The interface admin distance and management interface concepts are described
in sections 21.2.6 and 21.2.7.
1 The
exception is interface vlan1 (VID 1). If vlan1 does not exist, or if it is created without an
address method defined, vlan1 will default to acquire its address dynamically via DHCP.
2 Interfaces for PPP over serial port (modem<ID>) are only available for products equipped a
serial port.
3 For PPP interfaces, the IP address assignment is handled by the PPP configuration, see section 33.1.7.
4 When using PPPoE the default PPP interface MTU is 8 bytes less than the associated VLAN
interface MTU, which is typically 1500 bytes.
5 On new interfaces, all management services except Telnet are allowed by default.
6 Only layer-2 SSL interfaces have MAC addresses. As of WeOS v4.21.1 the auto mode picks a
random MAC address, however, this may change in the future WeOS releases.
© 2017 Westermo Teleindustri AB
479
Westermo OS Management Guide
Version 4.21.1-0
VLAN network interfaces will be named according to the associated VLAN ID, e.g.,
the interface of VLAN 100 will be named vlan100. PPP, GRE and SSL interfaces will
be named according to their associated instance ID, e.g., pppoe0 is the interface
of PPPoE instance ”0”, modem0 is the interface of serial/modem instance ”0”,
and so on.
To communicate with the switch via a newly created interface, an IP address must
be assigned to the interface, see section 21.2.5.
When creating a PPP instance of type PPPoE, the admin distance and management interface properties of the associated VLAN network interface are inherited
by the PPP interface. This inheritance does not work in the reverse direction
though, i.e., if the PPP instance is removed, the management and admin distance properties of the PPP interface are not passed back to the associated VLAN
interface.
Note
With PPPoE, one must specify which VLAN interface to run PPPoE over, e.g.,
see interface vlan4 in fig. 21.1. The resulting PPP interface will be said to
”own” the associated VLAN interface. As of WeOS v4.21.1, it is not possible
to access a switch via ”owned” VLAN interfaces — access is only possible
via the PPP interface.
21.2.4
VLAN Interface MAC address
Each VLAN network interface will be assigned a MAC address (also known as
the Ethernet address, the link address, the hardware address, or the IEEE EUI-48
address).
In WeOS products, each Ethernet port (or DSL port) is assigned a MAC address,
and a VLAN interface will by default inherit its MAC address from one of its member ports. It is also possible to manually configure a MAC address for a VLAN
interface, see section 21.6.13.
The algorithm to assign VLAN interface MAC address uses the following preference order:
1. If the interface has been configured with a custom MAC address, use that
address as the interface MAC address.
2. If the VLAN has one or more ports assigned untagged, use the MAC address
of the ”lowest” untagged port as the interface MAC address.
480
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
3. If the port has one or more ports assigned tagged, use the MAC address of
the ”lowest” tagged port as the interface MAC address.
4. Use the MAC address of the channel (section 15.1.6) associated with the
VLAN.
Consider the sample configuration in fig. 21.1. When all interfaces get their MAC
address automatically, interface vlan1 inherits the MAC address of port 1, vlan2
inherits its MAC from port 4, vlan3 from port 7 (assuming port 6 is tagged on
VLAN 3), and interface vlan4 from port 10.
Note
For the automatic MAC assignment methods (steps 2-4 above), the MAC address may change when the set of ports associated with the VLAN changes.
When this happens, the WeOS device will submit a gratuitous ARP to update
stale ARP caches in neighbour nodes.
For VLANs created dynamically (section 15.1.7), no associated network interface
is created. Thus, for such VLANs no interface MAC address is needed.
21.2.5
IP address settings
Each network interface can be assigned a primary IP address and up to 8 secondary IP addresses, this is sometimes referred to as multinetting, but can also
be another address on the same subnet as the primary address. The primary IP
address can either be statically or dynamically assigned, depending on the address method configured for the interface (”inet static” or ”inet dynamic”).
The secondary IP addresses can only be statically configured, but can be used
with both static and dynamic primary address.
Options for configuring the primary address for different interface types:
ˆ VLAN interfaces: The primary IP address of a VLAN interface can be configured statically, or configured to acquire its address dynamically (DHCP). It is
also possible to have a VLAN interface without any IP address.
ˆ PPP interfaces: For PPP interfaces the address setting is set to dynamic, but
the actual IP address assignment is handled by the PPP configuration (IPCP),
see section 33.1.7.
ˆ GRE interfaces: For GRE interfaces, the primary IP address can only be configured statically.
© 2017 Westermo Teleindustri AB
481
Westermo OS Management Guide
Version 4.21.1-0
ˆ Loopback interface (lo): The primary IP address of the loopback interface
(lo) is permanently set to 127.0.0.1.
In the example below, interface vlan2 is assigned a static primary IP address
(”192.168.11.1”) and an additional secondary IP address (”192.168.12.1”),
i.e., multinetting is used. Here the IP address netmask (255.255.255.0) for both
addresses has been written in prefix length format (’/24’).
Example
example:/config/#> interface vlan2
example:/config/iface-vlan2/#> inet static
example:/config/iface-vlan2/#> address 192.168.11.1/24
example:/config/iface-vlan2/#> address 192.168.12.1/24 secondary
example:/config/iface-vlan2/#> end
example:/config/#>
Interfaces with dynamic address assignment use DHCP to acquire their IP address
from a DHCP server, or IPCP for PPP interfaces. If no DHCP server is present, the
interface will generally end up without any IP address. The exception is the interface with best admin distance, which will also acquire a link-local IP address1 . The
interface admin distance and link-local address concepts are further described in
section 21.2.6.
21.2.6
Dynamic Address Assignment and Admin Distance
An interface can be configured to retrieve its IP settings dynamically via DHCP
(VLAN interfaces) or IPCP (PPP interfaces). In addition to interface settings such
as IP address and netmask, the switch can also acquire general network settings
such as default gateway and DNS server(s) from the DHCP server, or via PPP.
More information on general network settings is given in section 21.3.
Multiple network interfaces can acquire their IP settings dynamically, but only
one default route, one set of DNS servers, one domain search path and one set
of NTP servers can be active at one time in the system. WeOS handles this using
a set of precedence rules. When setting up a device with automatic fail-over
between multiple upstream connections these rules are important to be aware
of.
Prior to WeOS 4.14.0 the precedence was handled by something called the primary interface. However, this has been replaced with the concept of administrative distance for both static routes and interfaces. Administrative distance is also
1 Assignment
482
of link-local address can be disabled, see section 21.6.12.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
available to dynamic routing protocols such as OSPF and RIP, see chapters 27
and 28, respectively.
The admin distance is a priority value ranging from 1–255, where 255 is treated
as infinite distance. E.g., a static route installed with distance 255 is guaranteed
to never be activated. WeOS makes use of this in fail-over scenarios with multiple
upstream interfaces and ping triggers.
The following list summarises the rules for dynamically retrieved settings and
how they are applied to the system.
ˆ Dynamic IP address and netmask are always set on the interface, without
affecting any secondary IP address configured statically.
ˆ Default route, domain search path, and DNS servers are always saved, but
not necessarily installed.
ˆ Default routes are installed with the configured interface admin distance
and the ’best’ route is set as the active default route in the system.
ˆ The interface with the best (lowest) distance wins. If that interface goes
down, the default route of the next best interface distance is activated.
ˆ If there are multiple interfaces with lowest distance, the system will select
one of those interface as ’best’. A user wishing to have full control of what
interface is ’best’ should assign a unique admin distance per interface.
Hint
Assign unique admin distance values to your interfaces.
ˆ A ping trigger can be associated with the interface distance setting. When
it signals loss of connectivity, the distance of the associated default route is
raised to infinity (255).
ˆ When the best upstream interface has been established, domain search
path, domain name servers (DNS) and network time protocol servers (NTP)
are set from that source, unless there exist statically configured settings.
ˆ Statically configured DNS, domain and NTP always win, regardless of any
distance.
ˆ NTP server may be acquired from a DHCP server when no NTP server has
been configured statically (see section 21.3.2).
© 2017 Westermo Teleindustri AB
483
Westermo OS Management Guide
Version 4.21.1-0
ˆ The ’primary’ setting in WeOS prior to 4.14.0 is converted to a distance
value: the primary interface gets distance 1 and all other interfaces get the
default distance, 16.
ˆ Static configuration of routes, including the default route, competes with
routes learned on DHCP client interfaces as well as routes from dynamic
routing protocols. An obvious benefit of this is to have a statically configured
fallback default route that is activated automatically when no better route
is available. This is often referred to as a floating static route.
ˆ The default gateway setting ”ip default-gateway <IPADDR>” is deprecated. Setting up a default gateway in the CLI will install a static default
route with distance 1. Use the route command instead (see section 21.7.3,
notice the new keyword ’default’ for ’0.0.0.0/0’
Example
example:/config/#> ip route default 192.168.11.1 10
In the example below interface vlan3 is configured to acquire its IP address via
DHCP with distance 1. The system default interface, vlan1 is moved to distance
200 and a floating static route to a gateway reachable via vlan1 is setup with
a distance of 200 as well. The default route acquired by DHCP on vlan3 will be
installed with distance 1 and will be made the active route.
Example
example:/config/#> interface vlan1
example:/config/iface-vlan1/#> address 192.168.11.2/24
example:/config/iface-vlan1/#> distance 200
example:/config/iface-vlan1/#> end
example:/config/#> interface vlan3
example:/config/iface-vlan3/#> inet dhcp
example:/config/iface-vlan3/#> distance 1
example:/config/iface-vlan3/#> end
example:/config/#> ip default 192.168.11.1 200
example:/config/#>
If no DHCP server is present, an interface configured to use DHCP client for address assignment will end up without any IP address. The exception is the DHCP
client interface with the best distance, which will also acquire a link-local IP address in the range 169.254.0.0/16 in addition to any address assigned via DHCP2 .
The link-local address is taken from the 169.254.0.0/16 range such that address
2 Assignment
484
of link-local address can be disabled, see section 21.6.12.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
collisions are avoided and that an interface is likely to get the same address every
time it comes up.
21.2.7
Management Interface
The operator can manage the switch remotely in several ways: Web (HTTP/HTTPS),
SSH, Telnet, SNMP and WeConfig (using the IPConfig service). As described in
chapter 7 it is possible to completely disable individual management services,
however, there are situations when an operator may wish to limit management
access to a certain network interface or VLAN. WeOS provides a powerful mechanism for controlling access to management services on a per interface basis. An
interface where one or more management services are enabled is referred to as
a management interface.
Interface vlan1:
management ssh http https ipconfig snmp
Interface vlan2:
no management
Interface vlan3:
management ssh
Internet
WeOS Switch/Router
1 2 3 4 5 6 7 8
VLAN 3
VLAN 1
VLAN 2
Figure 21.2: Management service filtering per interface.
Fig. 21.2 gives an example on the flexibility by the management interface feature
in WeOS. The switch has three network interfaces – one for each VLAN. VLAN
1 is the administrator’s local LAN with full management capabilities. VLAN 2
is another local LAN for regular in-house users, from which no management is
allowed. VLAN 3 is used as the upstream connection; in this example SSH is
allowed on this network interface, while other services are disabled.
© 2017 Westermo Teleindustri AB
485
Westermo OS Management Guide
Version 4.21.1-0
Note
WeOS use the term ”management interface” rather than ”management
VLAN”. This is because management is not be limited only to VLAN network interfaces. For example, the operator may wish to manage a switch
remotely through a modem connection (i.e., a PPP interface on a switch
equipped with a serial port).
The equivalent of a management VLAN can be setup by filtering out management services on all interfaces but the network interfaces associated
with that VLAN.
The default behaviour aims to avoid unintentional loss of management access
to the switch. Sections 21.2.2 and 21.2.3 describe the default settings for network interfaces, settings at factory default as well as settings for newly created
interfaces3 .
Warning
Access to management services on all interfaces is convenient, but may
pose a security risk if connected to an untrusted network. By default the
device is (typically) manageable via all network interfaces, it is therefore
strongly recommended that the operator use the interface management filter to only allow a select set of services, or none, on untrusted networks.
E.g., for an interface connected to the public Internet one should consider
disallowing all management services, or perhaps only allow management
via secure protocols such as SSH and HTTPS.
Also crucial to cyber security is the password policy and setting up adequately secure passwords when providing management access via an interface connected to an untrusted/public network.
A word of caution is in order, it is entirely possible to get locked out of a device
when setting up the management service filter. For devices with a console port
this may not be a problem, for others this is the time to be reminded about the
”crossed–cables factory reset” (section 7.1.3.3).
However, WeOS actually does implement some safeguards to prevent against
locking yourself out. If all management is disabled on all interfaces, the system
falls back to enabling secure shell, SSH, access on interface vlan1. Furthermore,
if Web (for instance) is the only management service allowed on any interface,
3 As mentioned in section 21.2.2 factory default on Falcon switches include a separate VLAN for
the xDSL port, and the associated interface (vlan1006) has management services disallowed for
security purposes.
486
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
but the Web server has been disabled, the same fall-back solution is triggered.
Hint
From security standpoint it is recommended to separate the management
interface from the upstream WAN interfaces, but also from interface vlan1
since it is also the fallback interface in WeOS.
E.g., use interface vlan1 as a LAN interface, with high interface distance,
and interface vlan2 as the upstream WAN interface, with distance 1.
If you, e.g., remove the unrelated VLAN 3 without assigning its ports to any
other VLAN, then WeOS will automatically place them as untagged in VLAN
1, the default/fallback VLAN. In most cases you do not want those ports
ending up on the upstream side . . .
21.2.8
Control Sending of ICMP Redirect
A WeOS router is able to send ICMP Redirect messages when it receives IP packets
which could have been routed more optimal. The topology shown in fig. 21.3 can
be used to illustrate a situation where ICMP Redirect is useful.
Internet/Intranet
R1
192.168.1.0/24
H1
optimal
path
R2
192.168.2.0/24
H2
Figure 21.3: Example where ICMP Redirect is useful.
Assume that Host 1 (H1) wishes to communicate with Host 2, and that H1 (only)
knows about its local subnet (192.168.1.0/24) and its default route pointing to
© 2017 Westermo Teleindustri AB
487
Westermo OS Management Guide
Version 4.21.1-0
Router 1 (R1). In this case all packets from H1 to H2 will go to R1, which in turn
sends them back on the same LAN to R2. The packets will be sent twice over
the LAN, resulting in waste of network capacity and increased delay. By enabling
sending of ICMP Redirect on R1, the router will send ICMP Redirect messages to
H1, informing the host that it can route packets directly to R2. If the host accepts ICMP Redirect messages, it will update its routing table and forward future
packets to H2 directly via R2.
In WeOS, the sending of ICMP Redirect messages can be enabled/disabled per
network interface. By default sending ICMP Redirect messages is enabled.
Note
A WeOS unit does not accept incoming ICMP redirect messages.
21.2.9
DHCP client identifier
When configuring an interface to use DHCP for address assignment, a WeOS unit
will provide a DHCP client-identifier (DHCP option 61) in its communication with
the DHCP server. As DHCP servers may use this Client ID when assigning host
specific IP settings, understanding how the Client ID is formed on WeOS units can
be useful:
ˆ MAC based: By default the WeOS DHCP client forms the Client ID based on
the MAC address of the network interface the DHCP client runs on. It forms
the Client ID by concatenating ’01’ and the ’MAC address’[1].
For example, a DHCP client running on a network interface with MAC address ’00:07:7c:12:34:56’ will send a Client ID as the following sequence of
hexadecimal numbers: ’0100077c123456’
In WeOS each VLAN network interface is automatically assigned a MAC address according to the algorithm specified in section 21.2.4. You may also
configure the MAC address manually.
488
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Note
In cases where several VLANs share the exact same port setup, the
automatic algorithm will assign the same MAC address to several VLAN
interfaces. This is usually harmless, but if you intend to let your WeOS
unit run a DHCP client on each of those VLAN interfaces you should
assign unique MAC addresses manually. Otherwise your DHCP server
may be confused when receiving different client requests reporting the
same MAC address.
ˆ Manual String: You can also configure the Client ID manually as an ASCII
text string. The client ID will then be sent as a sequence of hexadecimal
numbers, where each byte holds the ASCII value of the entered text string
(e.g., ’a’ is sent as hex ’61’, etc). You have the choice to prepend a ’00’ byte
to the string inline with [1], or to send the string as is (”raw string”).
The ASCII string can hold up 63 characters, and may contain the following
characters: a-z, A-Z, 0-9, hyphen (’-’) and period (’.’) The string may not
start or end with a hyphen (’-’) or period (’.’).
ˆ Hexadecimal Sequence: For more flexibility you can configure the Client ID
manually as a sequence of hexadecimal numbers (nibbles4 ). The sequence
should be an even number of nibbles up to 126 nibbles.
21.2.10
DHCP vendor class identifier
When configuring an interface to use DHCP for address assignment, a WeOS unit
will provide a DHCP vendor class identifier (DHCP option 60) in its communication
with the DHCP server.
By default, WeOS sends the software release identifier (e.g.,”WeOS v4.21.1”),
but it is also possible to configure the vendor class identifier manually as an
ASCII string (up to 63 characters) or hexadecimal sequence (up to 126 nibbles.)
4 A hexadecimal number can be in range ’0-f’ corresponding to decimal values 0-15. Each number is referred as a ’nibble’. Each nibble takes 4 bits, thus for every byte sent there are two
hexadecimal nibbles.
© 2017 Westermo Teleindustri AB
489
Westermo OS Management Guide
Version 4.21.1-0
21.3
General IP settings
The general IP settings provided fall into three categories:
ˆ Routing: Configuration of default gateway, static IP routes, and ability to
enable/disable IP forwarding (IP forwarding is available for products running
software level WeOS Extended).
ˆ IGMP: Configuration of IGMP snooping parameters such as querier mode,
query interval and static multicast router ports. (IGMP snooping is covered
in chapter 20.)
ˆ Services: Examples of include settings for DNS and DDNS servers, domain
search path, and NTP client settings.
21.3.1
Routing
To manage the WeOS unit remotely, it should generally be configured with a
default gateway. It is also possible to configure additional, static IP routes.
WeOS units running software level WeOS Extended are capable of IP forwarding,
i.e., it can route incoming IP packets to other interfaces and IP subnets. For
unicast, both static routing and dynamic routing (RIP and OSPF) are supported.
Units running WeOS Extended act as routers by default, i.e., IP forwarding is
enabled in the factory default setting.
WeOS units are also able to route IP multicast (static multicast routing). In addition, WeOS devices can efficiently distribute IP multicast packets in a switched
LAN by use of IGMP snooping.
This chapter only covers rudimentary routing features, such as enabling/disabling
IP forwarding and configuring a default gateway. WeOS routing support is described further in chapters 26-30. IGMP snooping support is covered in chapter 20.
21.3.2
Time synchronisation via NTP Server
The switch can synchronise its clock with an external time server via the NTP protocol. Up to 8 NTP servers can be configured, but it is also possible to acquire NTP
server(s) via DHCP when no static NTP server is configured (see section 21.2.6).
490
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
21.3.3
DNS client - setting DNS server and dynamic DNS
Most users find it is easier to refer to Internet hosts using domain names, e.g.,
http://www.example.com, than using IP addresses, e.g., http://93.184.216.
119. To facilitate the use of the Domain Name System (DNS), WeOS supports
configuration of up to two DNS server entries. It is also possible to configure a
domain search path. These settings can also be acquired dynamically via DHCP
or PPP (see section 21.2.6).
Use of domain names on a switch can be convenient, e.g., when setting up ping
triggers, VPN peers or when troubleshooting with tools such as ping or traceroute,
see section 7.1.9.
It is also convenient to communicate with the switch using domain names. When
the switch acquires its IP address dynamically (via DHCP or PPP), maintaining
the DNS server entry is cumbersome. To manage this situation, WeOS includes
support for dynamic DNS (DDNS). With DDNS enabled, the switch will update its
DNS server entry automatically when acquiring a new IP address.
Examples of supported DDNS providers are:
ˆ dyndns: http://www.dyndns.org,
ˆ freedns: http://freedns.afraid.org
ˆ no-ip: http://www.no-ip.com
See the CLI or Web online help for a more up-to-date list.
21.3.4
Proxy DNS server
WeOS units are able to act as DNS proxy servers (enabled by default). When
enabled, the unit will act as a DNS server and respond to DNS queries for known
hosts:
ˆ either statically added by the ”host” (section 21.7.9), see also the ”show
ip host” (section 21.7.29) command, or
ˆ hosts for which this unit acts as DHCP server (chapter 22), see also the
”show dhcp-clients” (section 22.3.22) command .
As DNS proxy, the WeOS the unit will also act as a caching DNS forwarder; DNS
queries of unknown hosts are forwarded to the unit’s own DNS server (see section 21.3.3 and the ”show ip name-server” command described in section 21.7.27),
© 2017 Westermo Teleindustri AB
491
Westermo OS Management Guide
Version 4.21.1-0
and the answer is cached for fast response of subsequent requests for the same
host. Domain specific overrides may be set as domain forwarding rules, see section 21.3.4.1.
When proxy DNS server is enabled on a WeOS unit, it will accept incoming DNS
packets on all its interfaces.
Hint
For security purposes you may wish to avoid accepting DNS packets on some
interfaces, e.g., your upstream interface towards the Internet. To block
such request you are recommended to configure appropriate deny filter
rules, e.g., ”filter deny in vlan1 dport 53 proto udp” and ”filter
deny in vlan1 dport 53 proto tcp” to block incoming DNS request on
interface vlan1. For more details on the WeOS firewall, see chapter 31.
Alternatively, disable the DNS proxy service.
For WeOS products running software level WeOS Standard attached directly
to the Internet, it is recommended to disable the DNS proxy service.
21.3.4.1
Conditional Domain Forwarding
The unit can also be configured to perform conditional forwarding based on the
domain name in the DNS query. By specifying a domain name and the IP(v4)
address of the DNS server to which queries should be forwarded, queries for that
domain and all its sub-domains will be forwarded to the specified DNS server.
Up to four rules may be specified. Multiple rules may have the same domain
name to provide DNS server redundancy. If a sub-domain rule is specified, the
sub-domain rule will have precedence over the domain rule.
For example, consider the following rules:
ˆ ”domain-forward example.net 10.10.10.10”
ˆ ”domain-forward sub1.example.net 20.20.20.20”
With these two rules, the unit will forward queries for www.sub1.example.net to
address 20.20.20.20, but queries for www.example.net to address 10.10.10.10.
Queries for domain names not matching any conditional domain rule will be forwarded to the default DNS server(s), see section 21.3.3.
492
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
21.4
Managing network interfaces via the web interface
This section covers network interface settings of the unit. Settings related to
IGMP snooping is described in section 20.2.
Menu path: Configuration ⇒ Network (IP) ⇒ Interface
Name
Enabled
Status
Distance
A unique identifier for the interface. Automatically generated
from VLAN/PPP/GRE/SSH identifier when the VLAN/PPP/GRE/SSH/ instance is created. lo is the loopback interface. (PPP, GRE and SSH
interfaces are available for WeOS Extended.)
Shows whether the interface is enabled or disabled. A green checkmark means the interface is enabled, and a dash means it is disabled.
The status of the interface, Up or Down.
The administrative distance value used for routes acquired on this
interface. Route selection is based on this number. A lower value
indicates a more preferred route.
Continued on next page
© 2017 Westermo Teleindustri AB
493
Westermo OS Management Guide
Version 4.21.1-0
Address
method
Address/
Netmask
Edit
Sort by
Continued from previous page
The IPv4 address assignment method used for the interface: Static
means the IPv4 address is configured manually, Dynamic means the
address is acquired automatically via DHCP (for VLAN interfaces) or
is part of the PPP configuration (for PPP interfaces), and Disabled
means IPv4 address assignment is disabled on the interface.
The IPv4 address, and its associated netmask, assigned to the interface. The netmask identifies what IP addresses are located on the
same subnet. Displays configured IP address, when address method
Static is used. Displays the dynamically assigned address, or Pending if Dynamic address method is set. Text Disabled is shown if IP
address assignment is disabled. Text Owned is shown when there is
a PPPoE interface associated with that VLAN interface. Secondary
addresses assigned to the interface are also listed.
Click this icon to edit the interface.
The list of interfaces may be sorted either in a default sort order,
or by the distance value. Select desired sort order and press apply
button.
When clicking the Edit icon for an interface you will be presented to its associated
edit page.
494
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Note: The user support to only display relevant input fields is only available when
using a JavaScript enabled browser.
MAC Address
Enabled
Distance
IP Address
Enabled
IP Address
Mode
Primary
Address
Secondary
Addresses
(Only applicable for VLAN interfaces.) The media access
control (MAC) address is used for controlling the communication on OSI layer 2. Shows the MAC-address associated to this interface.
The interface may be activated or deactivated by the Enabled setting. Click the check-box to activate/deactivate
the interface.
The administrative distance value used for routes acquired on this interface. Route selection is based on this
number. A lower value indicates a more preferred route.
(Only applicable for VLAN interfaces.) When disabling the
IP address, traffic may not be sent to the switch from
units connected to the VLAN associated with this interface. The address may be disabled to e.g. prevent administration access from specific VLANs. The IP address
mode field, and for static address mode the IP address
and netmask fields, will not be visible unless this box has
been checked.
Choose Static to manually configure IP address and netmask or Dynamic to let the unit query a DHCP server
for address information.(PPP interfaces can only be specified for dynamic IP address, but the actual IP address
assignment is handled by the PPP configuration, see section 33.2.)
The IPv4 address, and its associated netmask, assigned
to the interface. The netmask identifies what IP addresses
are located on the same subnet. Not applicable for PPP
and loopback interfaces. These fields will only be visible
if static IP Address Mode has been selected.
Address and netmask for the secondary IPv4-addresses
associated to this interface. These fields will only be visible if IP Address Enable has been checked. Up to eight
secondary IPv4-addresses may be associated to the interface. Click the plus sign to add new lines. Click the
to delete a row.
Continued on next page
© 2017 Westermo Teleindustri AB
495
Westermo OS Management Guide
Version 4.21.1-0
Client ID Type
(Only for
Address Method
”dynamic’)
Vendor Class
ID
(Only for
Address Method
”dynamic’)
MTU
496
Continued from previous page
Select the method to form the DHCP Client ID.
Auto
The Client ID will be ’01’+’MAC-Address of interface’.
Hex
Form Client ID manually from a hexadecimal
sequence. Limited to 1-63 octets.
String
Form Client ID manually from a text string.
Limited to 1-63 characters. Valid characters
are a-z, A-Z, 0-9, hyphen (’-’) and period (’.’).
The string shall not start or end with a hyphen
(’-’) or period (’.’). The Client ID is formed by
prefixing the string with ’00’.
Raw
Same as ’String’, except that no ’00’ prefix is
String
used.
For more information about Client ID, see section 21.2.9.
Select the method to form the DHCP Vendor Class ID.
Auto
The Vendor Class ID will be the Software release and number, (WeOS v4.21.1).
Hex
Form Vendor Class ID manually from a hexadecimal sequence. Limited to 1-63 octets.
String Form Vendor Class ID manually from a text
string. Limited to 1-63 characters. Valid characters are a-z, A-Z, 0-9, hyphen (’-’) and period
(’.’). The string shall not start or end with a
hyphen (’-’) or period (’.’).
For more information about Vendor Class ID, see section 21.2.10.
This option is not available for all interface types.
Override Set a non-default MTU size by entering
an override value.
Auto
The interface will let its MTU be the default MTU of the associated link type.
Continued on next page
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
TCP MSS
Management
Services
Continued from previous page
This option is not available for all interface types.
Override Limit TCP-MSS to the given number of
bytes.
Auto
Lets the TCP-MSS depend on the MTU
of the interface This will work fine
for typical TCP connections, but is not
likely to work over IPsec tunnels or
when additional IP header options are
in use.
Disabled Disables TCP-MSS clamping.
Check the boxes for the services that should be accessible from this interface.
Click the Apply button to save and apply the changes.
© 2017 Westermo Teleindustri AB
497
Westermo OS Management Guide
Version 4.21.1-0
21.4.1
Interface Status
Menu path: Status ⇒ Interface
Name
Enabled
Status
Distance
498
A unique identifier for the interface. Automatically generated
from VLAN/PPP/GRE/SSH identifier when the VLAN/PPP/GRE/SSH/
instance is created. lo is the loopback interface. (PPP, GRE and
SSH interfaces are available for WeOS Extended.)
Shows whether the interface is enabled or disabled. A green
check-mark means the interface is enabled, and a dash means
it is disabled.
The status of the interface, Up or Down.Text Owned is shown
when there is a PPPoE interface associated with the VLAN interface. The owner is also displayed within parenthesis.
The administrative distance value used for routes acquired on
this interface. Route selection is based on this number. A lower
value indicates a more preferred route.
Continued on next page
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Address
method
Address/
Netmask
Sort by
Continued from previous page
The IPv4 address assignment method used for the interface:
Static means the IPv4 address is configured manually, Dynamic
means the address is acquired automatically via DHCP (for VLAN
interfaces) or is part of the PPP configuration (for PPP interfaces),
and Disabled means IPv4 address assignment is disabled on the
interface.
The IPv4 address, and its associated netmask, assigned to the
interface. The netmask identifies what IP addresses are located
on the same subnet. Displays configured IP address, when address method Static is used. Displays the dynamically assigned
address, or Pending if Dynamic address method is set. Text Disabled is shown if IP address assignment is disabled. Secondary
addresses assigned to the interface are also listed.
The list of interfaces may be sorted either in a default sort order,
or by the distance value. Select desired sort order and press
apply button.
© 2017 Westermo Teleindustri AB
499
Westermo OS Management Guide
Version 4.21.1-0
21.5
Managing general IP settings via the web interface
This section covers general IP related settings of the unit. Settings related to
IGMP snooping are described in section 20.2.
21.5.1
Global Network Settings Overview
Menu path: Configuration ⇒ Network(IP) ⇒ Global settings
When entering the Network(IP) configuration page you will be presented to a list
of common network settings.
Global Settings (Default Gateway, Routing and DNS servers)
Configured
Default
Gateway
Active
Default
Gateway
500
Statically configured default gateway of the unit. This is
the IP address of the gateway to send packages to when
no more specific route can be found in the routing table.
Empty field indicates that no default gateway address has
been statically configured.
The currently active default gateway in use. N/A indicates
that no default gateway is in active use. A default gateway cannot be active if no route to the default gateway
is available.
Continued on next page
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Routing
Domain Name
Server(s)
Edit
Continued from previous page
(only for WeOS Extended) Routing, also known as IPforwarding, allows traffic to flow between VLANs. Use the
firewall to protect VLANs from unwanted traffic. Texts Enabled and Disabled shows routing status.
List manually configured DNS servers. An empty field
indicates that no DNS server has been manually configured.
Click this icon to edit ”this part” of the global settings.
These settings are described further in section 21.5.2.
To change the settings for a specific Interface click the associated edit icon which
will take you to the interface settings edit page. Interface settings are described
further in section 21.4.
© 2017 Westermo Teleindustri AB
501
Westermo OS Management Guide
Version 4.21.1-0
21.5.2
Edit Common Network Settings
Menu path: Configuration ⇒ Network (IP) ⇒ Global settings ⇒
Default Gateway
Routing
Name server 1
Name server 2
Statically configured default gateway of the unit.
This is the IP address of the gateway to send packages to when no more specific route can be found in
the routing table. Leave empty if no default gateway
is desired.
(only for WeOS Extended) Routing, also known as
IP-forwarding, allows traffic to flow between VLANs.
Use the firewall to protect VLANs from unwanted
traffic. Check this box to enable routing, uncheck to
disable.
IP address of (primary) DNS server.
IP address of (secondary) DNS server.
Click the Apply button to save and apply the changes.
502
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
21.5.3
NTP client
Menu path: Configuration ⇒ System ⇒ Date & Time
Figure 21.4: Switch date and time settings, NTP client
Current Date/Time
Remote NTP Server
Timezone
Shows current date and time. Click the
icon to
manually set date/time .
The IP address of a time server to be used to keep
the units calendar time synchronised. Leave empty
if you do not want to use a time server, or if NTP
server should be acquired via DHCP or PPP.
Select a timezone region to get adjusted local time.
© 2017 Westermo Teleindustri AB
503
Westermo OS Management Guide
Version 4.21.1-0
21.5.4
DDNS settings
Menu path: Configuration ⇒ Network (IP) ⇒ DDNS
Dynamic DNS (DDNS) provider settings
Enabled
Provider
SSL
Login
Password
Hostname
Interval
Check this box to enable Dynamic DNS, uncheck to disable.
Select DDNS provider. Example of supported providers:
dyndns http://www.dyndns.org,
freedns http://freedns.afraid.org, and
no-ip http://www.no-ip.com
See the online help for more.
Check this box if your DDNS provider supports HTTPS updates.
Set login username for the account at your DDNS provider
Set login password for the account at your DDNS provider
Set the DNS hostname, i.e., registered domain name which
should map to the IP address of this your switch. When
selecting freedns, the domain name must be followed by
a hash value (”HOSTNAME,HASH”); the hash is provided by
FreeDNS).
Set the interval by which DDNS verifies that the IP address
mapping at your DDNS provider matches the IP address of
your switch. Maximum 10 days (864000 seconds).
Click the Apply button to save and apply the changes.
504
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
21.5.5
Domain Proxy Settings
Menu path: Configuration ⇒ Network (IP) ⇒ DNS Proxy
Enabled
Forwarding Rules
If checked, the device acts as a DNS proxy server. If not
checked the device will not listen to and answer name
queries. See section 21.3.4 for more information.
Conditional domain forwarding rules:
Domain - The domain for which name queries should
be conditionally forwarded.
Name Server - The IPv4 address of the name server
to which the queries should be conditionally forwarded.
See section 21.3.4.1 for more information.
Click the Apply button to save and apply the changes.
© 2017 Westermo Teleindustri AB
505
Westermo OS Management Guide
Version 4.21.1-0
21.6
Managing network interfaces via the CLI
The available interface settings and monitoring commands are shown in the table
below:
Command
iface <IFNAME> inet <static|dynamic>
[no] enable [always]
[no] address <ADDRESS/LEN|
ADDRESS NETMASK> [secondary]
[no] primary
[no] distance <1-255>
[no] management <[ssh] [telnet] [http]
[https] [ipconfig] [snmp] | all>
[no] mtu <68-1500>
[no] tcp-mss <40-1460|auto>
[no] redirect
Default
Differs1
Enabled
Disabled
Section
Sec. 21.6.1
Sec. 21.6.2
Sec. 21.6.3
DEPRECATED
16
Enabled2
Sec. 21.6.4
Sec. 21.6.5
Sec. 21.6.6
Differs1
Differs1
Enabled
Sec. 21.6.7
Sec. 21.6.8
Sec. 21.6.9
Only if ”inet” is set to ”dynamic”
[no] clientid <hex|string|rawstring> <VALUE>
[no] vendor-classid <hex|string> <VALUE>
[no] zeroconf
Auto3
Auto4
Enabled
Sec. 21.6.10
Sec. 21.6.11
Sec. 21.6.12
Only for VLAN interfaces
[no] mac <X:X:X:X:X:X>
Auto
Sec. 21.6.13
Show interface status
show iface [IFNAME]
Sec. 21.6.14
1 Some
interface ”native” default settings depend on the interface type, see section 21.2.3.
Section 21.2.2 provides information on ”factory” default settings.
2 By default, all management services except Telnet are allowed on newly created VLAN and
PPP interfaces.
3 By default the Client ID is formed by concatenating ’01’ and the ’MAC address’, see section 21.2.9.
4 The Vendor Class ID defaults to the release name (WeOS v4.21.1).
506
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
21.6.1
Manage Network Interfaces
Syntax iface <IFNAME> inet <static|dynamic>
Context Global Configuration context
Usage Enter Interface Configuration context, and specify IP address assignment
method.
ˆ ”static” means static IP address assignment. The IP address is configured via the ”[no] address <ADDRESS/LEN|ADDRESS NETMASK>” command, see section 21.6.3.
ˆ If ”dynamic” is selected, the switch attempts to acquire its address via
DHCP (VLAN interfaces) or IPCP (PPP interfaces). If no DHCP server is
available, the interface will generally end up without an IP address. The
exception is the interface with best admin distance, which also gets a
link-local IPv4 address5 .
Use ”show iface” to show network interface configuration information of
all interfaces. Use ”show iface [IFNAME]” to show configuration information for a specific interface (also available as ”show” command within the
Interface Configuration of that specific interface).
Default values ”static” for VLAN and GRE interfaces, and ”dynamic” for PPP
interfaces. For VLAN interfaces there is one exception – If vlan1 does not
exist, or if it is created without an address method defined, vlan1 will default
to acquire its address dynamically via DHCP.
21.6.2
Interface Administrative Mode (Enabled or Not Enabled)
Syntax [no] enable [always]
Context Interface Configuration context
Usage Bring interface up/down. Note, even if an interface is configured administratively up, its operational status may still be down if the associated VLAN
(or PPP instance) is not up.
Use command ”enable” to configure an interface as up, and ”no enable”
to configure the interface as down.
5 Assignment
of link-local address can be disabled, see section 21.6.12.
© 2017 Westermo Teleindustri AB
507
Westermo OS Management Guide
Version 4.21.1-0
On LAN/VLAN interfaces, it is possible to circumvent the link status propagation property by configuring an interface as always up (”enable always”).
However, disabling link status propagation may significantly impact layer-3
protocols such as RIP, OSPF, VRRP, and more -– the protocols will have to
fall-back to other methods to detect link-down, e.g. hello message timeout
and similar. Do not use the ”enable always” setting unless you really know
what you are doing.
Note
An interface configured as always up will in SNMP report ifOperStatus
”testing(3)”.
Use ”show enable” to show whether this interface is configured as administratively enabled (up) or disabled (down).
Default values Enabled (”enable”)
21.6.3
IP Addresses (primary and secondary)
Syntax [no] address <ADDRESS/LEN|ADDRESS NETMASK> [secondary]
Context Interface Configuration context
Usage Set static IP address and netmask for an interface.
When static address assignment is chosen (”inet static”, see section 21.6.1),
the ”address” command can be used to the primary IP address of the
interface, as well as secondary IP addresses of the interface (using the
”secondary”) keyword.
When dynamic address assignment is chosen (”inet dynamic”, see section 21.6.1), the ”address” command is limited to assign secondary IP addresses.
Up to 8 secondary addresses can be configured for an interface.
It is possible to specify the boundary between the network part and the
host specific part of the IP address either as a prefix length (e.g. ”address
192.168.0.1/24”) or as a regular netmask (e.g., ”address 192.168.0.1
255.255.255.0”).
Use ”show address” to show the IP address setting for this interface.
508
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Default values Disabled (no address). That is, newly created interfaces have
no IP address configured, see also section 21.2.3.
21.6.4
Primary Interface
Syntax [no] primary
Context Interface Configuration context
Usage This command is deprecated and only kept for backwards compatibility
when upgrading. It is recommend to instead use the interface admin distance setting (section 21.6.5).
An old configuration file with this setting is converted to set the selected
interface as distance 1 and keep other interfaces at their default distance of
16.
For more information, see section 21.2.6.
21.6.5
Interface Administrative Distance
Syntax [no] distance <1-255> [trigger ID]
Context Interface Configuration context
Usage Administrative distance for routes learned on this interface.
Static routes learned dynamically, e.g. via DHCP, will be installed in the
routing table with this administrative distance.
Possible values are 1-255, where 1 is the best and 255 is infinity, it will be
visible in the routing table but will never be activated.
Use the form no distance to reset the value to its default value, 16. Use
distance 255 to prevent routes from ever being activated.
A trigger ID may be set, e.g., for monitoring an upstream network with a
ping trigger, and dynamically adjusting the default route to infinite distance.
Effectively switching to another upstream interface not only on link loss.
For more information, see section 21.2.6.
Default values 16 (no distance)
Notes:
© 2017 Westermo Teleindustri AB
509
Westermo OS Management Guide
Version 4.21.1-0
ˆ A PPP interface created via PPPoE will ”inherit” the admin distance setting from its associated VLAN interface.
ˆ The old primary setting on an interface is converted to distance 1 and
all other interfaces are shifted downwards in priority.
ˆ This setting does not apply to protocols such as RIP and OSPF.
21.6.6
Management Service Filtering
Syntax [no] management <[ssh][telnet][http][https][ipconfig][snmp]|all>
Context Interface Configuration context
Usage Filter management services on this interface.
The setting controls what services are allowed to use on this network interface. E.g., ”management ssh https” adds SSH and HTTPS to the set of services accessible for traffic entering via this interface, and ”no management
http” disallows management via unencrypted HTTP on this interface.
Use ”no management” to filter out access to all management services on
this interface.
Use ”management all” to allow all management services on this interface.
Use ”show management” to show the list of currently allowed services via
this interface.
Default values All services except ”telnet” are allowed.
Note: PPP interfaces created via PPPoE will inherit the management settings
from its associated VLAN interface.
21.6.7
Interface MTU Size
Syntax [no] mtu <68-1500>
Context Interface Configuration context
Usage Configure a non-default maximum transmission unit (MTU) size (in bytes)
for this interface. The MTU size is the packet size a network interface will
pass to the link layer for transmission, i.e., the maximum payload of the link
layer protocol.
510
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
The default is to let the MTU depend on the type of link layer (auto mode).
For interfaces associated with Ethernet and DSL links this implies a default
MTU of 1500 bytes.
For PPP interfaces (PPPoE), the MTU is set to 8 bytes less than the MTU of
the associated VLAN interface, which typically implies a PPP interface MTU
of 1492 bytes (1500 − 8). This value is set at the time of PPP interface
creation; if the VLAN interface MTU is changed afterwards, the PPP interface
MTU is not updated automatically. Note: The operational MTU can change
based on the PPP connection negotiation, see section 33.3.19.
The MTU of GRE interfaces defaults to 1476 bytes.
Use ”mtu <68-1500>” to set a non-default MTU size. Use ”no mtu” to specify that the interface should let its MTU be the default MTU of the associated
link type.
Use ”show mtu” to how the interface maximum transfer unit (MTU) size
setting.
Default values
ˆ VLAN interfaces: Auto (”no mtu”) For Ethernet and DSL links, this implies MTU 1500 bytes.
ˆ GRE interfaces: 1476 bytes (”mtu 1476”)
ˆ PPP interfaces (PPPoE): Typically 1492 bytes (”mtu 1492”, i.e., 8 bytes
less than the associated VLAN interface)
21.6.8
Interface TCP MSS Size
Syntax [no] tcp-mss <40-1460|auto>
Context Interface Configuration context
Usage Enable/disable TCP-MSS clamping on this interface.
TCP-MSS clamping is used to limit the packet size (or more precisely, limit
the ”maximum TCP segment size”) of TCP connections over the given interface, and is useful in situations where path MTU discovery of some reason
does not work.
Enabling TCP-MSS clamping implies additional packet processing, thus it degrades routing performance somewhat. It is disabled by default on most
© 2017 Westermo Teleindustri AB
511
Westermo OS Management Guide
Version 4.21.1-0
interface types (exception is PPP interface of type PPPoE).
Use ”tcp-mss <BYTES>” to limit TCP-MSS to the given number of bytes.
Use ”tcp-mss auto” to let the TCP-MSS depend on the MTU of the interface
(”MTU-40”, i.e., interface MTU minus typical size of IP and TCP headers).
This will work fine for typical TCP connections, but is not likely to work over
IPsec tunnels or when additional IP header options are in use.
Use ”no tcp-mss” to disable TCP-MSS clamping.
Use ”show tcp-mss” to show the interface maximum TCP segment size
(MSS).
Default values Disabled (no tcp-mss) (Exception: ”tcp-mss 1412” for PPPoE
PPP interfaces.)
21.6.9
Sending ICMP Redirect messages
Syntax [no] redirect
Context Interface Configuration context
Usage Enable/disable sending of ICMP Redirect messages. When enabled on a
WeOS router, the router will send ICMP Redirect messages when detecting
that packets coming in on this interface have a more optimal route towards
the destination.
Use ”redirect” to enable sending of ICMP Redirect, and ”no redirect” to
disable it.
Use ”show redirect” to show if sending of ICMP Redirect is enabled or
disabled.
Default values Enabled
21.6.10
DHCP Client ID
Syntax [no] clientid <hex|string|rawstring> <VALUE>
Context Interface Configuration context
512
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Usage Configure the Client ID that is sent in DHCP requests. This command is
only applicable when DHCP is used for address assignment (section 21.6.1).
ˆ Auto: (Default) Use ”no clientid” to form the Client ID automatically
by concatenating ’01’ and the MAC address of the interface.
ˆ Hex: Use ”clientid hex <VALUE>” to form the Client from a manual
sequence of hexadecimal numbers.
ˆ String: Use ”clientid string <VALUE>” to form the Client ID from a
manual string. A prefix ’00’ will be added.
ˆ Raw String: Use ”clientid rawstring <VALUE>” to form the Client
ID from a manual string, without any ’00’ prefix.
The Client ID is restricted to the length of 1-63 string characters or hexadecimal octets. When using the ”string” or ”rawstring” method, valid
characters are a-z, A-Z, 0-9, hyphen (’-’) and period (’.’). The string shall not
start or end with a hyphen (’-’) or period (’.’).
Use ”show clientid” to show the Client ID setting for this interface.
Default values Auto (form Client ID from MAC address)
Example
Example
”clientid hex c0ffee” > 3d 03 c0 ff ee
”clientid hex c0:ff:ee” > 3d 06 c0 ff ee
”clientid string coffee” > 3d 07 00 63 6f 66 66 65 65
”clientid rawstring coffee” > 3d 06 63 6f 66 66 65 65
”no clientid” > 3d 07 01 00 00 5e 00 53 01
The last example assumes the interface has the MAC address 00:00:5e:00:53:01.
21.6.11
DHCP Vendor Class ID
Syntax [no] vendor-classid <hex|string> <VALUE>
Context Interface Configuration context
© 2017 Westermo Teleindustri AB
513
Westermo OS Management Guide
Version 4.21.1-0
Usage Configure the Vendor Class ID (DHCP option 60) sent in DHCP requests.
This command is only applicable when DHCP is used for address assignment
(section 21.6.1).
ˆ Auto: (Default) Use ”no vendor-classid” to form the Vendor Class ID
automatically as the release name (WeOS v4.21.1).
ˆ Hex: Use ”vendor-classid hex <VALUE>” to form the Vendor Class ID
from a manual sequence of hexadecimal numbers.
ˆ String: Use ”vendor-classid string <VALUE>” to form the Client ID
from a manual string.
The Vendor Class ID is restricted to the length of 1-63 string characters or
hexadecimal octets. When using the ”string” method, valid characters are
ASCII 32-126 except double quote (”), i.e., ASCII 34. Double quotes are used
to delimit strings including space.
Use ”show vendor-classid” to show the Vendor Class ID setting for this
interface.
Default values Auto (form Vendor Class ID from software release name)
Example
Example
vendor-classid hex c0ffee > 3c 03 c0 ff ee
vendor-classid hex c0:ff:ee > 3c 03 c0 ff ee
vendor-classid string coffee > 3c 06 63 6f 66 66 65 65
vendor-classid string ’’coffee mug’’ > 3c 0a 63 6f 66 66 65 65 20 6d 75 67
no vendor-classid > reset to send software release name, e.g.,
WeOS v4.21.1 as ASCII string.
21.6.12
Enable/disable link-local address
Syntax [no] zeroconf
Context Interface Configuration context
Usage Enable/disable assignment of link-local IP address. This command is only
applicable when DHCP is used for address assignment (section 21.6.1).
514
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
If no DHCP server is available, an interface will generally end up without an
IP address. The exception is the interface with best admin distance, which
by default is also assigned a link-local IPv4 address.
When link-local address is undesirable, the ”no zeroconf” setting can be
used to disable it on this interface.
Use ”show zeroconf” to show the current setting.
Default values Enabled
21.6.13
VLAN Interface MAC address
Syntax [no] mac <X:X:X:X:X:X>
Context Interface Configuration context
Usage Configure a specific MAC address for this (VLAN) interface. The address
is given as a colon-separated hexadecimal string of numbers, e.g., ”mac
00:00:5e:00:53:01”. Leading zeros can be ignored. Uppercase or lowercase letters can be used.
Use ”no mac” specify that the interface should get its MAC address automatically.
Use ”show mac” to show the interface MAC setting for this (VLAN) interface.
For more information, see section 21.2.4.
Default values Auto (no mac)
Example
example:/config/iface-vlan1/#> mac 00:00:5e:00:53:01
example:/config/iface-vlan1/#>
21.6.14
Show Network Interface Status
Syntax show iface [IFNAME]
Context Admin Exec context.
© 2017 Westermo Teleindustri AB
515
Westermo OS Management Guide
Version 4.21.1-0
Usage Show status information for this interface (or all interfaces). If dynamic
address assignment is configured on an interface, this command will display
the IP address acquired.
Default values Unless a specific interface is specified, status for all interfaces
will be shown.
516
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
21.7
Managing general IP settings via the CLI
The available general IP settings and monitoring commands are shown below.
Command
Configure general IP settings
ip
[no] default-gateway <IPADDR>
[no] route <NETWORK/LEN>
<GATEWAY|IFNAME> [DISTANCE]
[no] forwarding
[no] name-server <IPADDR>
[no] domain <DOMAIN>
[no] domain-proxy
[no] domain-forward <DOMAIN>
<IPADDR>
[no] host <FQDN | HOSTNAME>
<IPADDR>
[no] ddns
[no] provider <dyndns|freedns|no-ip>
[no] ssl
[no] login <USERNAME>
<PASSWORD>
[no] hostname <HOSTNAME>[,HASH]
[no] interval <SECONDS>
icmp
[no] broadcast-ping
[no] ntp
[no] enable
[no] server <FQDN|IPADDR>
[no] enable
[no] poll-interval <SECONDS>
[no] sntp
[no] server <FQDN|IPADDR>
[no] poll-interval <SECONDS>
© 2017 Westermo Teleindustri AB
Default
Section
(DEPRECATED)
Section 21.7.1
Section 21.7.2
Section 21.7.3
Distance 1
Enabled
Disabled
Disabled
Enabled
Section
Section
Section
Section
Section
21.7.4
21.7.5
21.7.6
21.7.7
21.7.8
Section 21.7.9
Disabled
dyndns
Disabled
Disabled
Section
Section
Section
Section
21.7.10
21.7.11
21.7.12
21.7.13
Disabled
600
Section 21.7.14
Section 21.7.15
Section 21.7.16
Enabled
Section 21.7.17
Disabled
Section 21.7.18
Enabled
Section 21.7.19
N/A
Section 21.7.20
Enabled
Section 21.7.21
600 sec
Section 21.7.22
(DEPRECATED) Section 21.7.23
Disabled
Section 21.7.24
600 sec
Section 21.7.25
Continued on next page
517
Westermo OS Management Guide
Version 4.21.1-0
Command
Show general IP status
show ip route
show ip name-server
show ip domain
show ip host
show ntp
21.7.1
Continued from previous page
Default
Section
Section
Section
Section
Section
Section
21.7.26
21.7.27
21.7.28
21.7.29
21.7.30
Manage Global IP Settings
Syntax ip
Context Global Configuration context
Usage Enter IP Configuration context
Use ”show ip” to show general IP configuration settings.
Default values Not applicable.
21.7.2
Configure IP Default Gateway
Syntax [no] default-gateway <ADDRESS>
Context IP Configuration context
Usage This command is deprecated and only kept for backwards compatibility
when upgrading. It is recommend to instead use the route command since
it also has the distance attribute.
A default route configured using this command will always get a distance
of 1. With multiple upstream WAN connections using PPPoE or DHCP it is
recommended to use the route command instead.
Use ”show gateway” to show configured default gateway.
Default values Disabled (”no default-gateway”)
518
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
21.7.3
Configure Static IP Routes
Syntax [no] route <NET MASK | NET/LEN> <GATEWAY | IFNAME> [DISTANCE]
Context IP Configuration context
Usage Add or remove a static IP route, including default routes.
The network boundary of the destination subnet can be given as a netmask
(e.g., ”route 192.168.3.0 255.255.255.0 192.168.0.1”) or as a prefix
length (e.g., ”route 192.168.3.0/24 192.168.0.1”).
System default routes are setup using the subnet 0.0.0.0 with prefix length
0, but the key keyword ’default’ is much easier to use ”route default
192.168.0.1”. The optional distance is useful when setting up backup
routes in multiple upstream scenarios where interfaces acquire default routes
using PPPoE or DHCP.
The destination network is however typically located remotely (specify the
next hop gateway, e.g., ”route 192.168.3.0/24 192.168.0.1”), but it is
also possible to use the static route command to specify additional directly
connected subnets (specify the local interface, e.g., ”route 192.168.3.0/24
vlan1”).
Use the ”no”-form to remove a static route, e.g., ”no route 192.168.3.0/24
192.168.0.1”.
Use ”show route” to list configured static routes.
Default values Using ”no route” (without a subnet address, etc.) removes all
configured static routes.
21.7.4
Manage IP Forwarding
Syntax [no] forwarding
Context IP Configuration context
Usage (only for WeOS Extended) Enable/disable IPv4 routing.
Use ”show forwarding” to show whether IP forwarding (routing) is enabled
or disabled.
Default values Enabled (”forwarding”)
© 2017 Westermo Teleindustri AB
519
Westermo OS Management Guide
Version 4.21.1-0
21.7.5
Name Server (DNS)
Syntax [no] name-server <ADDRESS>
Context IP Configuration context
Usage Add/remove name-server (DNS). Two name-servers can be configured call the same ”name-server” command twice.
Run ”no name-server <ADDRESS>” to remove a specific name server, or
”no name-server” to remove all configured name servers.
If a name server is not configured using the ”name-server” command,
name server(s) (and domain search path) can be acquired dynamically from
an interface with DHCP address assignment.
Use ”show name-server” to show configured name servers.
Default values Disabled (”no name-server”) Running ”no name-server” (without specifying any name removes all configured name servers.
21.7.6
Domain Search Path
Syntax [no] domain <DOMAIN>
Context IP Configuration context
Usage Add/remove domain search path. A single search path can be added.
Run ”no domain” to remove the domain search path.
If a name server is not configured using the ”name-server” command, domain(s) can be acquired dynamically from an interface with DHCP address
assignment.
Use ”show domain” to show configured domain search path.
Default values Disabled (”no domain”)
21.7.7
Enable/Disable DNS proxy service
Syntax [no] domain-proxy
Context IP Configuration context
520
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Usage Enable or disable DNS proxy support. When enabled, the unit will act as
a DNS server and respond to DNS queries for known hosts:
ˆ either statically added by the ”host” (section 21.7.9), see also the
”show ip host” (section 21.7.29) command, or
ˆ hosts for which this unit acts as DHCP server (chapter 22), see also the
”show dhcp-clients” (section 22.3.22) command .
Furthermore, the unit will act as a caching DNS forwarder; DNS queries of
unknown hosts are forwarded to the unit’s own DNS server (see the ”show
ip name-server” command described in section 21.7.27), and the answer
is cached for fast response of subsequent requests for the same host.
Use command ”domain-proxy” to enable the DNS proxy service, and ”no
domain-proxy” to disable it.
Use ”show domain-proxy” to view the current setting.
Default values Enabled (”domain-proxy”)
Example
example:/#> show ip host
127.0.0.1 localhost
127.0.1.1 example.local example
192.168.3.11
mypc
example:/#> ping mypc
Press Ctrl-C to abort PING mypc (192.168.3.11): 56 data bytes
64 bytes from 192.168.3.11: seq=0 ttl=64 time=1.049 ms
64 bytes from 192.168.3.11: seq=1 ttl=64 time=0.627 ms
ˆC
--- mypc ping statistics --2 packets transmitted, 2 packets received, 0% packet loss
round-trip min/avg/max = 0.627/0.838/1.049 ms
example:/#> show dhcp-clients
Lease Time MAC Address
IP Address
Hostname
Client ID
===============================================================================
120
00:07:7c:03:ec:02 192.168.5.106
alice
01:00:07:7c:03:ec:02
example:/#> ping alice
Press Ctrl-C to abort PING alice (192.168.5.106): 56 data bytes
64 bytes from 192.168.5.106: seq=0 ttl=64 time=1.182 ms
64 bytes from 192.168.5.106: seq=1 ttl=64 time=0.754 ms
ˆC
--- alice ping statistics --2 packets transmitted, 2 packets received, 0% packet loss
round-trip min/avg/max = 0.754/0.968/1.182 ms
example:/#>
© 2017 Westermo Teleindustri AB
521
Westermo OS Management Guide
Version 4.21.1-0
21.7.8
Add conditional domain forwarding entry
Syntax [no] domain-forward <DOMAIN> <IPADDR>
Context IP Configuration context
Usage Add conditional domain forwarding entries. When an entry exists for a
domain, name lookups for the domain will be forwarded to the DNS server
specified as IP-address. The server with most specific domain match will be
used.
Name lookups for other domains will be handled as described in section 21.3.4.
The commands ”domain-proxy” and ”no domain-proxy” (section 21.7.7)
enables and disables the conditional domain forwarding as well.
Use ”show domain-forward” to view the current setting.
Default values Not applicable (no forwarding entries configured)
Example
example:/#> configure ip
example:/config/ip/#> domain-forward example1.org 192.168.2.2
example:/config/ip/#> domain-forward ext.example1.org 192.168.2.3
example:/config/ip/#> show domain-forward
2 domain forwarding entries configured
Domain
Name Server
==============================================================================
example1.org
192.168.2.2
ext.example1.org
192.168.2.3
example:/config/ip/#> leave
example:/#>
ping server.example1.org
Press Ctrl-C to abort PING mypc (192.168.2.40): 56 data bytes
64 bytes from 192.168.2.40: seq=0 ttl=64 time=1.049 ms
64 bytes from 192.168.2.40: seq=1 ttl=64 time=0.627 ms
ˆC
--- mypc ping statistics --2 packets transmitted, 2 packets received, 0% packet loss
round-trip min/avg/max = 0.627/0.838/1.049 ms
example:/#>
ping server.ext.example1.org
Press Ctrl-C to abort PING mypc (192.168.2.140): 56 data bytes
64 bytes from 192.168.2.140: seq=0 ttl=64 time=1.049 ms
64 bytes from 192.168.2.140: seq=1 ttl=64 time=0.627 ms
ˆC
--- mypc ping statistics --2 packets transmitted, 2 packets received, 0% packet loss
round-trip min/avg/max = 0.627/0.838/1.049 ms
522
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
The first name lookup for server.example1.org will be forwarded to name-server
192.168.2.2 and the second lookup for server.ext.example1.org will be forwarded
to name-server 192.168.2.3 due to the matching rules.
21.7.9
Add static hostname lookup entry
Syntax [no] host <FQDN | HOSTNAME> <IPADDR>
Context IP Configuration context
Usage Add or delete entries in the static hostname resolution table (host table).
The table is both used when resolving hostnames of DNS requests originating from the unit itself (e.g., when running ”ping www.example.com” from
the CLI command line), and when responding to DNS queries from hosts
(assuming this unit is configured as DNS proxy, see section 21.7.7).
ˆ Hostnames containing a dot (”.”) are interpreted as fully qualified domain names (FQDN).
ˆ Hostnames without a dot are interpreted as simple hostnames. The system will both be able to resolve DNS queries for the hostname, as well as
hostname concatenated with the unit’s domain search path. Use ”show
ip domain” (section 21.7.28) to view the unit’s search path domain.
Example
example:/#> configure
example:/config/#> ip
example:/config/ip/#> domain example.org
example:/config/ip/#> host mypc 192.168.10.1
example:/config/ip/#> host www.anotherexample.org 10.0.0.1
example:/config/ip/#> leave
example:/#> show ip hosts
127.0.0.1 localhost
127.0.1.1 example.local example
192.168.10.1
mypc
mypc.example.org
10.0.0.1
www.anotherexample.org
example:/#> ping mypc
Press Ctrl-C to abort PING mypc (192.168.10.1): 56 data bytes
64 bytes from 192.168.10.1: seq=0 ttl=64 time=8.291 ms
64 bytes from 192.168.10.1: seq=1 ttl=64 time=0.650 ms
ˆC
--- mypc ping statistics --2 packets transmitted, 2 packets received, 0% packet loss
round-trip min/avg/max = 0.650/4.470/8.291 ms
example:/#>
© 2017 Westermo Teleindustri AB
523
Westermo OS Management Guide
Version 4.21.1-0
Use ”no host <HOSTNAME>” to remove a specific entry in the host table,
and ”no host” to remove all configured entries in the host table.
Use ”show host” to view the currently configured static host entries.
Default values Not applicable (no static host entries configured)
21.7.10
Manage DDNS Settings
Syntax [no] ddns
Context IP Configuration context
Usage Enter DDNS Configuration context. Upon entering the context, the DDNS
service will be enabled. However, it will not be activated until valid DDNS
parameters (login, etc.) are configured. Use ”no ddns” to disable the DDNS
service.
Use ”show ddns” to show configured DDNS settings (also available as ”show”
command within the DDNS Configuration context).
Default values Disabled (”no ddns”)
21.7.11
Set DDNS Provider
Syntax [no] provider <dyndns|freedns|no-ip>
Context DDNS Configuration context
Usage Set DDNS provider. Example of supported providers:
dyndns http://www.dyndns.org,
freedns http://freedns.afraid.org, and
no-ip http://www.no-ip.com
For a complete list of supported DDNS providers, type ”help provider”.
Use ”no provider” to return to the default provider setting.
Default values dyndns
524
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
21.7.12
Enable HTTPS Updates
Syntax [no] ssl
Context DDNS Configuration context
Usage Enable/disable HTTPS updates, if the provider supports it.
Use ”show ssl” to show whether HTTPS updates is enabled or disabled.
Default values Disabled (HTTP)
21.7.13
Set DDNS Login and Password
Syntax [no] login <USERNAME> <PASSWORD>
Context DDNS Configuration context
Usage Set login username and password for your account at your DDNS provider
(see section 21.7.11). Use ”no login” to remove a configured DDNS login
setting.
Default values Disabled
21.7.14
Set DDNS Hostname
Syntax [no] hostname <HOSTNAME>[,HASH]
Context DDNS Configuration context
Usage Set the DNS hostname, i.e., registered domain name which should map
to the IP address of this your switch.
When selecting ”provider freedns”, the domain name must be followed
by a hash value (”hostname HOSTNAME,HASH”); the hash is provided by
FreeDNS).
Default values Disabled
21.7.15
Set DDNS interval
Syntax [no] interval <SECONDS>
© 2017 Westermo Teleindustri AB
525
Westermo OS Management Guide
Version 4.21.1-0
Context DDNS Configuration context
Usage Set the interval by which DDNS verifies that the IP address mapping at
your DDNS provider matches the IP address of your switch. Maximum 10
days (864000 seconds).
Use ”no interval” to return to the default provider setting.
Default values 600 (seconds)
21.7.16
Manage ICMP Settings
Syntax icmp
Context IP Configuration context
Usage Enter ICMP Configuration context.
Use ”show icmp” to show ICMP settings (also available as ”show” command
within the ICMP Configuration context).
Default values Not applicable.
21.7.17
Enable/disable Broadcast Ping
Syntax [no] broadcast-ping
Context ICMP Configuration context
Usage Define whether the switch should respond to broadcast ”ping” (ICMP Echo
Request) messages or not. Responding to broadcast ping is convenient
when troubleshooting the network, but can in some situations be considered a security risk.
Use ”no broadcast-ping” to disable responding to broadcast ping messages.
Use ”show broadcast-ping” to show whether the switch is configured to
respond to broadcast ping messages or not.
Default values Enabled (”broadcast-ping”)
526
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
21.7.18
Manage NTP Settings
Syntax [no] ntp
Context Global Configuration context
Usage Enter NTP Configuration context by using the ”ntp” command.
Use ”no ntp” to remove all configured NTP settings.
Use ”show ntp” to show NTP settings (also available as ”show” command
within the NTP Configuration context).
Default values Not applicable.
21.7.19
Enable/Disable NTP Settings
Syntax [no] enable
Context NTP Configuration context
Usage Enable or disable configured NTP settings.
Use ”enable” to enable/activate configured NTP settings. Use ”no enable”
to disable/deactivate configured NTP settings (the settings are not removed,
only deactivated).
Use ”show enable” to show whether NTP settings are enabled or disabled.
Default values Enabled
21.7.20
Manage (remote) NTP Server(s)
Syntax [no] server <FQDN|IPADDR>
Context NTP Configuration context
Usage Add, delete, or manage a remote NTP server with specified IP Address
or domain name, to set the time on this unit. Up to 8 NTP servers can be
configured. With the ”server <FQDN|IPADDR>” you enter the NTP Remote
Server Configuration context for that specific NTP server.
If no (remote) NTP server is configured, the unit can acquire NTP server(s)
dynamically from an interface with DHCP address assignment.
© 2017 Westermo Teleindustri AB
527
Westermo OS Management Guide
Version 4.21.1-0
Use ”no server <FQDN|IPADDR>” to remove a specific NTP server, or ”no
server” to remove all configured NTP servers.
Use ”show server” to show settings for all configured NTP servers, or ”show
server <FQDN|IPADDR>” to show NTP settings for a specific NTP server (also
available as ”show” command within the NTP Remote Server Configuration
context).
Default values Not applicable
21.7.21
Enable/Disable (remote) NTP Server
Syntax [no] enable
Context NTP Remote Server Configuration context
Usage Enable or disable configured settings for this NTP server.
Use ”enable” to enable/activate configured NTP server settings. Use ”no
enable” to disable/deactivate configured NTP server settings (the settings
are not removed, only deactivated).
Use ”show enable” to show whether NTP server settings are enabled or
disabled.
Default values Enabled
21.7.22
Set NTP Server Poll Interval
Syntax [no] poll-interval <30-720>
Context NTP Remote Server Configuration context
Usage Set NTP server poll interval (in seconds) for this NTP server. ”no poll-interval”
will reset the poll interval to its default (600 seconds).
Use ”show poll-interval” to show configured poll interval.
Default values 600 (seconds)
21.7.23
Manage NTP Client Settings (Deprecated)
Syntax [no] sntp
528
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Context Global Configuration context
Usage Enable NTP client and enter NTP Client Configuration context by using
the ”sntp” command.
Use ”no sntp” to disable the NTP client service.
Use ”show sntp” to show NTP client settings (also available as ”show” command within the NTP Client Configuration context).
Note
The NTP Client Configuration context is deprecated and kept for backwards compatibility. NTP client settings is instead handled as part of
other NTP settings in the NTP Configuration, see section 21.7.18.
Default values Not applicable.
21.7.24
Set NTP Client (Remote) Server Address (deprecated)
Syntax [no] server <FQDN|IPADDR>
Context NTP Client Configuration context
Usage Set IP Address, or domain name, of NTP Server used by this client.
A single NTP server IP address, or a fully qualified domain name, FQDN, can
be configured. If disabled, NTP server can be acquired dynamically from an
interface with DHCP address assignment.
”no server” to remove a configured NTP server address.
Use ”show server” to show NTP client (remote) server setting.
Note
The ”server” command within NTP Client Configuration context is deprecated and kept for backwards compatibility. NTP client settings for
remote server is instead handled as part of other NTP settings in the
NTP Remote Server Configuration, see section 21.7.20.
Default values Disabled
© 2017 Westermo Teleindustri AB
529
Westermo OS Management Guide
Version 4.21.1-0
21.7.25
Set NTP Poll Interval (deprecated)
Syntax [no] poll-interval <30-720>
Context NTP Client Configuration context
Usage Set NTP server poll interval (in seconds). ”no poll-interval” will reset
the poll interval to its default (600 seconds).
Use ”show poll-interval” to show configured poll interval.
Note
The ”poll-interval” command within NTP Client Configuration context is deprecated and kept for backwards compatibility. NTP client settings for remote server is instead handled as part of other NTP settings
in the NTP Remote Server Configuration, see section 21.7.22.
Default values 600 (seconds)
21.7.26
Show IP Forwarding Table
Syntax show ip route
Context Admin Exec context
Usage Show IP Forwarding table (summary of configured routes and routes acquired dynamically).
Default values Not applicable.
21.7.27
Show Name Server and Domain Search Path Status Information
Syntax show ip name-server
Context Admin Exec context
Usage Show name-server and domain search path status information (statically
configured or acquired dynamically)
530
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
21.7.28
Show Domain Search Path Status Information
Syntax show ip domain
Context Admin Exec context
Usage Show domain search path status information (statically configured or acquired dynamically)
Example
example:/#> show ip domain
example.org
example:/#>
21.7.29
Show local host table
Syntax show ip hosts
Context Admin Exec context
Usage Show the local hostname resolution table. Static hostname resolution
entries configured with the ”host” command (section 21.7.9) are listed, as
well as entries for the unit itself (localhost and entries for the unit’s own
hostname, see section 8.3.2).
Example
example:/#> show ip hosts
127.0.0.1 localhost
127.0.1.1 example.local example
192.168.10.1
10.0.0.1
example:/#>
21.7.30
mypc
mypc.example.org
www.anotherexample.org
Show NTP Status Information
Syntax show ntp [verbose]
Context Admin Exec context
© 2017 Westermo Teleindustri AB
531
Westermo OS Management Guide
Version 4.21.1-0
Usage Show NTP status information. An asterisk ’*’ shows which NTP server
is used to synchronise the time. For more information, use ”show ntp
verbose”.
Example
example:/#> show ntp
NTP Client/Server running as PID: 805
210 Number of sources = 2
MS Name/IP address
Stratum Poll Reach LastRx Last sample
===============================================================================
^* ntp-anycast.kth.se
2
9
37
370
+222us[ -916ms] +/22ms
^- cecar.ddg.lth.se
2
9
37
370 -8317us[-8317us] +/81ms
532
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
21.8
Feature Parameters
MAX_CHARACTERS_CLIENTID
MAX_HEX_NIBBLES_CLIENTID
MAX_CHARACTERS_VENDORCLASSID
MAX_HEX_NIBBLES_VENDORCLASSID
© 2017 Westermo Teleindustri AB
63
126
63
126
533
Westermo OS Management Guide
Version 4.21.1-0
Chapter 22
DHCP Server
The WeOS DHCP server is capable of handing out IP settings to hosts (DHCP
clients) on local and remote IP subnets. For each defined IP subnet, the DHCP
server can assign IP addresses dynamically from a pool of addresses, but also
statically based on
ˆ the port the (DHCP) client is connected to (”one IP per port”, DHCP option 82),
ˆ the DHCP client identifier provided by the connecting client,
ˆ the MAC address of the connecting client, or
ˆ a combination of DHCP option 82 and DHCP client identifier
To serve clients on remote IP subnets, DHCP relay agents would be used to forward the DHCP messages between the clients and the DHCP server. In WeOS you
can even configure a DHCP relay agent on the same unit as the DHCP server –
this is useful if you wish to hand out addresses per port (DHCP option 82) on the
DHCP server unit itself. For more information on configuring DHCP relay agents,
see chapter 23.
The WeOS DHCP server is also able to act as a (proxy) DNS server for the DHCP
clients it serves (see section 21.3.4).
Being part of an embedded system, the WeOS DHCP server does not store the
current set of leases in persistent storage. In most use cases this is fine, however if it necessary that the current lease table survives a reboot you are recommended to use a dedicated DHCP server instead.
534
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
22.1
Overview of DHCP Server Support in WeOS
The table below presents a summary of DHCP server functionality in WeOS.
Feature
Web
General DHCP Server Functionality
Enable DHCP Server
X
Define subnets to serve
X
Caching DNS server
X
Enable/Disable Ping check
X
Server Listening UDP Port
Server Source UDP Port
Address lease preemption
X
CLI
X
X
X
X
X
X
X
General Description
Section 22.1.1
Secs. 22.1.1, 21.3.4
Section 22.1.3
-”-”-”-
Client configuration settings
Client IP assignment and matching
Address pool
X
X
Section 22.1.2.1
Per port (Option 82)
X
X
Secs. 22.1.2.1, 22.1.4
Per client-ID
X
X
Secs. 22.1.2.1, 22.1.4
Per MAC
X
X
Secs. 22.1.2.1, 22.1.4
Per port & client-ID
X
X
Secs. 22.1.2.1, 22.1.4
Deny client service
X
X
Section 22.1.2.1
Additional client configuration parameters
Default Gateway
X
X
Section 22.1.2.2
DNS Server
X
X
-”Domain search path
X
X
-”NTP Server
X
X
-”Hostname
X
X
-”TFTP Server Name
X
X
-”TFTP Server Address
X
X
-”TFTP File
X
X
-”Lease time
X
X
-”DHCP Server Status
List current clients
© 2017 Westermo Teleindustri AB
X
535
Westermo OS Management Guide
Version 4.21.1-0
22.1.1
Introduction to WeOS DHCP server support
Intranet/Internet
DHCP pools:
A: 192.168.1.100−150/24
gw 192.168.1.1
B: 192.168.2.100−150/24
gw 192.168.2.1
DHCP pools:
A: 192.168.1.100−150/24
gw 192.168.1.1
B: 192.168.2.100−150/24
gw 192.168.2.1
C: 192.168.3.100−150/24
gw 192.168.3.1
DHCP
Server
.1
.1
192.168.1.0/24
192.168.2.0/24
PC1
PC2
a)
Intranet/Internet
192.168.1.0/24
PC1
10.0.1.1
.1
DHCP
Relay
Server
Agent
.1
.1
192.168.2.0/24
PC2
192.168.3.0/24
PC3
b)
Figure 22.1: Sample DHCP use cases: (a) DHCP server serving local subnets, and
(b) serving local and remote subnets.
DHCP servers are typically used to dynamically assign IP settings (IP address,
netmask, default gateway, etc.) to hosts on the local subnet, see fig. 22.1a. The
server maintains an address pool for each served subnet, from which it assigns
addresses to DHCP clients currently present on that LAN. Addresses in the pool
are maintained dynamically - they are assigned to clients for a configurable time
(DHCP lease time), and if a client goes away, that address can be reused and
assigned to another client.
The DHCP server also hands out configuration settings for default gateway and
DNS server(s). For local clients as in fig. 22.1a, the DHCP server unit will commonly act as default gateway and DNS server1 too.
To provide DHCP service on multiple subnets throughout your infrastructure, you
could either deploy a DHCP server on each subnet, or you could use DHCP relay
agents to forward DHCP packets between the remote subnet and a central DHCP
server, as shown in fig. 22.1b.
When configuring the server, there is no major difference if the subnet is local or
remote – you will simply define which subnets to serve. When the server receives
a DHCP message, it will automatically detect which subnet the request originated
from and thereby be able to hand out an address from the pool it has defined for
that subnet.
1A
536
WeOS unit acts as (proxy) DNS server by default, see section 21.3.4.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
In addition to handing out addresses dynamically from a pool, it is possible to
assign addresses more specifically based on the client’s MAC address, the client
identifier (client-ID) included in the DHCP messages from the client, or the physical port where the client is connected. More information on this is given in sections 22.1.2.1 and 22.1.5.
The DHCP server unit will by default accept incoming DHCP packets on any of
its interfaces, including the loopback interface ”lo”. (The exception are those
interfaces where a DHCP relay agent has been configured on the local unit, see
section 22.1.5 – there DHCP packets will be handled by the relay agent.)
Hint
For security purposes you may wish to avoid accepting DHCP packets on
some interfaces, e.g., your upstream interface towards the Internet. To
block such request you are recommended to configure appropriate deny
filter rules, e.g., ”filter deny in vlan1 dport 67 proto udp” to block
incoming DHCP request on interface vlan1. For more details on the WeOS
firewall, see chapter 31.
By default the DHCP server will check that an address is not in use before offering
it to a client. In some rare cases it may be useful to disable this.
22.1.2
Available features and scope of configuration
Configuration of DHCP server settings can be done at three levels of scope:
ˆ Global level: Settings in the global scope can be seen as default values.
They are valid for communication with all DHCP clients, unless overridden
by a corresponding setting at subnet or host level.
ˆ Subnet level: Settings at the subnet level apply to a specific IP subnet.
They override corresponding settings done at global level, but may themselves be overridden by settings at host level. Some settings only apply to
the subnet level, e.g., specifying the IP range for the address pool.
ˆ Host level: Settings at host level applies to individual DHCP clients. They
override corresponding settings at global or subnet level. Some settings
only apply to the host level, e.g., specifying the IP address to hand out to a
specific host. Assigning IP settings per host is referred to as ”static lease”,
as opposed ”dynamic” assignment from a pool.
© 2017 Westermo Teleindustri AB
537
Westermo OS Management Guide
Version 4.21.1-0
Below is a summary of server settings for client configuration and DHCP server
features, and the possible level of scope. More details of the individual settings
are found in later sections.
Setting
Settings for client configuration
Address pool
Address (individual)
Deny service
Netmask
Gateway/Router
Name server
Hostname
Domain path
NTP Server
TFTP-server address
TFTP-server name
TFTP file
Lease-time
Other Server settings
Ping-check
Server-port
Client-port
Preempt static lease
22.1.2.1
Global
Subnet
Host
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
IP address assignment
The addresses can either be assigned dynamically from an address pool, or be
assigned statically depending on the client’s MAC, its DHCP client identifier, or
the port to which it is connected.
ˆ Address pool: For each subnet served it is possible to define a pool of
addresses for dynamic assignment. The default range is ”100-199”, e.g.,
10.10.2.100-10.10.2.199 on the 10.10.2.0/24 subnet.
It is possible to disable dynamic address allocation using the ”no pool”
syntax in the CLI. This is mostly useful in combination with fixed assignment.
538
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
ˆ Static lease: Instead of handing out addresses from a dynamic pool, the
WeOS DHCP server enables you to assign addresses with more fine grain
control. You can match these host specific assignments, based on the MAC
address of the client, the DHCP Client-ID provided by the client, or port to
which the client has connected (DHCP option 82).
The static lease matching method (MAC, Client-id, Option 82) can also be
used to deny clients an IP address. To specify this feature, use the keyword
”deny” instead of an IP address in the assignment command.
For more information on how to match incoming DHCP Requests to static
leases, see section 22.1.4.
22.1.2.2
Configuration Options other than IP address
In addition to IP address, the WeOS DHCP server allows you configure the following configuration options:
ˆ Netmask (DHCP Option 1): IP netmask is passed to the client in DHCP option 1. By default, the netmask is set to 255.255.255.0.
ˆ Router IP address (DHCP Option 3): The DHCP server will pass information
about what router (default gateway) the DHCP client should use. If you leave
this blank, the will automatically fill out a value likely to work for the client.
– Local clients: For DHCP requests originating on the local subnets, the
DHCP server will put its own IP address on that subnet as gateway IP
address.
– Remote clients: For DHCP requests originating on remote subnets, the
DHCP server will put the IP address of the relay agent as gateway IP
address.
The router/gateway IP is passed in DHCP option 3. By default, the gateway
setting is empty, i.e., the ”auto” behaviour described above is used.
As of WeOS v4.21.1 there is no way to hinder the DHCP server to send
the router/gateway IP address (option 3). This may change in future WeOS
releases.
ˆ DNS Server (DHCP option 6): It is possible to specify up to two DNS servers
to be passed to the DHCP client. If no DNS server is specified, the DHCP
server will fill in its own IP address as DNS server (the DHCP server unit will
© 2017 Westermo Teleindustri AB
539
Westermo OS Management Guide
Version 4.21.1-0
act as DNS forwarder and forward any (non-cached) incoming DNS requests
to the name-server(s) configured on the unit, see chapter 21).
As of WeOS v4.21.1 there is no way to hinder the DHCP server to send the
Domain Name Server option to the client. This may change in future WeOS
releases.
ˆ Hostname (DHCP Option 12): The DHCP server can be configured to pass
the hostname to the DHCP client.
ˆ Domain search path (DHCP option 15): The DHCP server can be configured
to pass a domain search path to the DHCP client. (Leaving the setting empty
implies that no domain search path is sent to the client.)
ˆ NTP Server (DHCP option 42): The DHCP server can be configured to pass up
to two NTP Servers to the DHCP client. (Leaving the setting empty implies
that no NTP server is sent to the client.)
ˆ Lease time (DHCP Option 51): The lease time can be configured in range
120-5256000 seconds or ”infinite”. It defaults to 864000 seconds (10 days).
ˆ TFTP server address: The DHCP server can be configured to pass the nextserver address siaddr in DHCP and BOOTP messages from the DHCP server,
i.e., the IP address of a TFTP server (or other type of file transfer server)
used by a BOOTP/DHCP client to retrieve a boot file (see below).
ˆ TFTP server name (DHCP Option 66): The DHCP server can be configured to
pass the next-server domain name. It can be used to inform BOOTP/DHCP
clients about their next-server to download a boot file, as an alternative to
the next-server address (see above).
The server name is typically passed within the sname field of a BOOTP/DHCP
message, but is instead sent as DHCP option 66 if option overloading applies
or if the client has requested DHCP option 66.
ˆ TFTP file (DHCP Option 67): The DHCP server can be configured to pass the
boot filename (as stored at the TFTP server).
The bootfile name is typically passed within the file field of a BOOTP/DHCP
message, but is instead sent as DHCP option 67 if option overloading applies
or if the client has requested DHCP option 67.
540
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
22.1.3
General DHCP Server settings
WeOS allows you to configure a set of general DHCP server settings. These are
advanced settings, and are primarily of interest to users with special demands.
The default values are sufficient in almost all use cases.
ˆ DHCP Server Listening UDP port: The DHCP server listens to UDP port 67 by
default (in-line with RFC2131[7]). It is possible to set the server port to a
different value. That may be of interest in some specific DHCP relay setups,
to avoid that the server receives packets directly from clients (in addition to
relayed packets).
Note
It is possible to configure the WeOS relay agent to forward DHCP messages to non-standard UDP ports on the server, see chapter 23.
ˆ DHCP Server Source UDP port (client port): The DHCP server will send packets with source UDP port 68 by default. It is possible to set the source UDP
port to a non-default value.
ˆ Enable/disable Duplicate Address Detection (ICMP Ping Check): Before a
DHCP server offers a client an address it will check that no-one is already
using that address. The server conducts this duplicate address detection
mechanism by attempting to ”ping” the IP address a couple of time to verify that it gets no response. Disabling ”ping check” can speed up address
assignment.
The ”ping check” mechanism is recommended for robustness and is enabled by default. Only consider disabling ”ping-check” if you are using static
leases.
Warning
The WeOS DHCP server does not store the lease table in persistent
storage. Disabling ”ping check” can therefore lead to situations where
a server reboot causes some host to be assigned an address, which
was already assigned to (and in use by) another host.
ˆ Preemption of static leases: When replacing a unit, which has been assigned
an address according to a static lease (e.g., matching on option 82 or clientid) it is possible to preempt the existing lease.
– Option 82 preemption: When assigning an address to a unit connecting
© 2017 Westermo Teleindustri AB
541
Westermo OS Management Guide
Version 4.21.1-0
via a specific port (option 82), a WeOS DHCP server assumes that an
existing lease should be preempted if it receives a new DHCP Request
from that port. That is, a WeOS DHCP server will always preempt static
leases when option 82 is used in the matching criteria; this also applies
when combining option 82 with client-id as match criteria. Before handing out the address to a new client, the server will run a ping-check to
verify that the old client is not active (given that ping-check is enabled).
– Client-id preemption: When matching is done based on client-id it is
possible to configure if the static lease should be preempted or not.
The server will still conduct a ping-check before reassigning the address
(see comment on ”Option 82 preemption” above).
– MAC address preemption: The DHCP server will treat requests from the
same MAC address as being the same client unit. Thus, preemption is
not relevant for static leases matching on MAC addresses.
22.1.4
Matching Static Lease Assignments
You can define specific configuration settings for different hosts (static leases).
Incoming DHCP requests can be matched to a static lease based on several criteria:
ˆ Client MAC: You can reserve a specific address to a client with a certain MAC
address.
ˆ Client identifier (option 61): You can reserve a specific address to a client including a certain client-identifier in its DHCP messages (DHCP option 61[1]).
In the DHCP server, you can specify the client-id as a hexadecimal sequence
(e.g., ”01485b392f34bc”) or as a text string such as ”foobar”.
Note
If the client-id is specified as a text string, it would match a DHCP option 61 holding a hexadecimal sequence of the corresponding ASCII
numbersa , e.g., ”foobar” would match an option 61 holding value
”666f6f626172” (hex). In addition, the WeOS DHCP server would accept the same string with a prepended ”00”; thus, ”foobar” would also
match an option 61 holding value ”00666f6f626172” (hex).
a American Standard Code for Information Interchange (ASCII), see e.g. http://en.
wikipedia.org/wiki/ASCII (accessed May 2009).
542
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Note
A WeOS unit acting as DHCP client sends a client-id consisting of a
concatenation of ’01’ and ’the MAC address of the interface’, see section 21.2.9.
ˆ Connected Port (option 82): The server can be configured to assign a specific
address to the client connected to a certain switch port (”one IP per port”).
This is useful when you wish to replace a client unit, such as a CCTV camera,
and ensure that the new unit gets the same IP as the replaced unit.
ˆ Combination of connected port (option 82) and client identifier (option 61):
You may to assign the address based on the combination of option 82 and 61.
This can be useful in situations where you have multiple hosts connected to
the same port of the (WeOS) relay agent (e.g., via a external hub). As long
as the client-id is unique per port on the (WeOS) relay agent, the combination of client-id (option 61) and connected port (option 82) can uniquely
identify the host.
ˆ Other combinations: Use of other combinations for matching is discouraged.
If other combinations are used, the behaviour is undefined.
As described in chapter 23, DHCP relay agents can add information to identify
the client’s port in a relay information option (DHCP option 82[33]). The DHCP
server can then extract relevant information (circuit-id and remote-id) and use
that when assigning the IP address.
The WeOS DHCP server allows for flexible specification of circuit-id and remote-id
(both as hexadecimal sequences and text strings), enabling it to work with relay
agents of various vendors. E.g., to make the DHCP server hand out a specific IP
address to a client unit attached to WeOS Relay Agent with default settings, the
DHCP server can be configured as follows:
ˆ Circuit-id: If the client is supposed to connect to Ethernet port 2, then specify
”Eth2” (string) for the circuit-id. If a slotted WeOS product is used, then
specify e.g., ”Eth3/5” for Ethernet port 5 on slot 3.
ˆ Remote-id: The remote-id is optional, but needed to distinguish between
relay agents on the same subnet. A WeOS relay agent defaults to using its
base MAC2 address as remote-id.. E.g., specify ”00077c8209d0” (hex) for a
WeOS relay agent with base MAC 00:07:7c:82:09:d0.
Note: to assign IP addresses per (local) ports on the DHCP server itself in
2 To
find the base MAC of your WeOS unit, see sections 4.4.2 (Web) or 7.3.2 (CLI).
© 2017 Westermo Teleindustri AB
543
Westermo OS Management Guide
Version 4.21.1-0
WeOS v4.21.1, you will need to setup a Relay Agent on the same unit (see
section 22.1.5).
A note on matching precedence order
A client request associated with a subnet served by the DHCP server may
match multiple static lease match entries, e.g., it may match one entry
based on its MAC and another entry based on the port (option 82). Such
ambiguities may be mitigated by avoiding use of different types of match
criteria (e.g., only use ”option 82”). When a client request matches multiple
static lease entries, they are evaluated in the following precedence order:
ˆ Matching port (option 82) and client-id (option 61) (first)
ˆ Matching port (option 82)
ˆ Matching client-id (option 61)
ˆ Matching MAC address
ˆ Assign address from pool (last)
22.1.5
Running a DHCP server and relay agent on the same unit
There are situations when you wish to run a DHCP relay agent (chapter 23) on
the same WeOS unit as your DHCP server.
ˆ IP per port on DHCP server unit: Section 22.1.5.1 describes how to use a
DHCP server and a relay agent to assign IP addresses per port on the DHCP
server unit itself.
ˆ Non-”DHCP snooping” relay agents in switched topologies: Section 22.1.5.2 explains how to handle non-”DHCP snooping” relay agents in
switched (as opposed to routed) topologies. (An alternative approach is to
let the DHCP server listen to a non-default UDP port, see section 22.1.3.)
22.1.5.1
IP per port on local DHCP server ports
With DHCP option 82, a relay agent can inform the DHCP server which port
(circuit-id) the client is connected to, thereby enabling the server to assign IP
addresses per port. In WeOS, the same approach is used when you wish to hand
out IP addresses per port on the DHCP server’s local ports.
Fig. 22.2 illustrates an example where the WeOS unit is configured to hand out
addresses on interface ”vlan2” (subnet 192.168.5.0/24). Regular hosts, such as
544
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
WeOS Router
DHCP Server
Relay Agent
1
2
3
4
5
6
VLAN 1
192.168.2.0/24
VLAN 2
192.168.5.0/24
.100
PC
.49
CCTV
Figure 22.2: Running both a DHCP Server and a DHCP Relay Agent on the same
unit enables you to assign IP address per port on the DHCP server unit.
the PC, will be assigned their IP addresses from an address pool, but the unit
attached to port 6 should always be assigned IP address 192.168.5.49. This
can be achieved by configuring a DHCP relay agent on interface ”vlan2”, and
to instruct the relay agent to forward DHCP request to the local DHCP server
(address ”127.0.0.1”). Relevant parts of the WeOS configuration is listed in
fig. 22.3.
The WeOS DHCP relay will by default pass its base-MAC address3 as remote-id
(”00:07:7c:00:30:b0” in the configuration example in fig. 22.3.). As the baseMAC is unit specific, this setting will not work if you wish to replace the unit,
but keep the same configuration file. In such situations, using ”system-name” or
”ip” as remote-id is recommended, see sections 23.2.1 (Web) and 23.3.9 (CLI)
for more information. An example using the system name as remote-id is given
in fig. 22.4.
3 To
find the base MAC of your WeOS unit, see sections 4.4.2 (Web) or 7.3.2 (CLI).
© 2017 Westermo Teleindustri AB
545
Westermo OS Management Guide
Version 4.21.1-0
Example
dhcp-server
subnet 192.168.5.0/24
pool 192.168.5.100 192.168.5.199
lease-time 864000
netmask 255.255.255.0
no gateway
no domain
end
host 1
match option82 circuit-id string "Eth6" remote-id hex 00:07:7c:00:30:b0
address 192.168.5.49
end
end
dhcp-relay
iface vlan2
server 127.0.0.1
option82 discard
end
Figure 22.3: Configuration example with DHCP relay and server on same unit,
here with base-MAC address as Option82 Remote-ID.
Example
system
hostname foobar
end
dhcp-server
subnet 192.168.5.0/24
pool 192.168.5.100 192.168.5.199
lease-time 864000
netmask 255.255.255.0
no gateway
no domain
end
host 1
match option82 circuit-id string "Eth6" remote-id string "foobar"
address 192.168.2.49
end
end
dhcp-relay
iface vlan2
server 127.0.0.1
option82 discard
remoteid-type system-name
end
Figure 22.4: Configuration example with DHCP relay and server on same unit,
here with system hostname as Option82 Remote-ID.
546
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
22.1.5.2
Handling non-snooping relay agents in switched topologies
As described in section 23.1.4, use of relay agents to add option 82 information
in switched topologies is challenging if the relay agents do not support DHCP
snooping. A (broadcast) DHCP message from a client will then result in two messages being forwarded towards the DHCP server - one relayed message including
option 82 information, and one regular message being switched and lacking option 82.
10.1.1.1/24
10.1.1.2/24
Non−’’DHCP snooping’’
Switch (e.g., "Gbit−")
module in RedFox)
WeOS Router/Switch
DHCP Server
Drop DHCP packets
lacking option 82
coming in on port 6
With
Option 82
Relay Agent
5
6
1
2
3
4
Relay Agent
5
Without
Option 82
DHCP Msg
6
1
2
3
4
.44
CCTV
Figure 22.5: A non-”DHCP snooping” relay agent (right unit) will likely result in
multiple ”copies” of the DHCP messages. This can be handled by running a DHCP
Relay Agent also the DHCP server unit (left unit).
Fig. 22.5 illustrates the situation. All ports are assumed to be on the same VLAN
(e.g., VLAN 1)
1. A broadcast DHCP message is sent by the PC on port 1 of the non-snooping
switch. That packet is forwarded onto all ports on the same VLAN including
port 5 towards the DHCP server.
2. The packet is also processed by the relay agent process, which adds option 82 information and relays the message (unicast) towards the DHCP
server.
3. If both DHCP requests would reach the DHCP server, it is likely that the PC
will be handed an address from the pool rather than an address dedicated
for that specific port. Or possibly the PC will get multiple responses to its
request.
In WeOS you can handle this by running a DHCP relay agent on the DHCP
server unit. The relay agent can be configured to drop DHCP packets not
© 2017 Westermo Teleindustri AB
547
Westermo OS Management Guide
Version 4.21.1-0
including option 82, thus only the relayed packet will be forwarded to the
DHCP server process.
Below (page 548) sample configurations for the DHCP server and DHCP relay
agent units are shown. The CCTV connected to port 14 of the (non-snooping)
relay agent should be assigned IP address 10.1.1.44/24.
Hint
An alternative approach is to let the DHCP server listen to a non-default UDP
port, see section 22.1.3. Then the DHCP relay agent must be configured to
send to this UDP port when relaying packets to the server.
Example
-- DHCP Server Unit (IP 10.1.1.1/24)
dhcp-server
subnet 10.1.1.0/24
pool 10.1.1.100 10.1.1.199
lease-time 864000
netmask 255.255.255.0
gateway 10.1.1.1
no domain
end
host 1
match option82 circuit-id string "Eth1" remote-id string "10.1.1.2"
address 10.1.1.44
end
end
dhcp-relay
iface vlan1
server 127.0.0.1
option82 discard
remoteid-type ip
port 6
option82 require
end
end
-- DHCP Relay Agent Unit (IP 10.1.1.2/24)
dhcp-relay
iface vlan1
server 10.1.1.1
option82 discard
remoteid-type ip
end
4 If the relay agent unit is a RedFox Industrial, the port labels would be written in slot/id form (1/1,
1/2, etc.). The server configuration would then reflect this, e.g., ”match option82 circuit-id
string "Eth1/2" remote-id string "10.1.1.2"” if the CCTV is connected to port 1/2.
548
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
22.2
Configuring DHCP Server Settings via the Web
The Web interface provides management of DHCP Server.
22.2.1
DHCP Server settings
Menu path: Configuration ⇒ Network (IP) ⇒ DHCP-Server
© 2017 Westermo Teleindustri AB
549
Westermo OS Management Guide
Version 4.21.1-0
Enabled
Subnets
Static DHCP
Check the box to enable the DHCP server. If you have a
JavaScript enabled browser the other settings will not be displayed unless you check this box.
Lists the configured DHCP subnets To add a Subnet click on
the New subnet button below the table. Click on the Edit icon
(
) to edit the settings for a specific Subnet. Clicking the edit
icon or the ”New Subnet” button will take you the ”Create/Edit
DHCP Subnet” page, see section 22.2.2.
The static leases for this subnet. To add a static lease, click
the New Lease button. Click on the Edit icon (
) to edit the
settings for an existing lease. Clicking the edit icon or the
”New Lease” button will take you to the ”Create/Edit DHCP
Host Settings” page, see section 22.2.3.
550
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
22.2.1.1
Advanced DHCP Server settings
Menu path: Configuration ⇒ Network (IP) ⇒ DHCP-Server ⇒ Advanced Settings
Ping Check
Boot Server
Address
Boot Server
Name
Boot File
Enables/disables the ICMP ping check. By default the DHCP
server will check that an address is not in use before offering it
to a client. In some rare cases it may be useful to disable this.
Default enabled
IP address for server from which the client should retrieve the
boot file.
DNS name for server from which the client should retrieve the
boot file.
Name of the boot file to retrieve from boot server.
© 2017 Westermo Teleindustri AB
551
Westermo OS Management Guide
Version 4.21.1-0
22.2.2
Create/Edit DHCP Subnet Settings
Menu path: Configuration ⇒ Network (IP) ⇒ DHCP-Server ⇒ New Subnet
Menu path: Configuration ⇒ Network (IP) ⇒ DHCP-Server ⇒
(Subnet)
On this page you can change the settings for the Subnet.
Address Pool
Lease Time
552
IP address pool from which the DHCP server will hand out
leases
DHCP address lease time (seconds) for addresses handed
out to DHCP clients
Continued on next page
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Netmask
Default Gateway
Name Servers
NTP Servers
Domain
Continued from previous page
The netmask option handed to DHCP clients.
The IP default gateway (default router) option handed to
DHCP clients.
The (DNS) name server option handed to DHCP clients.
The time server (NTP) option handed to DHCP clients.
Domain name search path option handed to DHCP clients
© 2017 Westermo Teleindustri AB
553
Westermo OS Management Guide
Version 4.21.1-0
22.2.3
Create/Edit DHCP Host Settings
Menu path: Configuration ⇒ Network (IP) ⇒ DHCP-Server ⇒ New Lease
Menu path: Configuration ⇒ Network (IP) ⇒ DHCP-Server ⇒
(Static DHCP)
On this page you can change the settings for the Host.
Lease
554
IP address for this lease. If left empty the DHCP server
will prohibit the host to be served.
Continued on next page
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Rule 1-3
Hostname
Domain
Lease Time
Default Gateway
Name Servers
NTP Servers
Boot Server
Address
Boot Server
Name
Boot File
Continued from previous page
Specify up to 3 match types for this host.
Sets the hostname that the server will hand out to this
host.
Specify the domain name search path option for this host.
DHCP address lease time (seconds) for this host.
The IP default gateway (default router) option handed to
this host.
The (DNS) name server option handed to this host.
The time server (NTP) option handed to this host.
Address of the TFTP server to hand out to this host.
Domain name of the TFTP server to hand out to this host.
File (at the TFTP server) to hand out to this host.
© 2017 Westermo Teleindustri AB
555
Westermo OS Management Guide
Version 4.21.1-0
22.3
Configuring DHCP Server Settings via the CLI
Command
Configure DHCP Server
[no] dhcp-server
[no] enable
[no] ping-check
[no] server-port <UDPPORT>
[no] client-port <UDPPORT>
[no] server-address <IPADDR>
[no] server-name <DOMAINNAME>
[no] file <FILENAME>
[no] host [INDEX]
[no] match <mac <MACADDR> |
clientid <hex|string> <CLIENTID> |
option82 [remote-id <hex|string>
<REMOTEID>] |
option82 [circuit-id <hex|string>
<CIRCUITID>]
[no] address <IPADDR|deny>
[no] gateway <IPADDR>
[no] name-server <IPADDR>
[no] hostname <HOSTNAME>
[no] domain <DOMAINNAME>
[no] ntp-server <IPADDR>
[no] lease-time <infinite|120-5256000>
[no] server-address <IPADDR>
[no] server-name <DOMAINNAME>
[no] file <FILENAME>
[no] preempt
[no] subnet <IPADDR[/LEN] | IPADDR [MASK]>
[no] netmask <NETMASK>
[no] pool <IPADDR_START>
<NUM|IPADDR_END>
[no] gateway <IPADDR>
[no] name-server <IPADDR>[,<IPADDR>]
556
Default
Section
Disabled
Enabled
Enabled
67
68
Disabled
Disabled
Disabled
1
Section
Section
Section
Section
Section
Section
Section
Section
Section
Section
22.3.1
22.3.2
22.3.3
22.3.4
22.3.5
22.3.6
22.3.7
22.3.8
22.3.9
22.3.10
Disabled
Disabled
Disabled
Disabled
Disabled
Disabled
864000
Disabled
Disabled
Disabled
Disabled
/24
Section
Section
Section
Section
Section
Section
Section
Section
Section
Section
Section
Section
Section
Section
22.3.11
22.3.12
22.3.13
22.3.14
22.3.15
22.3.16
22.3.17
22.3.6
22.3.7
22.3.8
22.3.18
22.3.19
22.3.20
22.3.21
Auto1
Empty2
Section 22.3.12
2
Empty
Section 22.3.13
Continued on next page
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Command
[no] domain <DOMAINNAME>
[no] ntp-server <IPADDR>
[no] lease-time <infinite|120-5256000>
Continued from previous page
Default Section
Disabled Section 22.3.15
Disabled Section 22.3.16
864000
Section 22.3.17
View DHCP Server Status
show dhcp-clients
22.3.1
Section 22.3.22
Manage DHCP Server
Syntax [no] dhcp-server
Context Global Configuration context
Usage Create, modify or remove a DHCP Server.
Enter DHCP server context. If this is a new DHCP server, the DHCP server
is created. As a side-effect, a caching (DNS) name server is started, which
forwards incoming DNS requests to the DNS server configured for the switch
(see chapter 21).
Use ”no dhcp-server” to remove an existing DHCP server.
Use ”show dhcp-server” to list all settings of a DHCP server. Alternatively,
you can run the ”show” command from within the DHCP server context.
Default values Disabled (No DHCP server configured)
22.3.2
Disable DHCP Server
Syntax [no] enable
Context DHCP Server Configuration context
1A
pool may be created automatically. See Section 22.3.21.
values have special meaning here. See Section 22.3.12 and Section 22.3.13.
2 Empty
© 2017 Westermo Teleindustri AB
557
Westermo OS Management Guide
Version 4.21.1-0
Usage Enable/disable the DHCP server. Useful to disable a fully setup DHCP
server before deployment, the configuration will remain dormant while disabled.
Use ”no enable” to disable the DHCP server (without losing the DHCP server
configuration) and ”enable” to enable the DHCP server.
Use ”show enable” to show whether the DHCP server is configured enabled
or disabled.
Default values Enabled
22.3.3
Disable ICMP ”ping” Check
Syntax [no] ping-check
Context DHCP Server Configuration context
Usage Enable/disable the ICMP ping check. By default the DHCP server will
check that an address is not in use before offering it to a client. In some
rare cases it may be useful to disable this.
Use ”no ping-check” to disable the ping-check mechanism, and ”ping-check”
to enable it.
Run ”show ping-check” to show whether the ping-check mechanism is configured enabled or disabled.
Default values Enabled
22.3.4
DHCP Server Listening UDP port
Syntax [no] server-port <UDPPORT>
Context DHCP Server Configuration context
Usage Set DHCP Server listening (UDP) port in range 1..65535. By default the
server listens to UDP port 67. Use ”server-port UDPPORT” to set a nondefault UDP port to listen on. See also section 23.3.5 for the corresponding
DHCP relay agent setting.
Use ”no server-port” to reset to default value (port 67).
Use ”show server-port” to show current server-port settings.
558
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Default values 67
22.3.5
DHCP Server Source/Client UDP port
Syntax [no] client-port <UDPPORT>
Context DHCP Server Configuration context
Usage Set DHCP Server source (UDP) port in range 1..65535. By default the
server sends DHCP messages with source UDP port 68. Use ”client-port
UDPPORT” to set a non-default UDP port to send from.
Use ”no client-port” to reset to default value (port 68).
Use ”show client-port” to show current client-port settings.
Default values 68
22.3.6
Next Server Address – BOOTP ”siaddr”
Syntax [no] tftp-server <IPADDR>
Context DHCP Server Configuration or DHCP Server Host Configuration context
Usage Set the next-server address siaddr in DHCP and BOOTP messages from
the DHCP server, i.e., the IP address of a TFTP server (or other type of file
transfer server) used by a BOOTP/DHCP client to retrieve a boot file.
Note
Using the ”tftp-server” command in DHCP Server Configuration will
apply to all DHCP messages from the server. Using the ”tftp-server”
command in DHCP Server Host Configuration will apply only to that
static lease entry.
Use ”no tftp-server” to remove a configured next-server address.
Use ”show tftp-server” to show the next-server setting.
Default values Disabled
© 2017 Westermo Teleindustri AB
559
Westermo OS Management Guide
Version 4.21.1-0
22.3.7
Next Server Name – BOOTP ”sname”, DHCP Option 66
Syntax [no] tftp-server-name <DOMAINNAME>
Context DHCP Server Configuration or DHCP Server Host Configuration context
Usage Set the next-server domain name. It can be used to inform BOOTP/DHCP
clients about their next-server to download a boot file, as an alternative to
the next-server address (see section 22.3.6).
The server name is typically passed within the sname field of a BOOTP/DHCP
message, but is instead sent as DHCP option 66 if option overloading applies
or if the client has requested DHCP option 66.
Note
Using the ”tftp-server-name” command in DHCP Server Configuration will apply to all DHCP messages from the server. Using the
”tftp-server-name” command in DHCP Server Host Configuration will
apply to that static-lease entry.
Use ”no tftp-server-name” to remove a configured next-server name.
Use ”show tftp-server-name” to show the next-server name setting.
Default values Disabled
22.3.8
Bootfile Name – BOOTP ”file”, DHCP Option 67
Syntax [no] bootfile <FILENAME>
Context DHCP Server Configuration or DHCP Server Host Configuration context
Usage Set the boot filename (as stored at the TFTP server).
The bootfile name is typically passed within the file field of a BOOTP/DHCP
message, but is instead sent as DHCP option 67 if option overloading applies
or if the client has requested DHCP option 67.
Note
Using the ”bootfile” command in DHCP Server Configuration will apply to all DHCP messages from the server. Using the ”bootfile” command in DHCP Server Host Configuration will apply to that static-lease
entry.
560
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Use ”no bootfile” to remove a configured bootfile name.
Use ”show bootfile” to show the next-server name setting.
Default values Disabled
22.3.9
Configure Host Entry (Static Lease)
Syntax [no] host [INDEX]
Context DHCP Server Host Configuration context
Usage Enter the DHCP Server Host Configuration to specify host specific DHCP
Server settings, i.e., static lease settings. This is typically used to configure
a static lease based on MAC, Client-ID or port ID (i.e., DHCP Option 82). Up
to 1024 can be configured. Each entry is given an index (default 1), e.g.,
”host 3” will enter the DHCP Server Host Configuration for entry number 3;
the entry will be created if it does not yet exist.
Use ”no host” to remove all configured static lease entries, and use ”no
host <INDEX>” to remove a specific static lease entry (e.g. ”no host 3”).
Use ”show host” to show a list configured static lease entries, and use
”show host <INDEX>” to show information on a specific static lease. Alternatively, you can run the ”show” command within the DHCP Server Host
Configuration context of that specific static lease.
Default values Default index is 1.
22.3.10
Configure Static Lease Match Setting
Syntax [no] match <mac <MACADDR> | clientid <hex|string> CLIENTID> |
option82 [remote-id <hex|string> <REMOTEID>] |
option82 [circuit-id <hex|string> <CIRCUITID>]>
Context DHCP Server Host Configuration context
Usage Specify the match type (mac, clientid or option82) to identify the host for
this entry, e.g., ”match mac 12:34:56:78:9a:bc:de”.
Use ”no match mac <MACADDR>”, ”no match clientid <hex|string> <CLIENTID>”,
or ”no match option82 <...>” to remove a specific match setting. Use
”no match” to remove all match settings.
© 2017 Westermo Teleindustri AB
561
Westermo OS Management Guide
Version 4.21.1-0
Use ”show match” to show the current match setting for this static lease.
Default values Not applicable. section 22.3.19.
22.3.11
Configure Host IP Address or Deny Service
Syntax [no] address <IPADDRESS|deny>
Context DHCP Server Host Configuration context
Usage Specify the IP address to assign to this host, e.g., ”address 192.168.1.51”.
Note
To hand out the specified address (e.g., ”192.168.1.51”) the DHCP
server must also be configured to serve the associated IP subnet, see
section 22.3.19 for information on the ”subnet” command. Other IP
settings (netmask, default gateway, etc.) will be inherited from settings
of the associated subnet.
Use ”address deny” to prohibit the host to be served by this DHCP server.
A host must either be assigned an IP address or explicitly be denied an
address. ”no address” is not a valid setting, i.e., then the host entry will
not be activated.
Use ”show address” to show the current address setting.
Default values None
22.3.12
Configure DHCP Server Default Gateway Option
Syntax [no] gateway <IPADDRESS>
Context DHCP Server Subnet Configuration or DHCP Server Host Configuration
context
Usage Specify the IP default gateway (default router) option for leases handed
to DHCP clients. A single default gateway can be specified. If no default
gateway is specified, the switch IP address on this interface will be provided
in the default gateway option for local DHCP clients (that is, the switch will
562
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
act as default gateway for hosts on local subnets), or the DHCP relay agent
IP address for DHCP requests on remote subnets.
Note
When acting as router for local DHCP clients, please remember to enable routing on this unit (chapter 21) and enable appropriate NAT and
firewall rules if necessary (chapter 31).
Use ”no gateway” to remove any statically configured default gateway option.
Use ”show gateway” to list the gateway option settings.
Default values Empty, this means that the switch IP address on this interface
will be provided in the default gateway option.
22.3.13
Configure DHCP Server Name Server Option
Syntax [no] name-server <IPADDRESS>[,<IPADDRESS>]
Context DHCP Server Subnet Configuration or DHCP Server Host Configuration
context
Usage Specify name server (DNS) options for leases handed to DHCP clients.
Up to two DNS name servers can be specified, either as comma separated
IP addresses on the command line, or by repeating the command for each
address.
Use ”no name-server” to remove all configured name server DHCP options.
If no name server is specified, the switch IP address on this interface will
be provided in the name server option (that is, the switch will act as DNS
name server for hosts on this interface. In this case, the switch will act as
a caching name server and forward any (non-cached) incoming requests to
the name-server(s) configured on the switch, see chapter 21).
Use ”show name-server” to list DNS name server option settings.
Default values Empty, this means that the switch IP address on this interface
will be provided in the name server option.
© 2017 Westermo Teleindustri AB
563
Westermo OS Management Guide
Version 4.21.1-0
22.3.14
Configure Host Hostname
Syntax [no] hostname <HOSTNAME>
Context DHCP Server Host Configuration context
Usage Specify the hostname e.g., ”hostname doorcamera”.
Sets the hostname that the server will hand out to this host. The command
”no hostname” will remove the hostname from this host.
Use ”show hostname” to show the current hostname setting.
Default values Disabled
22.3.15
Configure Domain Name Option
Syntax [no] domain <DOMAIN>
Context DHCP Server Subnet Configuration or DHCP Server Host Configuration
context
Usage Specify the domain name search path option for leases handed to DHCP
clients. A single domain name option can be specified.
Use ”no domain” to disable this option.
Use ”show domain” to list domain name option settings.
Default values Disabled, the domain name option will not be used.
22.3.16
Configure NTP Server Option (DHCP Option 42)
Syntax [no] ntp-server <IPADDR>
Context DHCP Server Subnet Configuration or DHCP Server Host Configuration
context
Usage Specify the NTP-server option (DHCP option 42) for leases handed to
DHCP clients, e.g., ”ntp-server 192.168.1.3”. Up to two NTP servers can
be specified.
Use ”no ntp-server <IPADDR>” to remove a specific NTP server, or ”no
ntp-server” to disable all NTP server options.
564
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Use ”show ntp-server” to list NTP-server option settings.
Default values Disabled, the NTP-server option will not be used.
22.3.17
Configure DHCP Server Lease Time
Syntax [no] lease-time <infinite|120-5256000>
Context DHCP Server Subnet Configuration or DHCP Server Host Configuration
context
Usage Specify the DHCP address lease time (seconds) for addresses handed out
to DHCP clients.
Use values in range ”120-5256000” to give lease-time in seconds (e.g.,
”lease-time 600” for 10 minutes), and use the special keyword ”infinite”
to denote an infinite lease time.
Use ”no lease-time” to reset the lease time setting to its default value.
Use ”show lease-time” to show the current lease-time setting.
Default values 864000 seconds (i.e., 10 days)
22.3.18
Lease preemption (Client-ID)
Syntax [no] preempt
Context DHCP Server Host Configuration context
Usage Enable/disable lease preemption. This setting only applies when the
static lease matches on client-id (option 61), see section 22.1.3.
Use command ”preempt” to enable lease preemption for this host entry.
Use command ”no preempt” to disable lease preemption for this host entry.
Use command ”show preempt” to show the current lease preemption setting for this host.
Default values Disabled (i.e., preemption is disabled for leases matching on
client-id)
© 2017 Westermo Teleindustri AB
565
Westermo OS Management Guide
Version 4.21.1-0
22.3.19
Configure DHCP Server Subnet
Syntax [no] subnet <IPADDR[/LEN] | IPADDR [NETMASK]>
Context DHCP Server Configuration context
Usage Specify a subnet for which the DHCP server will hand out IP addresses,
and enter the DHCP Server Subnet Configuration for that subnet. Optionally,
the subnet netmask can be specified as a prefix length or as a netmask,
with ”/24” (”255.255.255.0”) as default. It can later be changed with the
”netmask” command, see section 22.3.21.
Use ”no subnet” to remove all configured subnets, and use ”no subnet
IPADDR” to remove a specific subnet.
Use ”show subnet” to show a list configured subnets for the DHCP server,
and use ”show subnet IPADDR” to show information on a specific subnet
(alternatively, you can run the ”show” command within the DHCP Server
Configuration context of that specific subnet).
The DHCP server can handle up to 64 subnets.
Default values Default prefix length is 24 (i.e., netmask 255.255.255.0).
22.3.20
Configure DHCP Subnet Netmask
Syntax [no] netmask <NETMASK>
Context DHCP Server Subnet Configuration context
Usage Specify/modify the netmask for the subnet to serve, e.g.,
”netmask 255.255.128.0”.
Use ”no netmask” to reset the netmask to its default value.
Use ”show netmask” to show the current netmask setting.
Default values The netmask defaults to ”255.255.255.0”, however, a different
netmask can be specified in the ”subnet” command, see section 22.3.19.
22.3.21
Configure DHCP Server Address Pool
Syntax [no] pool <IPADDRESS_START> <NUM|IPADDRESS_END>
566
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Context DHCP Server Subnet Configuration context
Usage Specify the IP address pool from which the DHCP server will hand out
leases. The end of the address range can be specified as an IP address
(”IPADDRESS_END”), or as a number (”NUM”). ”NUM” specifies the number of
addresses in the pool, thus ”IPADDRESS_END” is computed as
”PADDRESS_START + NUM − 1”.
Use ”no pool” to disable dynamic address assignment. When disabled,
only static host entries are allowed in the range defined by the subnet itself
and the netmask option.
Use ”show pool” to see the IP addresses in the pool.
Default values A pool based on the configured subnet is automatically setup
when creating a new DHCP subnet.
22.3.22
Show list of current DHCP clients
Syntax show dhcp-clients
Context Admin Exec context
Usage Show list of current DHCP clients.
Default values Not applicable
Example
example:/#> show dhcp-clients
Lease Time MAC Address
IP Address
Hostname
Client ID
===============================================================================
864000
00:07:7c:8a:e2:41 192.168.2.109
01:00:07:7c:8a:e2:41
*
example:/#>
© 2017 Westermo Teleindustri AB
567
Westermo OS Management Guide
Version 4.21.1-0
Chapter 23
DHCP Relay Agent
This chapter describes WeOS DHCP Relay Agent support. For information on
WeOS DHCP Server support, see chapter 22.
DHCP Relay Agents relay DHCP messages between DHCP clients on a local LAN
to a central DHCP Server, usually located on a remote network. The two most
common reasons for using DHCP relay agents are:
ˆ Centralised management: Deploying and managing a DHCP server on every
LAN in your network is cumbersome. By use of relay agents, a central DHCP
server can be used, and the management effort is substantially reduced.
Furthermore, if the relay agent is located in a router or switch on the local
LAN, there is no additional equipment cost.
ˆ Assigning IP address per port (DHCP Option 82): In some topologies, you
may wish to assign IP addresses based on the switch port a DHCP client
connects to. By running a DHCP Relay Agent in the local switch/router, it
can include port information when forwarding the DHCP messages (DHCP
Option 82).
For redundancy purposes, the WeOS DHCP Relay Agent enables you to specify up
to two DHCP servers, to which the Relay Agent forwards incoming DHCP requests.
In case you wish to hand out addresses per port on the DHCP server unit (as
opposed to the DHCP relay agent), WeOS allows you to achieve this by running
a relay agent on the DHCP server unit, see the chapter on DHCP server (section 22.1.5).
568
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
23.1
Overview of DHCP Relay Agent Support
The table below lists the features available in the WeOS DHCP Relay Agent.
Feature
General DHCP Relay settings
Enable/disable Relay Agent
Define interfaces to serve
DHCP server IP address
DHCP server UDP port
DHCP Option 82
Enable/Disable DHCP Option 82
Default Policy
Default Circuit-ID type
Remote-ID
DHCP Proxy Mode
Force DHCP Option 54
Per-Port DHCP Relay settings
Enable/Disable DHCP Relay
DHCP Option 82
Policy
Circuit-ID type
23.1.1
Web
CLI
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
General Description
X
Section 23.1.1
-”-”-”Section 23.1.2
-”-”-”-”Section 23.1.3
-”-
X
X
Section 23.1.4
X
X
X
X
Section 23.1.2
-”-
Introduction to DHCP Relay Agents
One of the main reasons for using DHCP relay agents is to simplify DHCP management in larger infrastructures. Instead of deploying and managing a DHCP
server on every LAN, a DHCP relay agent present on the LAN can forward DHCP
messages between local DHCP clients, and a central DHCP server.
Fig. 23.1 can be used to illustrate the use of DHCP relays and a central DHCP
server.
ˆ (V)LAN interfaces: The DHCP relay agents (here RA1-RA3) serve DHCP clients
(here PC1-PC6) on the local LANs. A DHCP relay can serve a single LAN (Relay Agent 1 & 3) or multiple LANs (Relay Agent 2). In WeOS the LANs to
serve is selected by configuring which (VLAN) network interfaces the relay
agent should listen on.
© 2017 Westermo Teleindustri AB
569
Westermo OS Management Guide
Version 4.21.1-0
DHCP pools:
A: 192.168.0.100−150/24, gw 192.168.0.1
B: 192.168.1.100−150/24, gw 192.168.1.1
C: 192.168.2.100−150/24, gw 192.168.2.1
D: 192.168.3.100−150/24, gw 192.168.3.1
DHCP
Server
192.168.100.1
Intranet/Internet
DHCP
Relay
Agent
(RA1)
.1
192.168.0.0/24
PC1
DHCP
Relay
Agent
(RA2)
.1
.1
192.168.1.0/24
PC2
PC3
DHCP
Relay
Agent
(RA3)
.2
Router
.1
192.168.2.0/24
PC4
192.168.3.0/24
PC5
PC6
Figure 23.1: Sample topology where DHCP relay agents serve local DHCP clients,
and forwards DHCP requests to/from a central DHCP server.
ˆ DHCP Servers: The relay agent must also know where to forward the DHCP
requests from the local PCs, i.e., the relay agent must be configured with IP
address of the DHCP server (here 192.168.100.1). As of WeOS v4.21.1, the
relay agent can be configured with up to two DHCP servers. When configuring two DHCP servers, the DHCP relay will forward the DHCP requests to
both servers, thereby providing redundancy.
DHCP servers listen to UDP port 67 by default. It is possible configure the
WeOS relay agent to forward packets to a different port on the server, see
also sections 23.1.5 and 22.1.3.
ˆ Address pools: The DHCP server will in turn be configured with appropriate
address pools (here denoted A-D), from which it can hand out addresses to
the local PCs.
When a DHCP relay agent receives a DHCP request from a PC, it will add its
local IP address into the giaddr field of the DHCP message when forwarding
it to the server (e.g., RA1 will set giaddr to 192.168.0.1) when forwarding
requests from PC1 to the DHCP server). Based on the giaddr, the DHCP
server can distinguish which pool to hand out address from (here ”A”).
570
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
The DHCP server should also be configured with other relevant settings, e.g.,
default gateway, lease times, etc. (see chapter 22).
ˆ Running relay agents on routers or switches: Relay agents can be run as
dedicated servers (RA3), but are typically located inside the local routers
(RA1 and RA2). By running the relay agents inside the routers, deployment and management costs are reduced, since no additional equipment
is needed.
Although not shown in fig. 23.1, it is also possible to run relay agents on
(layer-2) switches. This is useful when you wish to assign IP addresses based
on the physical port the PC connects to (see section 23.1.2 for information
on DHCP Option 82). In such use cases, you may also wish to run several
relay agents within the same LAN – section 23.1.4 provides more information
on running relay agents in switched networks.
As of WeOS v4.21.1, it is only possible to run a single relay agent instance per
WeOS unit. This is no major limitation, but implies, e.g., that a relay agent serving multiple LANs (RA2 in fig. 23.1) cannot be configured to forward the DHCP
requests from different LANs to different sets of DHCP servers.
23.1.2
DHCP Option 82
The relay agent information option (DHCP option 82, see RFC3046[33]) enables
a relay agent to pass information to the DHCP server regarding which port the
DHCP request came in on. Thus, an option 82 aware DHCP server would be able
to assign IP settings (IP address, etc.) to a PC based on the port the PC connects
to.
The DHCP option 82 contains two sub-options, Circuit ID and Remote ID:
ˆ Circuit ID: The circuit ID identifies the port on the relay agent, where the
DHCP request was received. Since the circuit ID can only be considered
unique within the reporting relay agent, the DHCP server generally needs
to consider both the circuit ID and an identifier of the specific relay agent
(e.g., giaddr or option-82 remote ID, see below) when processing the DHCP
request.
In WeOS the circuit ID can be set according to the following methods:
– Disabled: When circuit ID is disabled, no circuit ID sub-option is passed
as part of the Relay Agent Information option (DHCP option 82).
© 2017 Westermo Teleindustri AB
571
Westermo OS Management Guide
Version 4.21.1-0
– Port Name: Selecting the port name method implies that the circuit ID
will be represented as Type appended by the port identifier, e.g., Eth1
and DSL1 on a single slot product, or Eth1/1 and DSL1/1 on a multi-slot
product (see section 10.1.1 for more information on WeOS port naming
conventions).
– Port Description: By selecting the port description method, the circuit
ID will be represented by the port description setting of the associated
port. However, as of WeOS v4.21.1 the port description (chapter 10)
cannot yet be configured. Until configuration of port description is supported, the circuit ID will fall-back to using the port name, see above.
– Manual: You can configure the Circuit-ID manually per port. The Circuit
ID will be sent as a byte sequence (max 9 bytes), and you can choose
to enter your manual circuit ID setting either as an ASCII string (max 9
characters) or as hexadecimal number (max 18 hex characters).
ˆ Remote ID: According to RFC3046[33], the purpose of the remote ID is to
enable the DHCP relay agent to supply a trusted unique identifier of the
DHCP client. In practice, it is commonly used as an identifier of the relay
agent itself – the option 82 aware DHCP server can then base the IP address
assignment on the combination of circuit ID and remote ID. In WeOS the
remote ID can be set according to the following methods:
– Disabled: When remote ID is disabled, no remote ID sub-option is passed
as part of the Relay Agent Information option (DHCP option 82).
– MAC: By selecting the MAC method, the unit’s base MAC address (6
bytes, hexadecimal) will be used as remote ID. See sections 4.4.2 (Web)
and 7.3.2 (CLI) for information on how to read the unit’s base MAC address.
– IP: By selecting the IP method, the relay agent will use the IP address
of the interface where the DHCP request came in as remote ID (i.e., the
giaddr). E.g., if RA2 in fig. 23.1 receives a DHCP request from PC4, it
would use 192.168.2.1 as remote ID.
– System Name: By selecting the System Name method, the unit’s configured hostname/system name will be used as remote ID. See sections 8.2
(Web) and 8.3.2 (CLI) for information on how to configure the unit’s hostname/system name.
– Manual: It is also possible to set a manual value for the Remote ID,
either as a hexadecimal or string value.
572
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
When configuring a DHCP relay agent in WeOS, use of the relay agent information
option is by default disabled. When enabling DHCP option 82, the relay agent will
add its relay information option to incoming DHCP requests, unless the request
already contains a relay agent information option (added by some ”downstream”
relay agent)1 .
Below the possible policy settings are listed how the relay agent should handle
incoming DHCP requests already containing a relay agent information option. The
policy can both be specified globally (i.e., per relay agent), as well as on per port
basis.
ˆ Discard: Drop requests already containing a relay agent information option.
ˆ Forward: If the request already contains a relay agent information option,
keep that entry when forwarding the request towards your DHCP server(s).
ˆ Replace: If the request already contains a relay agent information option, replace that with your own DHCP option 82 field when forwarding the request
towards your DHCP server(s).
ˆ Append: If the request already contains a relay agent information option,
append your own relay agent information option field when forwarding the
request towards your DHCP server(s).
ˆ Require: Discard requests lacking a relay agent information option. If the
request already contains a relay agent information option, keep that entry
when forwarding the request towards your DHCP server(s). This option may
be useful in topologies including a mix of relay agents supporting and not
supporting DHCP snooping (see sections 23.1.4, and 22.1.5.2).
When handling DHCP requests already containing a relay agent information option, the following mechanisms apply to all policies:
ˆ Dropping requests lacking a giaddr: As of WeOS v4.21.1, incoming requests
containing a relay agent information option, but lacking a giaddr, will be
discarded.
ˆ Keeping existing giaddr: When forward a request which already contains a
relay agent information option, the giaddr field will be unchanged.
As of WeOS v4.21.1 no validation is performed by the relay agent on relay agent
information option field(s) included in DHCP messages returned from the DHCP
1 The exception is when policy ”Require” is configured - then the packet will be discarded if it
does not contain a relay agent information option.
© 2017 Westermo Teleindustri AB
573
Westermo OS Management Guide
Version 4.21.1-0
Server. The relay agent information is always removed2 before passing it back to
the DHCP client (PC), or to a relay agent closer to the PC. This behaviour may give
problems at downstream relay agents when using the Forward, Append, Replace,
and Require policies. WeOS handling of packets on the return path from the DHCP
server may be modified in upcoming WeOS releases.
23.1.3
DHCP Proxy Mode
According to the RFC2131[7] a DHCP relay agent would only be involved in the
initial DHCP message, while subsequent DHCP lease renew messages would be
sent directly between client and server, as shown in fig. 23.2.
For many use-cases however, this behaviour is not desirable. In particular with
Option 82 (see section 23.1.2) all DHCP messages from the client to the server
need to have this extra piece of information appended so that the server can
properly identify the client. This is called DHCP Proxy Mode, or DHCP Server
Identifier Override, defined in RFC5107[22].
Most modern DHCP servers support RFC5107[22], which is a sub-option to Option
82. But some older DHCP servers do not and for this particular case the WeOS
relay agent can be configured to forcibly override Option 54, the Server Identity
field. In effect, making sure that the client will send all DHCP messages via the
relay agent, see fig. 23.3.
Hence, there are two levels of DHCP Proxy Mode support in WeOS.
ˆ Hint to server: The WeOS relay agent adds sub-option 11 to option 82 in
all DHCP messages forwarded to the server, to hint the server to fill in the
IP address of the relay agent in the DHCP server identity field in the server
responses. If the server supports RFC5107[22], the relay agent will act as a
server proxy towards the client (fig. 23.3).
Note
The WeOS DHCP server (chapter 22) supports RFC5107[22].
ˆ Force identity override: The WeOS relay agent can force server identity
override by updating the packets sent towards the DHCP client. This feature can be useful in situations where the DHCP server does not support
RFC5107[22]. Forcing DHCP server identity override is disabled by default.
2 If
574
more than one relay information option is included, the last option is removed.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
DHCP
DHCP
Relay
Client
DHCP Discover
DHCP Discover
DHCP Offer
DHCP Offer
time
DHCP
Server
DHCP Request
DHCP Request
DHCP Ack
DHCP Ack
DHCP Renew Request
DHCP Ack
Figure 23.2: Typically only the initial DHCP exchange is done via the relay agent,
while lease renew messages are sent directly (unicast) between client and server.
23.1.4
Relay Agents in Switched Networks
The DHCP protocol uses layer-2 broadcast (Destination MAC: ff:ff:ff:ff:ff) for some
of its protocol messages. Therefore, a (broadcast) DHCP packet coming in to a
switch, will typically be flooded on all ports of the same LAN. This is illustrated in
fig. 23.4a):
ˆ A broadcast DHCP message comes in on port ”A” of the switch (step ”1a”).
ˆ The message is broadcasted unmodified on all other ports within the LAN
(here ports ”B”-”F”), see step ”1b”.
ˆ In this case, the switch is also running a DHCP relay service on the LAN.
The relay agent will process the incoming DHCP packet, and forwards it to
the configured DHCP server, which here happens to reside in the direction
of port ”E” (step ”2”). The packet in step ”2” is modified as compared to
the initial broadcast packet: It is sent as unicast to the DHCP server, and it
contains the relay agents IP address as giaddr. If the relay agent has DHCP
option 82 enabled, such information is also added.
© 2017 Westermo Teleindustri AB
575
Westermo OS Management Guide
Version 4.21.1-0
DHCP
DHCP
Relay
Client
DHCP Discover
DHCP Offer
time
DHCP Request
DHCP
Server
DHCP Discover
DHCP Offer
DHCP Request
DHCP Ack
DHCP Ack
DHCP Renew Request
DHCP Renew Request
DHCP Ack
DHCP Ack
Figure 23.3: DHCP Proxy Mode, all messages goes via the relay agent.
As seen in fig. 23.4a), using (layer-2) switches as DHCP Relay Agents can result
in multiple versions of a DHCP message to be sent towards the DHCP server:
the original request being switched/broadcasted, and the one being relayed by
the relay agent process. This will not cause any problems if the DHCP server
is located on some remote network; then only the relayed packet will reach the
server. However, if the DHCP server is located within the same LAN, adequate
support is needed at the DHCP server to know which request to serve and which
to ignore (see section 22.1.5.2 in the DHCP server chapter for more information).
The number of ”copies/versions” of a DHCP request can increase further if a
LAN consists of several switches with DHCP relay agents (discussed later on, see
fig. 23.5).
To mitigate multiplication of broadcast DHCP messages, some switches support
DHCP snooping (see also section 23.1.5 for an alternative approach). With DHCP
snooping enabled on an Ethernet/DSL port, all DHCP packets will pass through
the DHCP relay agent – this includes broadcast and unicast DHCP packets, both
DHCP requests (to server) or DHCP responses (from server) coming in on that
port. Fig. 23.4b) shows the result when a broadcast DHCP packet comes in on a
port with DHCP snooping enabled.
576
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Relay
Agent
Towards
DHCP
Server
Relay
Agent
2
E
A
1b
B
C
F
D
1b
1a
1b
1b
1b
1b
Broadcast
DHCP
packet
a) No DHCP snooping support
Towards
DHCP
Server
2
E
A
1b
B
C
F
D
1a
Broadcast
DHCP
packet
b) DHCP snooping supported
Figure 23.4: Propagation of DHCP broadcast packets in switches running DHCP
relay agents. All ports are on the same (V)LAN. The switch in figure a) does not
support DHCP snooping, while the switch in figure b) supports DHCP snooping.
When configuring a WeOS relay agent on a VLAN interface, all ports on that VLAN
will have DHCP snooping enabled
- the exception is products lacking hardware support for DHCP snooping3 . More
fine-grained control to enable/disable DHCP snooping per port may be supported
in later WeOS versions.
DHCP relay service can be disabled on a per port basis. If DHCP relaying is disabled on an Ethernet/DSL port, incoming DHCP packets will be switched as other
layer-2 packets (no DHCP snooping), and the DHCP relay agent on the switch will
ignore DHCP requests entering the switch on that port.
Fig. 23.5 presents an example where multiple relays are located within the same
VLAN – port 1-6 on all RA units are in the same VLAN, while port 7 on RA1 and
RA2 are associated with another VLAN used and used as upstreams interface. The
topology in fig. 23.5 utilise several WeOS features to achieve a robust network:
FRNT (chapter 16) is used to handle single link failures within the local network.
VRRP (chapter 30) is used to handle router redundancy (RA1 and RA2). A second
DHCP server to protect against DHCP server failure4 .
3 In WeOS products, DHCP Snooping is supported on all Ethernet/DSL ports, except for ports of
switchcore(s): MV88E6095, MV88E6185 and MV88E6046. Please see Detailed System Overview
page in the Web (section 4.4.2) or use the ”show system-information” in the CLI (section 7.3.2)
to find information about what switchcore(s) is used in your product.
4 As of WeOS v4.21.1, the WeOS DHCP server (chapter 22) does not provide dedicated DHCP
server failover support, but you can achieve redundancy using ”static” address assignment (no
address pools) with the same configuration at both DHCP servers.
© 2017 Westermo Teleindustri AB
577
Westermo OS Management Guide
Version 4.21.1-0
DHCP
DHCP
Server
Server
IP:10.1.2.3
Internet/Intranet
Routers (possibly
running VRRP)
7
7
RA1
6
RA2
5
6
1 2 3 4
5
1 2 3 4
FRNT
RA3
5
6
1 2 3 4
Focal Point
RA4
5
6
1 2 3 4
RA5
5
6
1 2 3 4
Figure 23.5: Example with multiple DHCP Relay Agents within the same VLAN
(port 1-6 on all RAs are assumed to be on the same VLAN, e.g., VLAN 1).
The relay agents (RA1-RA5) server DHCP clients connecting to the local access
ports (ports 1-4), and will relay each request (unicast) to the configured DHCP
server(s). Below a sample DHCP relay configuration is shown, which would be
suitable for all relay agents in fig. 23.5.
Example
dhcp-relay
iface vlan1
server 10.1.2.3
option82 discard
port 5-6
no enable
end
end
ˆ DHCP relay has been enabled on interface vlan1 (this assumes that ports
1-6 are all associated with VLAN 1).
ˆ A single DHCP server has been configured (here 10.1.2.3). As of WeOS
v4.21.1, up to two DHCP servers can be configured.
ˆ Option 82 is enabled, with policy discard. Option 82 information will be
added to all incoming requests. Packets which already include option 82
578
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
information will be discarded. Default settings for circuit-id (port name) and
remote-id (base-MAC) will be used.
ˆ DHCP requests coming in on port 5 or 6 will be ignored by the relay agent.
No DHCP snooping will be done on those ports, thus a DHCP request being
relayed by RA4 to the DHCP server, will be forwarded through RA5 like any
other packet.
23.1.5
Mitigating duplication of DHCP messages by using a different server port
An alternative to address the issue with multiple DHCP requests in switched
topologies with non-snooping relay agents is to let the DHCP server listen on a
non-standard UDP port (section 22.1.3). The DHCP relay agent can be configured
to forward its packets to this server port (section 23.1.1), thus all relayed packets
will reach the server. Packets coming directly from the client will be dropped by
server, since they are sent to the regular DHCP server UDP port (67).
Example
-- DHCP Server configuration of non-standard listen port
example-server:/#> configure
example-server:/config/#>
example-server:/config/#> dhcp-server
example-server:/config/dhcp-server/#> server-port 6767
example-server:/config/dhcp-server/#> leave
Stopping DHCP/DNS Server ................................... [ OK ]
Starting DHCP/DNS Server ................................... [ OK ]
Configuration activated. Remember "copy run start" to save to flash (NVRAM).
example-server:/#>
-- DHCP Relay Agent configuration of non-standard server port
example-relay:/#> configure
example-relay:/config/#> dhcp-relay
example-relay:/config/dhcp-relay/#> server-port 6767
example-relay:/config/dhcp-relay/#> leave
Stopping DHCP/DNS Server ................................... [
Starting DHCP/DNS Server ................................... [
Configuration activated. Remember "copy run start" to save to
Starting DHCP Relay Agent .................................. [
example-relay:/#>
© 2017 Westermo Teleindustri AB
OK ]
OK ]
flash (NVRAM).
OK ]
579
Westermo OS Management Guide
Version 4.21.1-0
23.2
Configuring DHCP Relay Agent via the Web
The Web interface provides management of the DHCP Relay Agent.
23.2.1
DHCP Relay Agent settings
Menu path: Configuration ⇒ Network (IP) ⇒ DHCP-Relay
Figure 23.6: DHCP Relay Agent settings
580
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Listening
Interfaces
DHCP
Servers
Global
Option 82
Settings
The Listening Interface specifies on which interface(s) the
relay agent will listen for client requests. DHCP server responses may come in through any interface.
The DHCP Servers settings determine to which DHCP servers
each DHCP client request will be sent. At most two servers
may be configured.
The Global Option 82 Settings determine how the DHCP
Relay Agent Information option, also known as Option 82, will
be handled. The policy specify how to treat incoming client
requests that already contain an Agent Information option.
ˆ Disable: Do not add option 82 field. Any existing option
82 will be retained.
ˆ Forward: Adds a new option 82 or forwards any existing
option 82.
ˆ Append: Appends a new option 82 in addition to any
existing option 82.
ˆ Discard: Drops the whole packet if it contains an option
82.
ˆ Replace: Removes any existing option 82 and adds a
new option 82.
ˆ Require: Requires that the incoming packet contains an
option 82 otherwise it will be dropped.
The Circuit ID setting determines how the Circuit-Id field of option 82 will be
filled. It can be one of None, Port Name and Port Description. None will
leave this field with zero length, Port Name will fill this field with the port type
and name of the port as seen on front foil, stripped of any whitespace. E.g. Eth6
for Ethernet port 6. Lastly Port Description will use the description given to the
port in the port settings.
In a similar fashion the Remote ID tells how the Remote-id field of option 82 will
be set. None set its length to zero, IP sets it to the IP address of the inbound
interface. MAC uses the base MAC address of the unit. System Name uses
the hostname of the system. It is also possible to set a Manual value for the
Remote-ID, either as a hexadecimal or string value.
© 2017 Westermo Teleindustri AB
581
Westermo OS Management Guide
Version 4.21.1-0
23.2.2
DHCP Relay Agent Per-Port Settings
Menu path: Configuration ⇒ Network (IP) ⇒ DHCP-Relay Agent ⇒ Port Specific
Settings Show
Figure 23.7: DHCP-Relay Agent Per-Port Settings page
Enabled
Option 82
Policy
Option 82
Circuit ID
582
The Enabled checkbox tells whether to enable the relay agent
on this port, i.e. whether to listen for client requests on this
port or not. If enabled, you can override the global settings.
See section 23.2.1 for an explanation of the different policy
options. In the port specific section, the Policy setting has
an additional option Global, indicates that the global policy
setting (see fig. 23.6) will be used for this port.
See section 23.2.1 for an explanation of the different circuit ID
types. In the port specific section, the Circuit ID setting has
additional options for the Circuit ID type.
ˆ Global: Indicates that the global circuit ID setting (see
fig. 23.6) will be used for this port.
ˆ Manual (hex) and Manual (string): A user specified
hex or string value will be used as circuit ID. Value is entered in the Manual Circuit ID field.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
23.3
Configuring DHCP Relay Agent via the CLI
Command
Configure DHCP Relay Agent
[no] dhcp-relay
[no] enable
[no] iface <IFACE>
[no] server <IPADDR>
[no] server-port <PORT>
[no] force-server-identity
[no] option82 <forward|discard|append|
replace|require>
[no] circuitid-type <portname|
portdescription>
[no] remoteid-type <mac|ip|
system-name|manual <hex|string> <VALUE>>
port <PORTLIST|all>
[no] enable
[no] option82 <auto|forward|discard|
append|replace|require>
[no] circuitid-type <auto|portname|
portdescription|
manual <hex|string> <ID>>
View DHCP Relay Agent Settings
show dhcp-relay
dhcp-relay
show port [PORTLIST]
23.3.1
Default
Section
Enabled
Disabled
Disabled
Disabled
Disabled
Disabled
Section
Section
Section
Section
Section
Section
Section
”portname”
Section 23.3.8
”mac”
Section 23.3.9
Enabled
”auto”
Section 23.3.10
Section 23.3.11
Section 23.3.12
”auto”
Section 23.3.13
23.3.1
23.3.2
23.3.3
23.3.4
23.3.5
23.3.6
23.3.7
Section 23.3.14
”all”
Section 23.3.15
Manage DHCP Relay Agent
Syntax [no] dhcp-relay
Context Global Configuration context
© 2017 Westermo Teleindustri AB
583
Westermo OS Management Guide
Version 4.21.1-0
Usage Create, modify or remove the DHCP Relay Agent.
Enter DHCP relay agent context.
Use ”no dhcp-relay” to remove an existing DHCP relay configuration.
Default values Not applicable.
23.3.2
Enable DHCP Relay Agent
Syntax [no] enable
Context DHCP Relay Configuration context
Usage Enable the DHCP Relay Agent.
Default values Enabled.
23.3.3
Listening Interfaces
Syntax [no] iface <IFACE>
Context DHCP Relay Configuration context
Usage Specify the interfaces that the relay agent will listen to.
Note
If the interface includes a aggregate the physical ports that is a part of
the aggregate will be represented with the aggregate port.
Default values Not applicable.
23.3.4
DHCP Servers (IP addresses)
Syntax [no] server <ADDRESS>
Context DHCP Relay Configuration context
Usage Specify the DHCP server that the relay agent will forward requests to.
Default values Not applicable.
584
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
23.3.5
DHCP Server UDP port
Syntax [no] server-port <UDPPORT>
Context DHCP Relay Configuration context
Usage Specify the DHCP server UDP port that the relay agent will forward requests to. See also section 22.3.4 for the corresponding DHCP relay agent
setting.
Use ”no server-port” to reset to default value (port 67).
Use ”show server-port” to show current server-port settings.
Default values 67
23.3.6
Force DHCP Server Identity Override
Syntax [no] force-server-identity
Context DHCP Relay Configuration context
Usage By enabling the force DHCP server override setting, the DHCP relay agent
can work-around older DHCP servers that do not support RFC5107[22] (a
hint/extension to Option 82) by overriding Option 54 in the server response
to the client with the relay agents IP address.
It is recommended to leave this setting disabled and instead either use the
WeOS DHCP server, or upgrade to another RFC compliant DHCP server.
Use ”force-server-identity” to enable force DHCP server override and
use ”no force-server-identity” to disable it.
Use ”show force-server-identity” to show the current setting.
Default values Disabled
23.3.7
Option 82
Syntax [no] option82 <forward|discard|append|replace|require>
Context DHCP Relay Configuration context
© 2017 Westermo Teleindustri AB
585
Westermo OS Management Guide
Version 4.21.1-0
Usage Enable or disable the addition of option 82, a.k.a. relay agent information,
to DHCP requests. The policy for how to handle any existing option 82 can
optionally be specified as follows.
Forward
Adds a new option 82 or forwards any existing option 82.
Append
Appends a new option 82 in addition to any existing option 82.
Discard
Drops the whole packet if it contains an option 82.
Replace
Removes any existing option 82 and adds a new option 82.
Require
Requires that the incoming packet contains an option 82 otherwise it
will be dropped.
Default values Option 82 is disabled by default, if enabled and policy is omitted
it defaults to forward.
23.3.8
Circuit ID Type
Syntax [no] circuitid-type <portname | portdescription>
Context DHCP Relay Configuration context
Usage Specify how the circuit id in option 82 will be set. portname will use
the name of the port as it is printed on the front foil plus the port type.
For Ethernet ports it will be Eth, so e.g. requests coming in on port 6 will
have the Circuit ID set to “Eth6”. portdescription is currently the same as
portname but will use the port description set in the port configuration, as
soon as that feature is released.
Default values portname.
23.3.9
Remote ID Type
Syntax [no] remoteid-type <mac | ip | system-name | manual <hex|string>
<VALUE>>
586
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Context DHCP Relay Configuration context
Usage Specify how the remote id in option 82 will be set.
ˆ Use command ”remoteid-type mac” to use the unit’s base MAC address as remote-id.
ˆ Use command ”remoteid-type ip” to use the unit’s IP address (IP address of the interface where the DHCP client request comes in, see section 23.3.3).
ˆ Use command ”remoteid-type system-name” to use the unit’s hostname as remote-id.
ˆ Use command ”remoteid-type manual string <VALUE>” and ”remoteid-type
manual hex <VALUE>” set remoteid-type manually as a string or hex
value.
ˆ Use command ”no remoteid-type” to skip sending a remote-id.
ˆ Use command ”show remoteid-type” to show the current remoteid
type setting.
Default values mac
23.3.10
Manage DHCP Relay Agent Per-Port Settings
Syntax port <PORT|PORTS>
Context DHCP Relay Configuration context
Usage Modify DHCP Relay Agent configuration for one or several ports.
Default values Not applicable.
23.3.11
Enable/disable DHCP Relay Agent per port
Syntax [no] enable
Context DHCP Relay Configuration context
Usage Enable or disable the DHCP Relay Agent on a port.
Default values Enabled.
© 2017 Westermo Teleindustri AB
587
Westermo OS Management Guide
Version 4.21.1-0
23.3.12
Option 82 policy per port
Syntax [no] option82 <auto|forward|discard|append|replace|require>
Context DHCP Relay Port Configuration context
Usage Enable or disable the addition of option 82 on one ore more ports. The
auto policy uses the same a policy as specified in the DHCP Relay context.
Default values auto.
23.3.13
Option 82 Circuit ID per port
Syntax [no] circuitid-type <auto|portname|portdescription>
Context DHCP Relay Port Configuration context
Usage Specify how the circuit id in option 82 will be set for this port. In addition
to the keywords defined in section 23.3.8 auto can be used, meaning the
configured circuit ID type in DHCP relay context.
Default values auto.
23.3.14
Show DHCP Relay Agent Settings
Syntax show dhcp-relay Also available as ”show” command within the DHCP
Relay Configuration context context.
Context Global Configuration context
Usage Show DHCP relay agent settings.
Default values
23.3.15
Show DHCP Relay Agent Per-port Settings
Syntax show port [PORTLIST] Also available as ”show” command within the
DHCP Relay Port Configuration context.
Context DHCP Relay Configuration context
588
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Usage Show DHCP relay agent per port settings. Furthermore, not only the circuit ID type settings are listed, but also the resulting circuit ID.
Default values If no PORTLIST is given, settings are listed for all ports associated with the given (VLAN) interfaces (see also section 23.3.3).
Example
example:/config/dhcp-relay/#> show port
Port
Enabled Policy Circuit-ID type
(Circuit ID)
==============================================================================
Eth 1
NO
auto
auto
(Eth1)
Eth 2
NO
auto
auto
(Eth2)
Eth 3
YES
auto
auto
(Eth3)
Eth 4
YES
auto
auto
(Eth4)
Eth 5
YES
auto
auto
(Eth5)
Eth 6
YES
auto
auto
(Eth6)
example:/config/dhcp-relay/#>
© 2017 Westermo Teleindustri AB
589
Westermo OS Management Guide
Version 4.21.1-0
Chapter 24
Alarm handling, Front panel
LEDs and Digital I/O
This chapter describes WeOS features for alarm and event handling (sections 24.124.3). The chapter also covers general information on functionality related to
Digital I/O and front panel LEDs (sections 24.4 and 24.5).
24.1
Alarm handling features
The table below summarises the WeOS alarm handling features.
Feature
Configure alarm triggers
Configure alarm actions
Configure alarm targets
View alarm status1
24.1.1
Web
X
X
X
X
CLI
X
X
X
X
General Description
Sections 24.1.1-24.1.3
Sections 24.1.1 and 24.1.4
Sections 24.1.1 and 24.1.5
Section 24.1.5
Introduction to the WeOS alarm handling support
The WeOS alarm handling support makes use of the following terminology:
1 In addition to monitoring alarm status via Web and CLI, there are other ways in which an
operator can get notified when an alarm is triggered.
590
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Alarm triggers
Alarm targets specify:
Alarm type and source(s), e.g., "temperature sensor 1"
Alarm thresholds and condition (e.g., ’alarm’ for
temperatures above 50 degrees.)
Alarm severity, e.g., DEBUG, INFO, etc.
Alarm action "index", i.e., point out the action to
invoke when the trigger becomes "active" and "inactive".
Alarm sources
Examples:
Temperature sensors
Digital In sensors
Link−alarm (ports/interfaces)
RMON counters (ports/interfaces)
FRNT Ring Status
Alarm actions
Alarm actions specify:
Alarm targets
Alarm targets
Examples:
ON LED
Status Relay
SNMP Trap
Logging
Figure 24.1: Overview of WeOS alarm entities: Alarm triggers monitor the state of
alarm source, and define conditions and thresholds when to invoke an associated
alarm action. The invoked alarm action specifies what alarm target(s) to use to
notify the operator.
ˆ Alarm sources: An alarm source is an object being monitored by an alarm
trigger, e.g., the link status (up/down) of an Ethernet port, the input byte
counter of a network interface, or the temperature value of a temperature
sensor. Alarm sources are described further in section 24.1.2.
ˆ Alarm trigger: An alarm trigger monitors alarm sources, and defines the
conditions when alarm events occur, i.e., when the trigger becomes active
(alarm situation) or inactive (normal situation).
In addition, the alarm trigger specifies the alarm action to be invoked once
an alarm event occurs. Alarm triggers are described further in section 24.1.3.
ˆ Alarm actions and alarm targets: When an alarm event occurs, the operator
can be notified via SNMP traps, logging, status relay (digital-out), and front
panel status LED. These notification mechanisms are referred to as alarm
targets.
Instead of mapping triggers directly to targets, a trigger is mapped to an
alarm action (profile). The alarm action defines what specific targets to use
when an alarm event occurs. For example, a link alarm trigger for ports 1-3
© 2017 Westermo Teleindustri AB
591
Westermo OS Management Guide
Version 4.21.1-0
can be mapped to a specific alarm action, which in turn specifies logging
and SNMP traps as targets. Alarm actions and targets are described further
in sections 24.1.4 and 24.1.5 respectively.
24.1.2
Alarm sources
As of WeOS v4.21.1 the following alarm sources are supported:
ˆ Power failure: If the unit is equipped with redundant power feed (or redundant power supply), an alarm can be triggered if one of the feeds lack input
power.
Note: if all power is lacking on all feeds, the unit is powerless and cannot
trigger alarms via SNMP traps or remote logging. To detect such a situation
remotely, the operator could poll the unit (e.g., by pinging the unit on a
regular interval). The drawback is that it is difficult to distinguish problems
in the intermediate network from problems in the monitored device.
An alternative is to use out-of-band signalling, e.g., via GPRS equipment connected to the status relay (digital-out) to get an alarm notification instantly
if a device goes down.
ˆ Link alarm: It is possible to configure link alarm triggers to react when a
link goes down (and up).
ˆ Digital-In: Alarms can be triggered depending on the presence of input voltage/current on the Digital-In pins of the Digital I/O connector.
ˆ Temperature sensor alarms: Temperature alarm triggers can be configured
to react when the temperature rises above (or falls below) some defined
threshold.
ˆ FRNT status: The FRNT ring status trigger will react when an FRNT ring is
broken or healed (intact)1 .
ˆ RiCo/Dual-Homing uplink status: The RiCo uplink status trigger will react if
a RiCo (or Dual-Homing) uplink comes up or goes down. Both physical link
status and Uplink Echo response are used to determine the uplink status.
ˆ Hardware failure: Hardware alarms triggers notifies that the unit has detected a hardware failure (typically if an unsupported SFP is inserted).
1 Only
592
an FRNT focal point can determine the ring status with certainty.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
ˆ SHDSL/xDSL SNR Margin: On devices with SHDSL/xDSL ports, alarms can be
triggered when the SNR margin falls below some configured threshold.
ˆ Link Fault Forward (LFF): On devices with SHDSL ports, alarms can be triggered when the remote SHDSL switch indicates it has link down on its Ethernet port. That is, this feature can be used in topologies where an Ethernet
is extended over an SHDSL link, and where the remote SHDSL switch (e.g.,
a DDW-120) is able to signal that the Ethernet link is down on its side.
ˆ Network Connectivity (Ping): It is possible to have a trigger to monitor
network connectivity by using the ping command to a specific host. The
remote node is considered unreachable if a configurable number of pings are
lost, and considered reachable if the same number of pings are successfully
received.
Note
Make sure the remote host responds to ICMP ping. A typical behaviour
of many hosts is that ICMP ping is blocked in the host’s firewall.
ˆ PoE Power Usage: On units supporting Power Over Ethernet (PoE), alarms
can be triggered when the total power usage raises above (or falls below)
some configured threshold.
ˆ Microlok Session Status On units running a Microlok Gateway (chapter 41,
an alarm can be triggered if any of the established sessions go down.
ˆ Address Conflict: An alarm can be triggered to monitor a designated VLAN
network interface for address conflicts.
24.1.3
Alarm triggers
An alarm trigger defines the rules for when alarm events should be generated
for a monitored alarm source. Alarm triggers also define which alarm action to
invoke when an alarm event occurs.
Currently supported alarm trigger types:
ˆ Power failure
ˆ Link alarm
ˆ Digital-In
ˆ Temperature
© 2017 Westermo Teleindustri AB
593
Westermo OS Management Guide
Version 4.21.1-0
ˆ FRNT ring status
ˆ RiCo uplink status (applies both to RiCo and Dual-Homing uplinks)
ˆ Hardware failure (The hardware failure alarm trigger is implicit, and cannot
be removed or modified.)
ˆ SNR margin (SHDSL and xDSL ports)
ˆ LFF (SHDSL ports)
ˆ Timer
ˆ Ping
ˆ PoE power usage
ˆ Microlok Session Status
ˆ Address Conflict
As the WeOS alarm handling support is designed to include triggers for additional
alarm sources, the following description is of more general nature, thus contains
more options than needed for the trigger types currently supported.
Note
As of WeOS v4.21.1 there is no support for making an alarm trigger persistent. When an alarm condition is no longer fulfilled, the trigger status
will become inactive. As alarms are not persistent, it is not possible for an
operator to clear (i.e., acknowledge) an alarm.
24.1.3.1
Specifying what alarm source(s) a trigger should monitor
Different types of alarm triggers operate on different types of alarm sources:
ˆ Power failure: A power failure trigger can monitor one or more power feed
sensors. Most WeOS products have two power feeds (single power supply),
with a sensor for each power feed. Typically a single power failure trigger is
used to monitor both power feed sensors.
ˆ Digital-In: A digital-in trigger can monitor one or more digital-in sensors.
WeOS products typically have a single digital-in sensor.
594
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
ˆ Link alarm: Link alarm triggers monitor the operational status (up/down) of
Ethernet or DSL ports. Thus when configuring a link alarm trigger the port
(or ports) to monitor should be specified.
Note
It is possible to define multiple link alarm triggers, where each trigger
can monitor different ports and be mapped to different alarm actions.
In the future, link alarm triggers can be extended to monitor the operational
status of network interfaces and VLANs in addition to physical ports (Ethernet, SHDSL, etc.).
ˆ RMON statistics (not yet supported): The alarm source for an RMON trigger is specified by two parameters: (1) the name of the statistics counter
(e.g., etherStatsPkts), and (2) the port (or list of ports) for which this counter
should be monitored.
Note
In WeOS the term RMON is used to refer to data traffic statistics in
general; not only to the Ethernet statistics defined in the RMON MIB.
Thus, if a counter from the IF-MIB (such as ifHCInUcastPkts is specified,
the alarm source could refer to network interfaces or VLANs as well as
a physical ports (Ethernet, SHDSL, etc.).
ˆ Temperature: Temperature triggers can apply to one or more temperature
sensors.
ˆ FRNT: FRNT triggers can apply to one or more FRNT rings (as of WeOS
v4.21.1 only a single FRNT ring is supported).
ˆ RiCo/Dual-Homing: RiCo uplink triggers can apply to one or more RiCo (or
Dual-Homing) uplink ports.
ˆ Timer: Timer triggers are configured to go off at given time interval. As
of WeOS v4.21.1, only daily timers are supported, e.g., ”timeout daily
02:30”, and only apply to ”log” and ”reboot” action targets.
ˆ SNR Margin: An SNR Margin trigger applies to one or more SHDSL/xDSL
ports.
ˆ LFF (Link Fault Forward): An LFF trigger applies to one or more SHDSL ports.
ˆ Ping: A connectivity checker, sends an ICMP ping in a configurable interval.
© 2017 Westermo Teleindustri AB
595
Westermo OS Management Guide
Version 4.21.1-0
ˆ PoE Power Usage: The WeOS PoE enabled units have a single PoE power
module, and its current usage level is used as trigger source (i.e., no need
to select a trigger source).
ˆ Address Conflict Interface: An Address Conflict trigger applies to a VLAN
network interface. The trigger monitors if any IPv4 address or MAC address
used by the WeOS unit is also used by another device on the VLAN. Only
one VLAN network interface can be configured per Address Conflict trigger.
Typically there would be no more than one trigger monitoring the status of a specific alarm source. However, in some cases it would make sense to have multiple
triggers monitoring a single alarm source. For example, one could define two
temperature triggers for a single temperature sensor, where one trigger reacts if
the temperature rises above a warning threshold (say 60◦ C), and the other if the
temperature gets critically high (say 75◦ C).
24.1.3.2
Alarm thresholds and trigger output
For the trigger to know when an alarm event has occurred, threshold values for
the monitored alarm sources must be configured. Alarm sources which are ’binary’ to their nature (link up/down, power up/down, digital-in high/low, etc.) have
thresholds defined implicitly.
For sources which can take values in a wider range (temperature, SNR Margin,
received packets within a given time interval, etc.) the alarm thresholds should
be configured. Fig. 24.2a) illustrates use of alarm thresholds for a temperature
trigger.
As can be seen in fig. 24.2a), two thresholds are used – a rising threshold and a
falling threshold. Alarm events will be generated when reaching the rising threshold on the way up, and the falling threshold on the way down. However, once a
rising alarm event has occurred, a new rising alarm event cannot be generated
(for that alarm source) before the value has fallen down to the falling threshold
(and vice versa). Thus, the use of separate rising and falling thresholds creates
a hysteresis mechanism, which avoids generating multiple alarm events when a
monitored value fluctuates around the alarm threshold.
Alarm targets such as the status relay (Digital-Out) and the ON LED provide a
summary alarm function (see section 24.1.5.1), and these targets assume that
every alarm trigger define the condition when the alarm is active (”alarm” situation) and inactive (”normal” situation). To define this the alarm condition configuration option is used. To warn the operator for high temperatures, the alarm
596
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Temperature
Alarm event
(rising/high)
Rising threshold
Falling threshold
Alarm event
(falling/low)
time
a) Temperature status and alarm events.
Alarm trigger "active"
Alarm trigger "inactive"
time
b) Trigger status with alarm condition "high".
Alarm trigger "active"
Alarm trigger "inactive"
time
c) Trigger status with alarm condition "low".
Figure 24.2: Example use of rising and falling thresholds for a temperature alarm
trigger (a), and alarm condition setting to affect active and inactive trigger status
(b and c).
condition should be set to ”high”, see fig. 24.2b). If we instead wish to warn the
operator for low temperatures, the alarm condition should be set to ”low”, see
fig. 24.2c). A corresponding example for a Digital-In trigger is shown in fig. 24.3.
Additional details on threshold settings and properties:
ˆ The rising threshold cannot be set lower than the falling threshold.
ˆ It is possible to use the same value for the rising and falling thresholds.
ˆ Rising alarm events occur if the current sample value is equal or above
the rising threshold, and the previously sampled value was below the rising
threshold. A rising alarm event will also occur if the first sampled value is
equal or above this threshold, and the condition variable is configured as
rising (or any of its equivalents: high or up).
© 2017 Westermo Teleindustri AB
597
Westermo OS Management Guide
Version 4.21.1-0
High (Voltage present)
Low (Voltage non−present)
time
a) Digital In Sensor state
Alarm trigger "active"
Alarm trigger "inactive"
time
b) Digital In Trigger State (Alarm condition set to "high")
Alarm trigger "active"
Alarm trigger "inactive"
time
c) Digital In Trigger State (Alarm condition set to "low")
Figure 24.3: Alarm condition example: The alarm trigger for digital-in can be
configured to become active when the signal is high (b) or when it is low (c).
ˆ Falling alarm events occur if the current sample value is equal or below the
falling threshold, and the previously sampled value was above the falling
threshold. A falling alarm event will also occur if the first sampled value is
equal or below this threshold, and the condition variable is configured as
falling (or any of its equivalents: low or down).
24.1.3.3
Sample types and interval
Two sample types are possible: absolute and delta sampling. With absolute sampling, the value is compared directly to the alarm thresholds. With delta sampling
it is the difference between the current sample and the previous sample which is
compared to the alarm thresholds.
Alarm sources of counter type, such as RMON data traffic statistics, are well
suited for delta sampling. As the delta is computed over a given time interval
(sample interval), the alarm thresholds should be configured with respect to the
configured sample interval.
Note
As of WeOS v4.21.1 only absolute sampling is supported, and the sampling
interval is not configurable for any trigger type.
598
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
24.1.3.4
Alarm severity
For each trigger it is possible to define the severity level of the associated alarm
events. The levels defined by Unix Syslog are used:
ˆ EMERG: System is unusable
ˆ ALERT: Action must be taken immediately
ˆ CRIT: Critical conditions
ˆ ERR: Error conditions
ˆ WARNING: Warning conditions
ˆ NOTICE: Normal, but significant, condition
ˆ INFO: Informational message
ˆ DEBUG: Debug-level message
It is also possible to configure severity level ”NONE”. Alarm events with severity
NONE will not cause SNMP traps to be sent or events to be logged, however, such
events can still affect the status relay (digital-out) and ON LED targets.
Note
Severity levels can be configured independently for the events when an
alarm trigger becomes ”active” and ”inactive”. Default severity level are
WARNING for ”active” alarm events and NOTICE for ”inactive” alarm events.
24.1.3.5
Mapping triggers to actions
Triggers can be mapped to alarm actions (profiles) that are invoked when an
alarm event occurs, for more information see section 24.1.4. However, it is also
possible to leave a trigger unmapped, e.g., when defining a ping trigger to adjust
VRRP priority dynamically (see section 30.1.1).
24.1.4
Alarm actions - mapping triggers to targets
Instead of mapping triggers directly to alarm targets, each trigger is mapped to
an alarm action (alarm action profile). The alarm action specifies which targets
© 2017 Westermo Teleindustri AB
599
Westermo OS Management Guide
Version 4.21.1-0
to use (SNMP traps, Logging, ON LED, and status relay) when an alarm event
occurs.
It is possible to configure several actions (action profiles). Each trigger can be
mapped to an individual action, but it is also possible for multiple triggers to
share the same action. This can be particularly useful when managing several
triggers of similar type, such as different types of RMON triggers.
By default a trigger is mapped to the default alarm action (index 1). The default
alarm action cannot be removed.
24.1.5
Alarm presentation (alarm targets)
When an alarm situation occurs, such as an FRNT ring failure, WeOS enables the
operator to be notified in numerous ways:
ˆ SNMP trap: Alarms can be configured to generate SNMP traps2 . See chapter 6 for general information on SNMP.
ˆ Log files and remote logging: Alarms can be logged locally or passed to a
remote logging server. See chapter 25 for general information on event and
alarm logging.
ˆ Status Relay: On units equipped with a Digital I/O contact, the status relay pins (Digital-Out) can be used as an alarm target. Similar to the ’ON’
LED, the status relay provides a summary alarm function, where the ’gate’
is closed when the switch is operating ’OK’, and open when any of the associated alarm triggers becomes active (or when the unit has no power).
See section 24.4 for general information on the Digital I/O.
ˆ ’ON’ LED: There are front panel LEDs which can indicate status of specific
ports or protocols. There is also a general status LED, which shows a green
light when the unit is operating ’OK’, but shows a red light as soon as any of
the associated alarm triggers becomes active. Thus, the ’ON’ LED provides
a summary alarm function.
See section 24.5 for general information on front panel LEDs.
ˆ Reboot: (USE WITH CARE) The reboot target is used to make the unit to
reboot upon a specified alarm event. The purpose is to provide a way to
2 As
600
of WeOS v4.21.1 there is no support for SNMP traps for timer or hardware alarms.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
reboot the unit on a regular basis (i.e., by mapping a timer trigger to an
action profile with target reboot, see section 24.3.2.9).
In addition, an operator can view the alarm status via the Web and CLI interfaces.
24.1.5.1
Summary alarm
The summary alarm in use by the status relay (digital-out) and ON LED targets
assumes that every alarm trigger define the condition when the alarm is active
(”alarm” situation) and inactive (”normal” situation).
ˆ For many triggers this definition is implicit, e.g., a link alarm is active when
the port (or interface) is down and inactive it is up.
ˆ Other triggers, such as temperature or digital-in sensor triggers allow for the
operator to define if the alarm is active: high or low temperature, voltage
signal present or not present, etc. See section 24.1.3.2, and in particular
figs. 24.2 and 24.3, for further information on the active and inactive trigger
states.
Working as a summary alarm, the status relay as well as the ON LED will indicate
’alarm’ as soon as any of the associated alarm triggers become active. For the
ON LED alarm is indicated with a ’red’ light, as shown in fig. 24.4. For the status
relay, alarm is indicated by having the gate in ’open’ state. See sections 24.4
and 24.5 for general information on Digital I/O and front panel LEDs.
Trigger 1
"active"
"inactive"
time
Trigger 2
"active"
"inactive"
time
Trigger 3
"active"
"inactive"
time
Target
ON LED
"red"
"green"
time
Figure 24.4: Summary alarm example with three alarm triggers mapped to the ON
LED alarm target. The ON LED indicates ’alarm’ (red) when any of the associated
triggers are active.
© 2017 Westermo Teleindustri AB
601
Westermo OS Management Guide
Version 4.21.1-0
24.1.5.2
Target Severity thresholds
As of WeOS v4.21.1 setting target severity thresholds is not yet supported.
For logging and SNMP trap targets it is possible to filter alarm events depending
on severity. E.g., if the SNMP trap target configures its severity threshold to
WARNING, only events of severity level WARNING or higher will cause SNMP traps
to be sent.
By default, both logging and SNMP trap targets have severity threshold set to
level INFO. See section 24.1.3.4 for information on how to classify the severity
for alarm triggers.
602
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
24.2
24.2.1
Managing Alarms via the Web
Show alarm status
Alarm status is presented in the System Overview and the Detailed System Overview
web pages, which are described in sections 4.4 and 4.4.2.
Fig. 24.5 shows the System Overview page when a Link Alarm is activated.
Figure 24.5: The basic system overview page with a link alarm activated.
© 2017 Westermo Teleindustri AB
603
Westermo OS Management Guide
Version 4.21.1-0
24.2.2
Trigger configuration overview page
Menu path: Configuration ⇒ Alarm ⇒ Triggers
When entering the Alarm configuration page you will be presented to a list of all
alarm triggers configured on your unit, see below.
Figure 24.6: The alarm trigger configuration overview page.
Trigger
Type
Enabled
Action
Source
Edit
Delete
New Trigger
604
The index number of this trigger.
The trigger type.
A green check-mark means the trigger is enabled,
and a dash means it is disabled.
The index of the action profile associated with this
trigger. The action profile controls what targets
(LED, Status Relay (Digital Out), SNMP traps and/or
Logging) to invoke for this alarm trigger.
A list of alarm sources associated with this trigger.
For link alarms, this is a list of port numbers, for a
power alarm it is the identifiers for the associated
power sensors, etc.
Click this icon to edit a trigger.
Click this icon to remove a trigger.
Click this button to create a new alarm trigger. You
will be presented to a form where you can configure
the new trigger.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
24.2.3
Create a new alarm trigger using the web interface
Menu path: Configuration ⇒ Alarm ⇒ Triggers ⇒ New Trigger
When clicking the New Trigger button you will be presented to list of trigger
types. Select the trigger type and click next to continue.
Figure 24.7: The trigger type selection page.
When clicking the Next button you will be presented to the New trigger page.
Figure 24.8: The alarm trigger creation page.
Type
The type of alarm trigger.
Continued on next page
© 2017 Westermo Teleindustri AB
605
Westermo OS Management Guide
Version 4.21.1-0
Enabled
Severity Active
Severity Inactive
Condition
Sensors
Threshold Rising
Threshold Falling
Action
Port
606
Continued from previous page
To enable the trigger - check the box, to disable uncheck the box.
Severity level when active
Severity level when inactive
Controls the condition for triggering (High/low)
The sensor source for this trigger
The Rising threshold is the higher threshold value for
the sensor. When the current sample value is higher
than this value, and the last sample was lower than
this value, an action is triggered. Valid for none binary
sensors such as temperature and SNR.
The falling threshold is the lower threshold value for
the sensor. When the current sample value is less than
this value, and the last sample was greater than this
value, an action is triggered. Valid for none binary sensors such as temperature and SNR.
Selects the action for the trigger
The ports on your switch is grouped as on the actual
hardware, in slots. To get alarms for a a specific port,
check the check-box located underneath the port label.
In the picture above you see ports 1/1, 1/2 and 2/1 are
marked as alarm sources for this link alarm trigger.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
24.2.4
Create a new alarm trigger with sensor value
Triggers controlled by an analogue sensor, must be configure with threshold
value. E.g. if you want to create a trigger that alarms if the temperature gets
above a given temperature, you must set the rising threshold value to the alarm
temperature. The falling thresholds may be set to the same value, but by using
different thresholds (rising higher than falling) one can avoid receiving multiple
events when the temperature fluctuates around the alarm threshold.
Figure 24.9: Example of a temperature trigger.
© 2017 Westermo Teleindustri AB
607
Westermo OS Management Guide
Version 4.21.1-0
24.2.5
Action configuration overview page
Menu path: Configuration ⇒ Alarm ⇒ Actions
When entering the Alarm action configuration page you will be presented to a list
of all alarm actions configured on your unit, see below.
Figure 24.10: The alarm action configuration overview page.
Action
Targets
Edit
Delete
New action
608
The index number of this action.
The targets for this action.
Click this icon to edit an action.
Click this icon to remove an action.
Click this button to add a new alarm action. You will
be presented to a form where you can configure the
new action.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
24.3
Managing Alarms via the CLI
The table below shows alarm management features available via the CLI.
Command
Configure Alarm Configuration Settings
alarm
[no] trigger <<INDEX> | <TYPE>>
[no] enable
[no] <port <PORTLIST> |
sensor <SENSORIDLIST> |
ring <FRNTINSTANCE>
timeout <TIMESPEC>
peer <FQDN|IPADDR>
iface <IFNAME>
[no] severity <<LEVEL> |
[active <LEVEL>] |
[inactive <LEVEL>]>
condition <high|low>
threshold <NUM | [rising <NUM>] |
[falling <NUM>]>
[no] interval <SECONDS>
[no] number <NUM>
[no] outbound <IFNAME>
[no] action <INDEX>
show types
[no] action <INDEX>
[no] target <[log] [snmp] [led] >
[digout] [reboot] [custom]>
[no] custom <COMMAND>
[no] summary-trap
Alarm Status
alarm
show
© 2017 Westermo Teleindustri AB
Default
Enabled
Section
Section
Section
Section
Section
24.3.1
24.3.2
24.3.3
24.3.4
Section 24.3.5
rising 0
falling 0
3
Disabled
1
log
Disabled
Section 24.3.6
Section 24.3.7
Section
Section
Section
Section
Section
Section
Section
24.3.8
24.3.9
24.3.10
24.3.11
24.3.12
24.3.13
24.3.14
Section 24.3.15
Section 24.3.16
Section 24.3.17
Section 24.3.18
609
Westermo OS Management Guide
Version 4.21.1-0
24.3.1
Managing Alarm Settings
Syntax alarm
Context Global Configuration context
Usage Enter the Alarm Configuration context.
Use command ”show alarm” to list an overview global alarm settings as
well as configured alarm triggers and actions.
Default values Not applicable.
24.3.2
Manage Alarm Triggers
Syntax [no] trigger <<INDEX> | <TYPE>>
Context Alarm Configuration context
Usage Enter the Alarm Trigger Configuration to create, remove or update an
alarm trigger.
ˆ Use ”trigger <TYPE>” to create a new trigger and enter the Trigger
context, e.g., ”trigger link-alarm” to create a new link-alarm trigger.
Use ”show types” (section 24.3.12) to list supported trigger types.
An index will be assigned to each created index. This index can be used
to update or remove the trigger, see items below.
ˆ Use ”trigger <INDEX>” to manage an existing trigger.
ˆ Use ”no trigger <INDEX>” to remove an existing trigger.
Use command ”show trigger” to list configured alarm triggers. This is
useful to find the index of a trigger, which is needed to edit or remove an
existing trigger, see above.
Default values Not applicable.
Some examples of alarm trigger configurations are given in sections 24.3.2.124.3.2.13. Details of individual alarm trigger configuration settings are given in
sections 24.3.3-24.3.11.
610
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
24.3.2.1
Link Alarm Trigger Configuration Example
Syntax trigger link-alarm
Context Alarm Configuration context
Usage Create a link-alarm trigger, and enter the Alarm Trigger Configuration
context for this trigger.
Additional settings for link-alarm triggers are listed below. The only mandatory setting is the list of ports - no link-alarm alarm events will occur until
ports are defined.
ˆ Port(s) (mandatory): Define the port or ports this link-alarm trigger is
associated with.
ˆ Enable/Disable: By default, the trigger is enabled.
ˆ Severity: By default, active severity is WARNING and inactive severity
is NOTIFY.
ˆ Action: By default, the trigger is mapped to the default action profile
(action 1).
Example
example:/#> configure
example:/config/#> alarm
example:/config/alarm/#> trigger link-alarm
Created trigger 2
example:/config/alarm/trigger-2/#> port 1-2
example:/config/alarm/trigger-2/#> end
example:/config/alarm/#> show
Trigger Type
Enabled Action Source
==============================================
1 power
YES
1 1 2
2 link-alarm YES
1 1 2
Action Targets
===============================================================================
1 snmp log led digout
===============================================================================
Summary alarm traps: Disabled
example:/config/alarm/#>
© 2017 Westermo Teleindustri AB
611
Westermo OS Management Guide
Version 4.21.1-0
24.3.2.2
Digital-In Trigger Configuration Example
Syntax trigger digin
Context Alarm Configuration context
Usage Create a digital-in trigger, and enter the Alarm Trigger Configuration context for this trigger.
Additional settings for digital-in triggers are listed below.
ˆ Sensor: By default, digital-in sensor with ID 1 is used. Use ”show env”
(in Admin Exec context) to list available sensors, see section 7.3.41.
ˆ Condition: By default, the alarm condition is set to low. That is, high is
considered normal and low is considered an alarm situation.
ˆ Enable/Disable: By default, the trigger is enabled.
ˆ Severity: By default, active severity is WARNING and inactive severity
is NOTIFY.
ˆ Action: By default, the trigger is mapped to the default action profile
(action 1).
Example
example:/#> configure
example:/config/#> alarm
example:/config/alarm/#> trigger digin
Created trigger 2
example:/config/alarm/trigger-2/#> end
example:/config/alarm/#> show
Trigger Type
Enabled Action Source
==============================================
1 power
YES
1 1 2
2 digin
YES
1 1
Action Targets
===============================================================================
1 snmp log led digout
===============================================================================
Summary alarm traps: Disabled
example:/config/alarm/#>
612
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
24.3.2.3
Power Trigger Configuration Example
Syntax trigger power
Context Alarm Configuration context
Usage Create a power trigger, and enter the Alarm Trigger Configuration context
for this trigger.
Additional settings for power triggers are listed below. The only mandatory
setting is the list of power sensors - no power alarm events will occur until
power sensors are defined.
ˆ Sensor: WeOS units typically have two power sensors; sensor 1 for DC1
and sensor 2 for DC2. Use ”show env” (in Admin Exec context) to list
available sensors, see section 7.3.41.
ˆ Enable/Disable: By default, the trigger is enabled.
ˆ Severity: By default, active severity is WARNING and inactive severity
is NOTIFY.
ˆ Action: By default, the trigger is mapped to the default action profile
(action 1).
Example
example:/#> configure
example:/config/#> alarm
example:/config/alarm/#> trigger power
Created trigger 1
example:/config/alarm/trigger-1/#> sensor 1,2
example:/config/alarm/trigger-2/#> end
example:/config/alarm/#> show
Trigger Type
Enabled Action Source
==============================================
1 power
YES
1 1 2
Action Targets
===============================================================================
1 snmp log led digout
===============================================================================
Summary alarm traps: Disabled
example:/config/alarm/#>
© 2017 Westermo Teleindustri AB
613
Westermo OS Management Guide
Version 4.21.1-0
24.3.2.4
SNR-Margin Trigger Configuration Example
Note, this setting only applies to units equipped with DSL ports.
Syntax trigger snr-margin
Context Alarm Configuration context
Usage Create a SNR-margin trigger, and enter the Alarm Trigger Configuration
context for this trigger.
Additional settings for SNR-margin triggers are listed below. The only mandatory setting is the list of (DSL) ports - no snr-margin alarm events will occur
until (DSL) ports are defined.
ˆ Port(s) (mandatory): Define the port or ports this SNR-margin trigger is
associated with.
Note: SNR-margin alarms can only be generated for ports where a connection has been established.
ˆ Alarm threshold: As of WeOS v4.21.1 the SNR-margin falling threshold
is set to 3 (dB) by default, and the rising threshold to 6 (dB) by default.
ˆ Enable/Disable: By default, the trigger is enabled.
ˆ Condition: By default, the alarm condition is set to low. That is, high is
considered normal and low is considered an alarm situation.
ˆ Severity: By default, active severity is WARNING and inactive severity
is NOTIFY.
ˆ Action: By default, the trigger is mapped to the default action profile
(action 1).
In this example an SNR-margin trigger is created for DSL ports 1/1 and 1/2,
with falling threshold 4 dB and rising threshold 6 dB.
614
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Example
wolverine:/#> configure
wolverine:/config/#> alarm
wolverine:/config/alarm/#> trigger snr-margin
Created trigger 2
wolverine:/config/alarm/trigger-2/#> port 1/1-1/2
wolverine:/config/alarm/trigger-2/#> threshold falling 4 rising 6
wolverine:/config/alarm/trigger-2/#> end
wolverine:/config/alarm/#> show
Trigger Type
Enabled Action Source
==============================================
1 power
YES
1 1 2
2 snr-margin YES
1 1/1 1/2
Action Targets
===============================================================================
1 snmp log led digout
===============================================================================
Summary alarm traps: Disabled
wolverine:/config/alarm/#>
24.3.2.5
Temperature Trigger Configuration Example
Syntax trigger temperature
Context Alarm Configuration context
Usage Create a temperature trigger, and enter the Alarm Trigger Configuration
context for this trigger.
Additional settings for temperature triggers are listed below. The only mandatory setting is the temperature sensor (or list of sensors) - no temperature
alarm events will occur until a sensor is defined.
ˆ Sensor(s): Define the temperature sensor(s) this temperature trigger is
associated with (default is temperature sensor is ”1”). Use ”show env”
(in Admin Exec context) to list available sensors, see section 7.3.41.
ˆ Alarm threshold: As of WeOS v4.21.1 the temperature falling threshold
and rising threshold are both set to 0◦ C by default.
ˆ Enable/Disable: By default, the trigger is enabled.
ˆ Condition: By default, the alarm condition is set to high. That is, temperatures below the falling threshold are considered normal, and temperatures above the rising threshold is considered an alarm situation.
© 2017 Westermo Teleindustri AB
615
Westermo OS Management Guide
Version 4.21.1-0
ˆ Severity: By default, active severity is WARNING and inactive severity
is NOTIFY.
ˆ Action: By default, the trigger is mapped to the default action profile
(action 1).
In this example two temperature triggers are created, one to give alarm if
the temperature drops below 10◦ C, and a second trigger to create an alarm
if the temperature rises above 60◦ C.
Example
example:/config/alarm/#> trigger temperature
example:/config/alarm/trigger-2/#> sensor 1
example:/config/alarm/trigger-2/#> threshold falling -10 rising -5
example:/config/alarm/trigger-2/#> condition low
example:/config/alarm/trigger-2/#> end
example:/config/alarm/#> trigger temperature
example:/config/alarm/trigger-3/#> sensor 1
example:/config/alarm/trigger-3/#> threshold falling 55 rising 60
example:/config/alarm/trigger-3/#> condition high
example:/config/alarm/trigger-3/#> end
example:/config/alarm/#> show
Trigger Type
Enabled Action Source
===============================================================================
1 frnt
YES
1 1
2 temperature YES
1 1
3 temperature YES
1 1
Action Targets
===============================================================================
1 snmp log led digout
===============================================================================
Summary alarm traps: Disabled
example:/config/alarm/#>
24.3.2.6
FRNT Trigger Configuration Example
An FRNT trigger exists in the factory default configuration. Thus, when FRNT
is enabled, FRNT alarms will be presented on the default alarm targets without
requiring the user to create a trigger.
Syntax trigger frnt
Context Alarm Configuration context
Usage Create an FRNT trigger, and enter the Alarm Trigger Configuration context
for this trigger.
616
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Additional settings for FRNT triggers are listed below.
ˆ Ring: By default, FRNT ring ID 1 is used (as of WeOS v4.21.1 only
a single FRNT ring is supported, thus other values are invalid.) Use
”show env” (in Admin Exec context) to list available sensors, see section 7.3.41.
ˆ Condition: By default, the alarm condition is set to down (or low). That
is, ring status up (high) is considered normal and ring down (low) is
considered an alarm situation.
ˆ Enable/Disable: By default, the trigger is enabled.
ˆ Severity: By default, active severity is WARNING and inactive severity
is NOTIFY.
ˆ Action: By default, the trigger is mapped to the default action profile
(action 1).
Example
example:/#> configure
example:/config/#> alarm
example:/config/alarm/#> trigger frnt
example:/config/alarm/trigger-2/#> end
example:/config/alarm/#> show
Trigger Type
Enabled
Action Source
===============================================================================
1 power
YES
1 1 2
2 frnt
YES
1 Instance 1
Action Targets
===============================================================================
1 snmp log led digout
===============================================================================
Summary alarm traps: Disabled
example:/config/alarm/#>
24.3.2.7
RiCo Uplink Trigger Configuration Example
Syntax trigger rico-uplink
Context Alarm Configuration context
Usage Create an RiCo uplink trigger, and enter the Alarm Trigger Configuration
context for this trigger. This trigger applies both to RiCo and Dual-Homing
uplinks.
© 2017 Westermo Teleindustri AB
617
Westermo OS Management Guide
Version 4.21.1-0
Additional settings for RiCo uplink triggers are listed below.
ˆ Port: One or more RiCo/Dual-Homing uplink ports can be set for the
trigger.
ˆ Condition: By default, the alarm condition is set to down (or low). That
is, uplink status up (high) is considered normal and uplink status down
(low) is considered an alarm situation.
ˆ Enable/Disable: By default, the trigger is enabled.
ˆ Severity: By default, active severity is WARNING and inactive severity
is NOTIFY.
ˆ Action: By default, the trigger is mapped to the default action profile
(action 1).
Example
example:/#> configure
example:/config/#> alarm
example:/config/alarm/#> trigger rico-uplink
example:/config/alarm/trigger-2/#> port 1/1
example:/config/alarm/trigger-2/#> end
example:/config/alarm/#> show
Trigger Type
Enabled
Action Source
===============================================================================
1 frnt
YES
1 Instance 1
2 rico-uplink
YES
1 1/1
Action Targets
===============================================================================
1 snmp log led digout
===============================================================================
Summary alarm traps: Disabled
example:/config/alarm/#>
24.3.2.8
LFF Trigger Configuration Example
Note, this setting only applies to units equipped with SHDSL ports.
Syntax trigger lff
Context Alarm Configuration context
Usage Create a Link Fault Forward (LFF) trigger, and enter the Alarm Trigger
Configuration context for this trigger.
618
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Additional settings for LFF triggers are listed below. The only mandatory
setting is the list of (SHDSL) ports - no LFF alarm events will occur until
(SHDSL) ports are defined.
ˆ Port(s) (mandatory): Define the port or ports this LFF trigger is associated with.
Note: LFF alarms are generated both when detecting that the remote
SHDSL switch indicated LFF, or when the SHDSL link is down.
ˆ Enable/Disable: By default, the trigger is enabled.
ˆ Condition: By default, the alarm condition is set to low. That is, high
(remote link ”up”) is considered normal and low (remote link ”down”) is
considered an alarm situation.
ˆ Severity: By default, active severity is WARNING and inactive severity
is NOTIFY.
ˆ Action: By default, the trigger is mapped to the default action profile
(action 1).
In this example an LFF trigger is created to monitor incoming LFF indications
on SHDSL port 1/1.
Example
wolverine:/config/alarm/#> trigger lff
wolverine:/config/alarm/trigger-2/#> port 1/1
wolverine:/config/alarm/trigger-2/#> end
wolverine:/config/alarm/#> show
Trigger Type
Enabled Action Source
===============================================================================
1 frnt
YES
1 1
2 lff
YES
1 dsl 1/1
Action Targets
===============================================================================
1 snmp log led digout
===============================================================================
Summary alarm traps: Disabled
wolverine:/config/alarm/#>
24.3.2.9
Timer Trigger Configuration Example
Syntax trigger timer
© 2017 Westermo Teleindustri AB
619
Westermo OS Management Guide
Version 4.21.1-0
Context Alarm Configuration context
Usage Create a timer trigger, and enter the Alarm Trigger Configuration context
for this trigger.
Additional settings for timer triggers are listed below.
ˆ Timeout time: As of WeOS v4.21.1, only daily timeouts can be specified,
e.g., ”timeout daily 02:30”
ˆ Enable/Disable: By default, the trigger is enabled.
ˆ Condition: The condition setting has no meaning for a timer trigger,
since as of WeOS v4.21.1 the timer trigger should not affect the ON LED
or the status relay (Digital Out) action targets.
ˆ Severity: By default, active severity is WARNING and inactive severity
is NOTIFY.
ˆ Action: By default, the trigger is mapped to the default action profile
(action 1).
In this example a timer trigger is created to force a switch reboot daily at
02:30 in the morning.
Example
example:/config/alarm/#> trigger timer
example:/config/alarm/trigger-2/#> timeout daily 02:30
example:/config/alarm/trigger-2/#> action 2
example:/config/alarm/trigger-2/#> end
example:/config/alarm/#> action 2
example:/config/alarm/action-2/#> target log reboot
example:/config/alarm/action-2/#> end
example:/config/alarm/#> show
Trigger Class
Enabled Action Source
===============================================================================
1 frnt
YES
1 Instance 1
2 timer
YES
2 daily 02:30
Action Targets
===============================================================================
1 snmp log led digout
2 log reboot
===============================================================================
Summary alarm traps: Disabled
620
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
24.3.2.10
Ping Trigger Configuration Example
Syntax trigger ping
Context Alarm Configuration context
Usage Create a ping trigger, and enter the Alarm Trigger Configuration context
for this trigger. The ping trigger monitors the network connectivity (i.e.,
network reachability) to a given host, using the ping command.
Associated with the ping trigger are the following settings:
ˆ peer: The host to test the connectivity against.
ˆ interval: the ping interval can be configured (see section 24.3.8)
ˆ number: a robustness threshold, i.e., number of failed (or successful,
depending on the condition) pings required to consider the remote host
to be unreachable (or reachable), see section 24.3.9)
ˆ outbound: to force ping to use a specific interface. Useful with dynamic
VRRP priority (see section 30.1.1), where you do not want to rely on the
system default gateway.
In this example a ping trigger is created and mapped to the default action profile, to indicate alarm when the peer become unreachable after 3
retries.
Example
example:/config/alarm/#> trigger ping
Trigger 2: Peer is mandatory
example:/config/alarm/trigger-2/#> peer bbc.co.uk
example:/config/alarm/trigger-2/#> number 3
example:/config/alarm/trigger-2/#> interval 3
example:/config/alarm/trigger-2/#> action 2
example:/config/alarm/trigger-2/#> end
example:/config/alarm/#> show
Trigger Type
Enabled Action Source
===============================================================================
1 frnt
YES
1 Instance 1
2 ping
YES
1 peer bbc.co.uk
Action Targets
===============================================================================
1 snmp log led digout
===============================================================================
Summary alarm traps: Disabled
example:/config/alarm/#>
© 2017 Westermo Teleindustri AB
621
Westermo OS Management Guide
Version 4.21.1-0
In this example a ping trigger is created to trigger the status relay (digital out) when the peer become reachable, to do this change the condition
argument (default: low).
Example
example:/config/alarm/#> trigger ping
Trigger 2: Peer is mandatory
example:/config/alarm/trigger-2/#> peer bbc.co.uk
example:/config/alarm/trigger-2/#> number 3
example:/config/alarm/trigger-2/#> interval 3
example:/config/alarm/trigger-2/#> condition high
example:/config/alarm/trigger-2/#> action 2
example:/config/alarm/trigger-2/#> end
example:/config/alarm/#> action 2
example:/config/alarm/action-2/#> target digout
example:/config/alarm/action-2/#> end
example:/config/alarm/#> show
Trigger Type
Enabled Action Source
===============================================================================
1 frnt
YES
1 Instance 1
2 ping
YES
2 peer bbc.co.uk
Action Targets
===============================================================================
1 snmp log led digout
2 log digout
===============================================================================
Summary alarm traps: Disabled
24.3.2.11
Address Conflict Trigger Configuration Example
Syntax trigger address-conflict
Context Alarm Configuration context
Usage Create an Address Conflict trigger, and enter the Alarm Trigger Configuration context for this trigger. Mandatory setting is the VLAN network interface
to detect conflicts on. Optional setting is by what interval (1-300 seconds)
active ARP probing for conflicts should run (default 60).
Note
One VLAN interface per is allowed. Create an additional trigger activate
address conflict detection on another VLAN.
ˆ Interface: Set a valid VLAN interface, such as ”iface vlan1”.
622
By
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
default ”no iface” is set, which implies inactive trigger.
tion 24.3.4.
See sec-
ˆ Interval: Set preferred sample interval in seconds (1-300). By default,
60 seconds. See section 24.3.8
ˆ Severity: By default, active severity is WARNING and inactive severity
is NOTIFY.
ˆ Action: By default, the trigger is mapped to the default action profile
(action 1).
Example
example:/config/alarm/#> trigger address-conflict
example:/config/alarm/trigger-1/#> iface vlan1
example:/config/alarm/trigger-1/#> interval 100
example:/config/alarm/trigger-1/#> end
example:/config/#> show alarm
Trigger Type
Enabled
Action Source
===============================================================================
1 address-conflict YES
1 vlan1
Action Targets
===============================================================================
1 snmp log led digout
===============================================================================
Summary alarm traps: Disabled
24.3.2.12
PoE Power Usage Trigger Configuration Example
Syntax trigger poe
Context Alarm Configuration context
Usage Create a PoE power usage trigger, and enter the Alarm Trigger Configuration context for this trigger. The power usage is defined as the percentage
of consumed/maximum power.
Additional settings for temperature triggers are listed below. The only mandatory setting is the temperature sensor (or list of sensors) - no temperature
alarm events will occur until a sensor is defined.
ˆ Alarm threshold: Set the threshold to usage level (1-99 (%)) when an
alarm is desired. By default, the rising threshold is set to 95(%) and the
falling threshold to 90(%).
© 2017 Westermo Teleindustri AB
623
Westermo OS Management Guide
Version 4.21.1-0
ˆ Enable/Disable: By default, the trigger is enabled.
ˆ Condition: By default, the alarm condition is set to high. That is, usage
levels below the falling threshold are considered normal, and temperatures above the rising threshold is considered an alarm situation.
ˆ Severity: By default, active severity is WARNING and inactive severity
is NOTIFY.
ˆ Action: By default, the trigger is mapped to the default action profile
(action 1).
In this example a PoE trigger is created to give an alarm if the usage rises
above 80%.
Example
viper:/config/alarm/#> trigger poe
viper:/config/alarm/trigger-2/#> threshold rising 80 falling 75
viper:/config/alarm/trigger-2/#> condition high
viper:/config/alarm/trigger-2/#> end
viper:/config/alarm/#> show
Trigger Type
Enabled
Action Source
===============================================================================
1 frnt
YES
1 Instance 1
2 poe
YES
1 1
Action Targets
===============================================================================
1 snmp log led
===============================================================================
Summary alarm traps: Disabled
viper:/config/alarm/#>
24.3.2.13
Microlok Trigger Configuration Example
Syntax trigger microlok
Context Alarm Configuration context
Usage Create a Microlok session summary alarm trigger, and enter the Alarm
Trigger Configuration context for this trigger.
Additional settings for Microlok triggers are listed below. As of WeOS v4.21.1
there can only be one Microlok Gateway instance, thus the gateway instance
(i.e., instance 1) is implicit.
624
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
ˆ Enable/Disable: By default, the trigger is enabled.
ˆ Severity: By default, active severity is WARNING and inactive severity
is NOTIFY.
ˆ Action: By default, the trigger is mapped to the default action profile
(action 1).
Example
example:/config/microlok-1/#> map station 1a serial 1 session-timeout 2000
example:/config/microlok-1/#> map station 1b serial 1 session-timeout 2000
example:/config/microlok-1/#> map station 2a remote 192.168.2.1 sessiontimeout 2000
example:/config/microlok-1/#> map station 2b remote 192.168.2.1 sessiontimeout 2000
example:/config/microlok-1/#> end
example:/config/#> alarm
example:/config/alarm/#> trigger microlok
example:/config/alarm/trigger-2/#> action 2
example:/config/alarm/trigger-2/#> end
example:/config/alarm/#> action 2
example:/config/alarm/action-2/#> target log digout
example:/config/alarm/action-2/#> end
example:/config/alarm/#> show
Trigger Type
Enabled
Action Source
===============================================================================
1 frnt
YES
1 Instance 1
2 microlok
YES
2 1
Action Targets
===============================================================================
1 snmp log led digout
2 log digout
===============================================================================
Summary alarm traps: Disabled
example:/config/alarm/#>
24.3.3
Enable/disable a Trigger
Syntax [no] enable
Context Alarm Trigger Configuration context
Usage Enable or disable an alarm trigger. A disabled trigger will keep its configuration settings, but will not affect any alarm targets.
Use ”enable” to enable and ”no enable” to disable a trigger.
© 2017 Westermo Teleindustri AB
625
Westermo OS Management Guide
Version 4.21.1-0
Use ”show enable” to show whether this trigger is enabled or disabled.
Default values Enabled
24.3.4
Manage alarm sources
Syntax [no] <port <PORTLIST> | sensor <SENSORIDLIST> |
ring <FRNTINSTANCE> timeout <daily <HH:MM>>>
Context Alarm Trigger Configuration context
Usage Specify which alarm sources the trigger should monitor. The command
syntax differs depending on the trigger type:
ˆ Use ”[no] port <PORTLIST>” to specify which port(s) a link-alarm or
rico-uplink trigger should apply to, e.g., use ”port 1/1,2/2-2/4” to
add ports 1/1, and 2/2-2/4 to the list of ports monitored by this trigger.
ˆ Use ”[no] ring <FRNTINSTANCE>” to specify which FRNT ring an FRNT
alarm trigger should apply to.
ˆ Use ”[no] sensor <SENSORIDLIST>” to specify which sensors a digital in, power or temperature trigger should apply to, e.g., use ”sensor
1,2” to add power sensors 1 and 2 to the list of power sensors monitored by this power trigger.
Use command show env (section 7.3.41) to list available sensors and
their index values.
ˆ Use ”[no] timeout <daily <HH:MM>>” to specify how often and when
an timer trigger should go off, e.g., use ”timeout daily 02:30” to
make the timer trigger to go off every day at 02:30 in the morning.
ˆ Use ”[no] peer <FQDN|IPADDR>” to specify the peer (domain name or
IP address) to test the connectivity to.
”no peer” will delete the configured peer, however, having a ping trigger without a configured peer is not a valid setting.
ˆ Use ”[no] iface <IFNAME>” to configure the VLAN interface to enable
address conflict detection on. This is a mandatory setting for Address
Conflict triggers. When not set, or a non-valid VLAN interface is set,
then no iface is written to configuration.
626
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Use ”no port <PORTLIST>” remove a specific set of ports, or ”no port” to
remove all ports from a trigger (the same goes for other source types).
If no sources are defined when exiting the trigger context, the trigger will
automatically be configured as disabled (see section 24.3.3).
Use command ”show ” to show the alarm sources associated with this trigger. The type of alarm source differs depending on the trigger type. See
section 24.3.4 for more information.
Default values
24.3.5
Alarm Event Severity
Syntax [no] severity <<LEVEL>|[active <LEVEL>]|[inactive <LEVEL>]>
Context Alarm Trigger Configuration context
Usage Specify the severity level of active and inactive alarm events detected by
this trigger. See section 24.1.3.4 for information on available severity levels.
Active and inactive severity levels can be configured together or independently.
”no severity” to will set severity to level NONE. Alarm events with severity
NONE will not cause SNMP traps to be sent or events to be logged, however,
such events can still affect the status relay (digital-out) and ON LED targets.
Use ”show severity” to show the severity setting (active and inactive severity) for this trigger.
Default values active warning and inactive notice
The examples below show how to set severity level for active and inactive alarm events together and how to set it individually. The final example
shows how to set severity ’NONE’ for both active and inactive events.
© 2017 Westermo Teleindustri AB
627
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/config/alarm/trigger-2/#>
example:/config/alarm/trigger-2/#>
active err, inactive err
example:/config/alarm/trigger-2/#>
example:/config/alarm/trigger-2/#>
active err, inactive debug
example:/config/alarm/trigger-2/#>
example:/config/alarm/trigger-2/#>
example:/config/alarm/trigger-2/#>
active none, inactive none
example:/config/alarm/trigger-2/#>
24.3.6
severity err
show severity
severity inactive debug
show severity
no severity
show severity
Configure Alarm Condition Setting
Syntax condition <high|low>
Alternate keywords are possible:
ˆ rising and up are equivalents to high.
ˆ falling and down are equivalents to low.
Context Alarm Trigger Configuration context
Usage Define whether the high or low trigger state should be considered the
alarm state, while the other is considered the normal state.
Some triggers, such as link-alarm and power triggers have a static (predefined) alarm condition setting. (Both link-alarm and power triggers have
condition set to low). For other triggers, the alarm condition setting is configurable.
See section 24.1.3.2 for more information.
Use ”show condition” to show the alarm condition setting for this trigger.
Default values Differs for different trigger types
24.3.7
Configure Rising and Falling Thresholds
Syntax threshold <NUM|[rising <NUM>]|[falling <NUM>]>
Context Alarm Trigger Configuration context
628
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Usage Set falling and rising thresholds. The thresholds may be set to the same
value, but by using different thresholds (rising higher than falling) one can
avoid receiving multiple events when the alarm source fluctuates around
the alarm threshold.
Triggers which are binary to their nature, such as link-alarm, power, and
digital-in triggers have implicit thresholds, which cannot be configured.
See section 24.1.3.2 for more information.
Use command ”show threshold” to show the trigger threshold setting (both
rising and falling thresholds) for this trigger.
Default values rising 0 and falling 0 (except for binary alarm sources)
24.3.8
Configure Interval (Ping or Address Conflict)
Syntax [no] interval <SEC>
Usage There are more than one use for interval option: Ping trigger interval and
Address Conflict trigger interval.
ˆ Ping trigger: Specify the interval between ICMP Pings, i.e., interval at
which ping messages are sent to probe the reachability to the peer.
ˆ Address Conflict: Specify the interval between active check after conflicting addresses, i.e., interval at which ARP messages are sent to probe
the VLAN for conflicting addresses.
Use command ”show interval” to show the configured interval setting.
Use command ”no interval” to reset interval to default.
Default values Differs:
ˆ Ping trigger: 3 (seconds)
ˆ Address Conflict trigger: 60 (seconds)
24.3.9
Configure Ping Robustness Number
Syntax [no] number <NUM>
Context Alarm Trigger Configuration context (ping trigger)
© 2017 Westermo Teleindustri AB
629
Westermo OS Management Guide
Version 4.21.1-0
Usage Specify the number of ICMP ping that should be lost (or received) to determine if a host is unreachable (or reachable).
Use command ”show number” to show the ping trigger robustness number
setting, i.e., the number of pings required to be lost before the peer is considered unreachable, or the number of pings required to succeed before the
peer is considered reachable.
24.3.10
Configure Ping Outbound Interface
Syntax [no] outbound <IFNAME>
Context Alarm Trigger Configuration context (ping trigger)
Usage Force pings to use a specific outbound interface. This is very useful when
tracking upstreams connectivity in a VRRP dynamic priority scenario (see
section 30.1.1). Because then you want to make sure the default gateway,
or any other route, is avoided.
Use ”no outbound” to disable the setting. This makes ping rely on network
routes and fall back to use the default gateway.
Use command ”show outbound” to show the configured outbound interface
for this ping trigger. When unset, ”Default Gateway” is shown and the
system will use the system default route, or a matching network route, for
ping packets.
Default values Disabled (default gateway)
24.3.11
Configure Trigger Action
Syntax [no] action <INDEX>
Context Alarm Trigger Configuration context
Usage Specify the action (profile) to be invoked when this trigger detects an
alarm event.
Use ”no action” to disable the mapping to an alarm action. E.g., when
in use by another subsystem (e.g., VRRP with dynamic priority, see section 30.1.1), or if you simply want to temporarily disable or debug your
alarms.
630
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Use command ”show action” to show the action profile mapped to this
trigger.
Default values 1 (default action)
24.3.12
Show Supported Trigger Types
Syntax show types
Context Alarm Configuration context
Usage List supported trigger types. These are the types to be used with the
”trigger <TYPE>” command (see section 24.3.2).
Default values Not applicable
24.3.13
Manage Alarm Actions
Syntax [no] action <INDEX>
Context Alarm Configuration context
Usage Create, remove or update an alarm action (profile). Use ”action <INDEX>”
to enter the Alarm Action Configuration context and create a new or update
an existing action.
Use ”no action <INDEX>” remove an existing action. The default action
(index 1) cannot be removed, but you can disable all targets.
Use command ”show action” to list all configured alarm action profiles, or
”show action <ID>” to show detailed configuration information on a specific action profile (also available as ”show” command within the Alarm Action Configuration of that profile).
Default values Not applicable.
24.3.14
Manage Action Targets
Syntax [no] target <[log] [snmp] [led] [digout] [reboot] [custom]>
Context Alarm Action Configuration context
© 2017 Westermo Teleindustri AB
631
Westermo OS Management Guide
Version 4.21.1-0
Usage Add or remove alarm target to an alarm action (profile).
ˆ led: Set ON/Status LED
ˆ log: Log status change to syslog
ˆ snmp: Generate an SNMP trap
ˆ digout: Control the status relay (digital out)
ˆ reboot: Reboot the unit. USE WITH CAUTION!
ˆ custom: Run any admin-exec level command. DEPRECATED!
Warning
The ”custom” target is for experimental purposes only! A .conf file containing ”target custom” and ”custom reboot” (see section 24.3.15)
will be translated to ”target reboot” automatically. That is to be
backwards compatible. Other ”custom” commands are not guaranteed
to be supported in future releases.
Use command ”show target” to show the alarm target(s) configured for
this action profile.
Default values target log (New action profiles has ”target log” as default.
24.3.15
Set Custom Action Target
Syntax [no] custom <COMMAND>
Context Alarm Action Configuration context
Usage Set custom action command. The custom target allows the user to connect, e.g., a timer trigger to a CLI Admin Exec level command, such as
”reboot”, see section 7.3.28.
Warning
This is a deprecated feature not guaranteed to be supported in future
releases. For experimental purposes only!
Use ”no custom” to remove a custom command.
Use command ”show custom” to show the configured custom action command configured for this action profile.
632
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Default values Disabled
Examples See section 24.3.2.9.
24.3.16
Enable/disable Summary Alarm Traps
Syntax [no] summary-trap
Context Alarm Configuration context
Usage Enable or disable summary alarm traps. When enabled, a trap will be sent
whenever the summary alarm status changes (from OK to Warning or vice
versa). The summary alarm status follows the status of the ON LED. See also
section 6.1.3 for more information summary alarm status and its associated
SNMP trap, and see sections 24.1.5.1 and 24.3.14 for more information on
the ON LED alarm target.
Use ”summary-trap” to enable and ”no summary-trap” to disable a SNMP
traps for the summary alarm status.
Use ”show summary-trap” to show whether summary alarm traps are enabled or disabled.
Default values Disabled
Example
example:#> configure
example:/config/#> alarm
example:/config/alarm/#> summary-trap
example:/config/alarm/#> show summary-trap
Enabled
example:/config/alarm/#> end
example:/config/#>
24.3.17
Handling Alarm Status
Syntax alarm
Context Admin Exec context
Usage Enter the Alarm Status context.
Default values Not applicable.
© 2017 Westermo Teleindustri AB
633
Westermo OS Management Guide
Version 4.21.1-0
24.3.18
Show overall alarm status
Syntax show
Context Alarm Status context
Usage Show status of all alarms.
Default values Not applicable.
634
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
24.4
Digital I/O
WeOS products are typically equipped with a Digital I/O connector as the one
shown in fig. 24.11. The location of the connector differs between products; on
RedFox Industrial it is located on the CPU card as shown in fig. 24.12).
1
2
3
4
Figure 24.11: Digital I/O connector.
The Pin-Out of the Digital I/O connector is typically as follows:
Position
1
2
3
4
Description
Status +
Status Digital-In +
Digital-In -
Note
For a detailed specification on the Digital I/O connector (definite pin-out
mapping, voltage levels, etc.), see the User Guide of your specific WeOS
product (section 1.5).
Status+
Status−
No. 1
No. 2
Westermo switch
Digital In+
Digital In−
No. 3
No. 4
© 2017 Westermo Teleindustri AB
635
Westermo OS Management Guide
Version 4.21.1-0
ON
DC1
DC2
IO
1
2
POWER
+DC1
CONSOLE
+DC2
COM
FRNT
COM
ST1
ST2
Figure 24.12: The Power and CPU module of a RedFox Industrial unit
As described in section 24.1, Digital-In can be used as an alarm source, while the
Status Relay (Digital-Out) is utilised as an alarm target (summary alarm).
ˆ The Digital-In alarm is triggered when there is lack of voltage on the DigitalIn pins. For information on appropriate voltage/current levels to trigger
alarms via Digital-In, see the User Guide of your specific product (section 1.5).
ˆ The Status Relay pins (Digital-Out) are internally connected to a gate. The
gate is open when the switch has no power, or when any alarm sources are
active. When the switch is operating normally (the switch has booted up,
and no alarm source is active), the gate is closed.
636
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
24.5
LEDs
The LED functionality when running WeOS is described in the User Guide of your
product (section 1.5). Here the information on LED functionality of all WeOS
products is summarised. Note that your product may not have all LED types
listed here.
LED
ON
Status
OFF
GREEN
RED
GREEN BLINK
RED BLINK
DC1
DC2
AC1
DC1
DC2
OFF
GREEN
RED
OFF
GREEN
RED
OFF
GREEN
OFF
GREEN
RED
OFF
GREEN
RED
Description
Unit has no power
All OK, no alarm condition.
Alarm condition, or until unit has started up.
(Alarm conditions are configurable, see sections 24.1-24.3.)
Location indicator (”Here I am!”). Activated
when connected to WeConfig Tool, or upon request from Web, or when entering the CLI configuration context. Duration of blinking: 10
seconds.
Location indicator (see previous item) or indication of pending cable factory reset, see section 7.1.3.3.
Unit has no power.
Power OK on DC1.
Power failure on DC1.
Unit has no power.
Power OK on DC2.
Power failure on DC2.
Unit has no power.
Power OK on AC1.
Unit has no power.
Power OK on DC1 and DC2.
Power failure on DC1 or DC2.
Unit has no power.
Power OK. Input voltage > 70% of minimum
nominal voltage.
Power failure. Input voltage < 70% of minimum nominal voltage.
Continued on next page
© 2017 Westermo Teleindustri AB
637
Westermo OS Management Guide
Version 4.21.1-0
LED
FRNT
RSTP
(formerly
ST1)
USR1/VPN3
(formerly
ST2)
Ethernet
ports
SHDSL
ports
Status
OFF
GREEN
RED
BLINK
OFF
GREEN
BLINK
OFF
GREEN
RED
OFF
GREEN
GREEN FLASH
YELLOW
OFF
GREEN
GREEN BLINK
GREEN FLASH
YELLOW
YELLOW BLINK
SHDSL
Link
Quality
Indicator
ADSL/
VDSL
638
All OFF
3 RED
1 GREEN
2 GREEN
3 GREEN
OFF
GREEN
Continued from previous page
Description
FRNT disabled
FRNT OK. (See also the FRNT Error item below.)
FRNT Error. A focal point can detect and indicate local FRNT errors (FRNT link down) as well
as FRNT errors elsewhere in the FRNT ring. A
member switch only detects and indicates local FRNT errors (FRNT link down).
Unit configured as focal point.
RSTP disabled.
RSTP enabled.
Unit elected as RSTP/STP root switch.
VPN disabled4 .
At least one VPN tunnel up and OK4 .
All VPN tunnels down4 .
No link.
Link established.
Data traffic indication.
Port alarm, or port is set in blocking state by
link redundancy protocol.
No SHDSL link.
SHDSL link established.
SHDSL link negotiation.
Data traffic indication.
Port alarm, or port is set in blocking state by
link redundancy protocol.
Only during unit startup. Firmware downloading to SHDSL chip.
No SHDSL link.
SNR below 3 dB. Unstable SHDSL link.
SNR 3-5 dB. Marginal SHDSL link.
SNR 6-9 dB. Normal SHDSL link.
SNR 10 dB or higher. Strong SHDSL link.
No xDSL link.
xDSL link established.
Continued on next page
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
LED
ports
TD
RD
Status
GREEN BLINK
OFF
GREEN FLASH
YELLOW FLASH5
OFF
GREEN FLASH
Continued from previous page
Description
xDSL link negotiation.
No serial data received.
Serial data received.
Error on RS-422/485 bus.
No serial data transmitted.
Serial data transmitted.
Additional explanations:
ˆ BLINK means that the LED is blinking with a frequency about 1 Hz.
ˆ FLASH means that the LED is blinking with a higher frequency.
ˆ SHDSL LEDs only apply to products with SHDSL ports.
ˆ xDSL (ADSL/VDSL) LEDs only apply to products with xDSL ports.
ˆ TD and RD LEDs only apply to products with serial port(s). As the WeOS
serial ports operate in DCE mode, TD denotes receiving, and RD denotes
transmitting serial data.
1 Some
Viper units (see the user guide of your product, listed in section 1.5) have two DC feeds
(DC1 and DC2) with a common power indicator LED named ”DC”.
2 Some Viper units (see the user guide of your product, listed in section 1.5) have a single DC
feed with an indicator LED named ”DC”.
3 The ”USR1” LED is referred to as ”VPN” on some WeOS products and ”ST2” on older RFI products.
4 Only for products with software level WeOS Extended. As of WeOS v4.21.1, the USR1/VPN LED
presents VPN status as described above. Alternative (configurable) use is intended but not yet
supported.
5 Only applicable for the DDW-x42-485[43] products.
© 2017 Westermo Teleindustri AB
639
Westermo OS Management Guide
Version 4.21.1-0
Chapter 25
Logging Support
This chapter describes WeOS support for alarm and generic event logging.
In WeOS general events detected by the system (such as user login attempts), as
well as alarm events defined by configured alarm triggers (see chapter 24) can
be logged for further analysis. Three logging methods are available:
ˆ Logging to file: General events and alarm events are always logged to a
local log file.
ˆ Logging to console: It is possible to direct logging messages to the console
port. Messages of severity level DEBUG or higher are shown on the console
port.
ˆ Logging to a remote syslog server: Logging messages can be sent to a
remote syslog server for further processing. Messages of severity level NOTICE or higher are forwarded to the remote syslog server(s).
As of WeOS v4.21.1 logging support is only available via the CLI. The severity
thresholds for console and remote syslog logging are not configurable, however,
such support is planned.
640
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
25.1
Logging Support in the web interface
Select the log file in the drop down list and press View to the display desired log
file.
Menu path: Maintenance ⇒View Log
Select the log file in the drop down list and press View to the display desired log
file.
© 2017 Westermo Teleindustri AB
641
Westermo OS Management Guide
Version 4.21.1-0
25.2
Managing Logging Support via the CLI
Command
Configuring Logging Settings
[no] logging
[no] console
[no] server <ADDRESS1 [ADDRESS2]>
Managing Log Files
dir <cfg://|log://|usb://>
copy <FROM_FILE> <TO_FILE>
erase <file>
show <running-config | startup-config |
factory-config | [<filesys>://]FILENAME>
25.2.1
Default
Section
Disabled
Section 25.2.1
Section 25.2.2
Section 25.2.3
Disabled
Section
Section
Section
Section
7.3.22
7.3.23
7.3.24
7.3.25
Managing Logging Settings
Syntax [no] logging
Context Global Configuration context
Usage Enter Logging Configuration context.
Use ”no logging” to disable all logging (to console and remote syslog
server).
Use ”show logging” to show logging configuration settings. Also available
as ”show” command within the Logging Configuration context.
Default values Disabled
25.2.2
Logging to console port
Syntax [no] console
Context Logging Configuration context
Usage Enable or disable console logging.
642
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Use ”console” to enable console logging, and ”no console” to disable console logging.
When enabled, general events detected by the system, as well as alarm
events associated with configured alarm triggers, will be presented on the
console port.
Use ”show console” to show whether console port logging is enabled or
disabled.
Default values Disabled
25.2.3
Logging to remote syslog server
Syntax [no] server <ADDRESS1 [ADDRESS2]>
Context Logging Configuration context
Usage Set remote syslog server(s) (IPv4 addresses). A maximum of two remote
syslog servers are supported. The syntax allows typing them in one line or
two separate lines.
Use ”no server <ADDRESS>” to remove a single server. Use ”no server”
to remove all servers.
When enabled, general events detected by the system, as well as alarm
events associated with configured alarm triggers, will be forwarded to the
configured syslog server via UDP to port 514. If two servers are configured,
messages are sent to both of them.
Use ”show server” to show whether remote syslog logging is enabled or
disabled. If enabled, the IP address(es) of the configured server(s) are presented.
Default values Disabled
© 2017 Westermo Teleindustri AB
643
Westermo OS Management Guide
Version 4.21.1-0
Part III
Router/Gateway Services
644
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Chapter 26
IP Routing in WeOS
In addition to switching (layer-2), WeOS devices (with proper WeOS level) are able
to route data packets (layer-3), i.e., they are routing switches. The WeOS routing
support includes static routing and dynamic unicast routing via OSPF and RIP,
static multicast routing, as well as other useful router features such as firewall,
NAT, and VRRP.
This chapter introduces the IP routing capabilities in WeOS in general. More information on dynamic routing is found in chapters 27 (OSPF) and 28 (RIP), while
static multicast routing support is described in chapter 29. Supplementary router
services are covered in the chapters to follow: VRRP in chapter 30, and firewall
and NAT in chapter 31.
Support for VPN and tunneling techniques are presented separately, see part IV.
26.1
Summary of WeOS Routing and Router Features
Table 26.1 presents the routing and router features available in WeOS.
26.1.1
Introduction to WeOS Routing and Router Features
IP routing enables us to connect our networks together, and to let (TCP/IP) devices
communicate across networks of different type and topology, and possibly over
multiple network ”hops” and long distances. A router looks at the destination
IP address carried within each IP packet, consults its routing table to make a
© 2017 Westermo Teleindustri AB
645
Westermo OS Management Guide
Version 4.21.1-0
Feature
Enable/disable routing
Default gateway
Static unicast routing
Blackhole routes
Dynamic unicast routing
- OSPF
- RIP (v1/v2)
Static multicast routing
View routing table
Router redundancy (VRRP)
Firewall and NAT
Web
X
X
X
CLI
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
General Description
Section 26.1.1
Section 26.1.1
Section 26.1.4
Section 26.1.4.3
Section 26.1.1, Chapter 27
Section 26.1.1, Chapter 28
Section 26.1.1, Chapter 29
Section 26.1.1, Chapter 30
Section 26.1.1, Chapter 31
Table 26.1: Summary of router and routing features.
routing decision, and forwards the packet onto the next router in the path to the
destination.
The routing table can either be managed manually via static IP routing, or automatically by using dynamic routing protocols, or a combination of both. Static IP
routing is usually fine for small IP networks, or networks with no redundant paths.
To manage routing in larger networks, it is preferred to use dynamic IP routing.
With dynamic routing, the routers will exchange routing information and build
up their routing tables dynamically. Furthermore, dynamic routing utilises network redundancy; if a link goes down, routers will inform each other and packets
will automatically be routed along another path. Thus, dynamic routing protocols perform a similar service in routed networks as FRNT (chapter 16) and RSTP
(chapter 18) perform in switched networks. The time to react on a topology
change is referred to as the convergence time. WeOS supports two dynamic
routing protocols: Open Shortest Path First (OSPF) and Routing Information Protocol (RIP). OSPF is the recommended over RIP, due to its superior convergence
characteristics.
OSPF and RIP are examples of unicast Interior Gateway Protocols (IGPs), which
means they can be used to handle routing within a routing domain, such as an
corporate network. This is also referred to as intra-domain routing, as opposed
inter-domain routing, which is commonly handled using the Border Gateway Protocol (BGP)1 . OSPF and RIP are covered in chapters 27 and 28 respectively.
1 As of WeOS v4.21.1, dynamic routing is limited to intra-domain (unicast) routing with RIP and
OSPF. WeOS does not support dynamic inter-domain routing via BGP (Border Gateway Protocol), or
646
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
IP multicast routing enables efficient distribution of multicast data in a routed
network. A source, such as an IP camera, will send its data to a specific multicast IP address (also referred to as a multicast group), and receivers (the group
members) will listen in to this address by joining the group. WeOS supports static
multicast routing, which enables the network manager to manually set the multicast routing entries in the routers. Dynamic multicast routing protocols, such as
DVMRP or PIM-SM, are not yet supported. See chapter 29 for more details on IP
multicast routing.
While dynamic routing protocols such as RIP and OSPF enable routers to find
redundant paths in case a link or router goes down, they do not enable end
devices (hosts) to use a second router if their regular router goes down.
To
support redundancy between hosts and routers the Virtual Router Redundancy
Protocol (VRRP) is used. With VRRP, a backup router will take over if a router
fails, and communication from connected hosts can continue automatically. VRRP
support is covered in chapter 30.
When a router is used as a company gateway to a public network, such as the
Internet, there is an obvious need to protect the local company network against
network intrusion and other attacks. It is also common that the hosts and routers
within the company network use private IP addresses. To protect the company
network and to enable the use of private IP addresses, WeOS includes firewall
and network address translation (NAT) support. Chapter 31 describes the NAT
and firewall features in WeOS.
Another need which occurs when connecting company networks to the Internet is
to ensure communication privacy. WeOS supports IPsec VPN and SSL VPN (OpenVPN) to establish secure communication over public networks. With VPNs, a company can secure communication between a head office and different branch offices by installing a WeOS device as VPN gateway at each site. WeOS VPN support
is covered in part IV.
26.1.2
Using a WeOS device as a switch or as a router
WeOS devices are both able to route and to switch packets, i.e., they are routing
switches. Switching is performed between ports in the same VLAN, while routing
is performed between IP subnets or network interfaces (please see fig. 21.1 in
section 21.2 for information on the distinction between ports, VLANs and network
dynamic multicast routing.
© 2017 Westermo Teleindustri AB
647
Westermo OS Management Guide
Version 4.21.1-0
interfaces in WeOS). Routing can be disabled, and the WeOS device will then act
as a VLAN capable switch.
26.1.3
Learning routing information from different sources
A WeOS device will learn about routing information by manual configuration (connected interfaces or static routes), dynamic address assignment (e.g., DHCP), or
via dynamic routing protocols (OSPF and RIP). As described in chapters 27 and 28,
a router is able to redistribute external routing information into an OSPF or RIP
routing domain.
In some situations a router will learn the route to the same destination through
different mechanisms. In this case, the route to use will depend on the administrative distance (or simply ”admin distance”) associated with the involved routing
mechanisms. A route with a lower admin distance will be prioritised over a router
with higher admin distance.
Connected routes are always preferred (they have admin distance ’0’ (zero)). In
WeOS the admin distance of static routes, and routes learnt dynamically via RIP
and OSPF can be configured, but defaults to the values shown in the table below.
Routes learnt dynamically via DHCP or PPP will have admin distance according to
the distance assigned to the associated interface, see section 21.2.6.
Administrative
Distance
Connected
Static
OSPF
RIP
0
1
110
120
Configuring static routes with higher administrative distance than set for OSPF
or RIP is also referred to floating static routes, see section 26.1.4.2 for further
details.
26.1.4
Static routing
WeOS supports static IP routing. With static routing a WeOS devices can specify
the next hop router to use to reach a given IP subnet, or add additional (directly
attached) subnets to a local interface.
648
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
26.1.4.1
Using Static Route with Next-Hop or Interface as target
When defining a static route, the target is typically an IP address, e.g., ”route
192.168.5.0/24 192.168.1.1” where ”192.168.1.1” would be the IP address
of the next-hop router towards the destination.
In other situations you could define the target as a network interface of your unit,
e.g., ”route 192.168.5.0/24 ssl0” where all traffic towards ”192.168.5.0/24”
would be sent via your SSL VPN interface (chapter 36).
Note
Using an interface as target of a static route is almost only used on pointto-point interfaces, e.g., SSL or GRE interfaces. In rare cases it can be used
on LAN interfaces when you have multiple subnets on a VLAN, but in those
cases it is often simpler to use a secondary IP address on that LAN interface,
see section 21.2.5.
26.1.4.2
Floating Static Routes - Administrative Distance for Static Routes
Floating static routes are static routes with higher administrative distance (see
section 26.1.3) than routes learnt dynamically, e.g., via routing protocols such as
OSPF and RIP, or via dynamic configuration protocols such as DHCP or IPCP (PPP).
An example where a default route acquired via DHCP is given precedence over a
floating static (default) route is given in section 21.2.6. To complement this, an
example where routes learnt via OSPF is given precedence over a floating static
route is illustrated in fig. 26.1.
In this example, the user could have used OSPF over the low-speed backup link,
but has instead chosen to use a floating static route. Relevant parts of the configuration at routers 1, 2 and 3 are shown below.
Router 1 injects a default route into the OSPF area, a defines a floating static
route towards 192.168.35.0/24 via Router2.
© 2017 Westermo Teleindustri AB
649
Westermo OS Management Guide
Version 4.21.1-0
Internet
(GW at 192.168.32.1)
192.168.32.0/24
.2
Router1
.1
.1
Static route
(floating)
OSPF
Modem
(Low speed)
192.168.33.0/24
192.168.34.0/24
Modem
Backup Link
(Low speed)
.2
.2
Router2
.1
Router3
192.168.35.0/24
.2
Figure 26.1: Use of floating static route for on low-speed backup link.
Example
#Router1
ip
route 0.0.0.0/0 192.168.32.1
route 192.168.35.0/24 192.168.33.2 200
end
router
ospf
network 192.168.34.0/24 area 0.0.0.0
distribute-default always
end
end
Router 2 defines a floating static default route towards via Router1, and injects a
default route into the OSPF area given that its floating default route is active (no
”always” attribute; compare with Router1 configuration).
650
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Example
#Router2
ip
route 0.0.0.0/0 192.168.33.1 200
end
router
ospf
network 192.168.35.0/24 area 0.0.0.0
distribute-default
end
end
Router 3 has no static routes, i.e., it only uses OSPF.
Example
#Router3
router
ospf
network 192.168.34.0/24 area 0.0.0.0
network 192.168.35.0/24 area 0.0.0.0
end
end
26.1.4.3
Blackhole routes
WeOS has a blackhole interface referred to as ”null0”. This interface is hidden
in the sense that it cannot be configured (no IP address, management settings,
etc.). The blackhole interface is useful to avoid routing loops in networks with
incomplete subnetting.
An example is shown in fig. 26.2. R1 has set a static route for the ”192.168.0.0/22”
range towards R2. R2 only has routes to a part of this range, i.e., the directly connected subnets ”192.168.0.0/24”, ”192.168.1.0/24” and ”192.168.2.0/24”, while
”192.168.3.0/24” is currently unused. As R2 has defined R1 as its default route, a
packet sent towards e.g., ”192.168.3.11” would bounce back and forth between
R1 and R2, unless R2 defines a blackhole route.
Note
In this example, the static blackhole route for ”192.168.0.0/22” has a shorter
prefix than the directly connected routes. Therefore only traffic in range
”192.168.3.0/24” will be sent to ”null0” as long as the interfaces to the
directly connected subnets are up.
© 2017 Westermo Teleindustri AB
651
Westermo OS Management Guide
Version 4.21.1-0
Static routes at R1:
"192.168.0.0/22 via 172.16.0.2"
"0.0.0.0/0 via 1.2.3.4"
Static routes at R2:
"192.168.0.0/22 via null0"
"0.0.0.0/0 via 172.16.0.1"
R2
R1
Internet
.1 172.16.0.0/30
(Next Hop
1.2.3.4)
.2
Other parts of 192.168.0.0/22
are sent to "null0"
.1
.1 .1
192.168.0.0/24
192.168.1.0/24
192.168.2.0/24
Figure 26.2: Use of blackhole route at router R2 to avoid a routing loop for addresses within range 192.168.2.0-192.168.255.255.
Use of blackhole routes is also useful when setting up SSL VPNs or IPsec VPNs.
By use of blackhole routes, you can avoid that private traffic to the peer side is
routed (unencrypted) towards the Internet when the VPN tunnel is down. See
section 36.1.6 for an example of using blackhole routes with SSL VPNs.
26.1.5
Limitations When Using RSTP and Routing
As of WeOS v4.21.1 a single RSTP instance per WeOS unit is supported. This
works fine in a switched environment where all VLANs on a switch can be added
to inter-switch ports, see also chapters 15 (VLAN) and 18 (RSTP/STP).
However, when using RSTP in a routed environment it is often needed to run a
separate instance of RSTP per VLAN. Otherwise there is a risk that RSTP incorrectly detects a loop (at layer-2) and blocks some port, even though there is a
”routing barrier”, which already handles the loop. The result of RSTP blocking
ports may be loss of connectivity at layer-3.
RSTP is typically enabled on all ports by default. When using the WeOS device as
a router, it is therefore recommended either to
ˆ disable RSTP as a whole, or
ˆ disable RSTP on all ports but one VLAN, or a group of VLANs with a shared
layer-2 backbone (such as a ring).
Support for multiple RSTP/STP instances is planned but not yet implemented.
652
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
26.2
View Unicast Routing Table and Manage Static
Unicast Routes via Web Interface
Web configuration of static unicast routes is presented in section 26.2.1, and
examination of the current (unicast) routing table via Web is covered in section 26.2.2.
Web configuration of static multicast routes and examination of the multicast
routing table is instead handled in chapter 29.
26.2.1
Managing Static Unicasts Routing via Web Interface
Menu path: Configuration ⇒ Routing ⇒ Static Route
The main static routing configuration page lists the currently configured static
routes.
Destination
Netmask
Distance
Gateway
Interface
The subnet to route
The netmask defining the subnet
The administrative distance used when selecting between
multiple routes to the same destination (floating static
route).
The destination gateway
The destination interface
Edit
Click this icon to edit a route.
Delete
Click this icon to remove a route. You will be asked to acknowledge the removal before it is actually executed.
© 2017 Westermo Teleindustri AB
653
Westermo OS Management Guide
Version 4.21.1-0
Menu path: Configuration ⇒ Routing ⇒ Static Route ⇒
Edit
The edit page, see table above for descriptions.
654
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
26.2.2
Examine Routing Table via the Web Interface
Menu path: Status ⇒ Routing ⇒ Routes
On this page the current IP routes are listed.
One or more codes describe which source the route has, and if it is selected.
C
K
S
R
O
>
*
Connected - A network is known by a direct connection to
the switch.
Kernel route
Static - A statically configured route.
RIP - The route is known through the RIP protocol.
OSPF - The route is known through the OSPF protocol.
Selected route
FIB route
© 2017 Westermo Teleindustri AB
655
Westermo OS Management Guide
Version 4.21.1-0
26.3
Enabling Routing, Managing Static Routing, etc.,
via CLI
The table below shows WeOS CLI commands relevant for handling static routing.
The detailed description of these commands is found in other chapters as listed
in the table.
Dynamic routing (RIP and OSPF) and other router related protocols (VRRP) share
a common router configuration context which is also listed in the table.
Command
Configure general routing settings
ip
[no] default-gateway <IPADDR>
[no] route <NETWORK NETMASK|
NETWORK/LEN> <GATEWAY|IFNAME>
[DISTANCE]
[no] forwarding
router
[no] ospf
[no] rip
[no] vrrp <ID>
Show general routing status
show ip route
26.3.1
Default
Section
DEPRECATED
Section 21.7.1
Section 21.7.2
Section 21.7.3
DISTANCE 1
Enabled
Section
Section
Section
Section
Section
21.7.4
26.3.1
27.3
28.3
30.3
Section 21.7.26
Manage Router Protocols
Syntax router
Context Global Configuration context
Usage Enter the Router Protocol Configuration context. From here you can configure dynamic routing protocols such as OSPF (section 27.3) and RIP (section 28.3) and, as well as other router related protocols such as VRRP (section 30.3).
Use ”show router” to list general router protocol settings (also available
”show” command within the Router Protocol Configuration context.
656
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Default values N/A
Example
Example
example:/config/#> router
example:/config/router/#> show
OSPF/RIP not enabled.
VRRP Instances =============================================================
ID
Interface
Router-ID
Priority
Address
============================================================================
1
vlan1
1
100
192.168.2.1
example:/config/router/#>
© 2017 Westermo Teleindustri AB
657
Westermo OS Management Guide
Version 4.21.1-0
Chapter 27
Dynamic Routing with OSPF
This chapter describes WeOS support for the OSPF dynamic routing protocol.
27.1
Overview of OSPF features
Feature
General OSPF settings
Router-id
OSPF Networks
Area type (regular, stub, NSSA)
Web
CLI
General Description
X
X
X
X
X
X
Redistribution (static, connected, RIP)
Distribute default route
Inter-area summarisation
Inter-area filtering
Passive interface default
X
X
X
X
X
X
X
X
X
X
Section 27.1.1.1
Section 27.1.1.1
Sections 27.1.1.2, and
27.1.1.4-27.1.1.5
Section 27.1.1.3
Section 27.1.1.3
Section 27.1.1.6
Section 27.1.1.6
Section 27.1.1.7
Per interface OSPF settings
Link cost
Passive interface
Authentication (MD5, plain)
Hello/Dead intervals
Designated Router priority
X
X
X
X
X
X
X
X
X
X
Section
Section
Section
Section
Section
658
27.1.1
27.1.1.7
27.1.1.8
27.1.1.9
27.1.1.10
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Note
As of WeOS v4.21.1 there is no support for ”load balancing” in case there
are multiple paths with equal cost to reach a destination.
When an OSPF configuration change is done in WeOS, OSPF will be restarted
on that router. Until the OSPF routing protocol has converged, this may
cause a temporary loss of connectivity in parts of your network.
27.1.1
OSPF introduction
Net−A
Net−B
Router−A
Router−B
Router−E
Net−E
Router−C
Router−D
Net−C
Net−D
Figure 27.1: Simple network topology with interconnected routers and networks.
Dynamic routing protocols such as OSPF and RIP (chapter 28) simplifies router
configuration, and improves network robustness.
ˆ Simplified configuration: Manual configuration of static routes is not needed,
and thereby a time consuming and error-prone procedure is avoided. In the
network shown in fig. 27.1, each router would only have to be configured
with information about its own identity and the IP subnets it is attached to.
Routers will then exchange this information, and be able to establish the
appropriate routing table by themselves.
© 2017 Westermo Teleindustri AB
659
Westermo OS Management Guide
Version 4.21.1-0
ˆ Improved robustness: If the topology changes, perhaps because a link failed,
routers will automatically detect this and inform each other. The data traffic
will be forwarded other ways, given that a redundant path to the destination
exists.
OSPF is an example of a link-state routing protocol. In a link-state routing protocol, each router announces information about its own identity (router-id), its directly connected networks, and its neighbour routers. This information is flooded
throughout the OSPF domain, and each router will store the information in a local
OSPF database. Each router will gain complete knowledge about every router
and link in the whole topology, and is therefore able to compute the best path
(the least cost path) to reach every destination1 .
For example, Router-A in fig. 27.1 would send out OSPF messages informing other
routers about its router-id, its connected networks, i.e., Net-A and the links towards routers A, B, and C, the identity of (and link to) to its neighbour routers (A,
B and C).
A major advantage of link-state routing protocols, such as OSPF, over distance
vector routing protocols, such as RIP, is the fast convergence after a topology
change. If a link goes down, information about this can be flooded rapidly to all
routers within the routing domain, and each router can then update their routing
table accordingly.
27.1.1.1
OSPF Router-ID and OSPF Networks
We use the example below to explain some essential OSPF parameter settings
(the example is for Router-A in fig. 27.2).
Example
router
ospf
router-id 10.0.11.1
network 10.0.1.0/24 area 0.0.0.0
network 10.0.2.0/24 area 0.0.0.0
network 10.0.3.0/24 area 0.0.0.0
network 10.0.11.0/24 area 0.0.0.0
end
end
1 In OSPF, a cost is associated with every link. As of WeOS v4.21.1, the default cost per link is
”10”. The link cost can be configured per interface, see section 27.3.16 for details.
660
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
10.0.11.0/24
10.0.12.0/24
.1
.1
.1
Router−A .1
.2
.1
10.0.1.0/24
Router−B
10
10
.
.1
0.
10.0.2.0/24
.0.
6.0
.1
3.
0/
10.0.5.0/24
24
.2
.2
.2
.2
24
.1
Router−C
.2
.1
/24 .2
Router−D
10
.0
0/
.7.
Router−E
.1
10.0.15.0/24
10.0.4.0/24
.1
10.0.13.0/24
.1
10.0.14.0/24
Figure 27.2: Example OSPF network with IP addresses and subnets.
The ”router-id” line states the identity of this OSPF router, and must be unique
within this OSPF routing domain.
ˆ The router-id is 32-bit value, and can be specified either as a regular integer
value, or in dotted-decimal form, just like an IP address.
ˆ It is common practise to set the router-id to one of the IP addresses assigned
to the router.
ˆ If no router-id is configured, WeOS will pick one of the router’s configured IP
addresses, and use that as router-id.
As mentioned in section 27.1.1, the router should inform the other routers about
its attached links and networks. However, a router will announce its networks
and links first when they are declared to be within the OSPF routing domain –
this is done via the ”network” command. Furthermore, a ”network” declaration
implies that OSPF messages will be exchanged through the corresponding network interface. (In some network setups one likes to include a subnet within the
OSPF domain, without activating OSPF on the corresponding interface. This can
be achieved by configured that interface as passive, see section 27.1.1.7.)
In the example above, Router-A has been configured to include and announce all
© 2017 Westermo Teleindustri AB
661
Westermo OS Management Guide
Version 4.21.1-0
its subnets in the OSPF domain (10.0.1.0/24, 10.0.2.0/24, etc.). From the example
we can also see that the ”network” declaration contains an area parameter.
OSPF areas are further explained in section 27.1.1.2.
27.1.1.2
OSPF hierarchy and areas
Being a link state protocol, OSPF requires routers to keep a lot of routing information in their database:
ˆ Each OSPF router will typically keep a database with information of every
router and link in the whole OSPF domain.
ˆ OSPF routers will also redistribute and keep routing information learnt from
external sources (static routes, routes learnt via other routing protocols,
etc.).
To reduce the burden of keeping keeping state information about the whole OSPF
domain, the domain can be split into OSPF areas. (For information on how to
avoid the need to keep information on external routing information, see section 27.1.1.4.)
Area 0.0.0.0
(Backbone area)
R
R
R
ABR
ABR
ABR
Area 0.0.0.1
R
ABR
Area 0.0.0.3
R
Area 0.0.0.2
R
R
R
R
R
Figure 27.3: Sample OSPF hierarchy with a backbone area and three other areas.
The routers in fig. 27.3 have been divided into four areas. When splitting the
network into multiple areas, each router will only have full knowledge of the
topology within their respective area. Routers will also keep summary information
about destinations outside their own area, but routers will not have knowledge
about the actual topology inside other areas.
662
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Each IP subnet can only be part of one OSPF area, and when configuring OSPF
networks you should also define which area it belongs to. The area identifier is a
32 bit value, which can be stated as a decimal value, but is commonly written in
dotted decimal form. E.g., ”network 10.0.1.0/24 area 0.0.0.0” is equivalent
to writing ”network 10.0.1.0/24 area 0”.
A router which have networks in different areas is called an area border router
(ABR). An example is given below.
Example
router
ospf
router-id 192.168.5.11
network 192.168.5.0/24 area 0.0.0.0
network 192.168.11.0/24 area 0.0.0.1
end
In OSPF, areas are organised in a two-level hierarchy. At the top we have area
0, which is referred to as the backbone area. As the hierarchy is limited to two
levels, every ABR must be connected to the backbone area. Direct connections
between areas at lower level is prohibited; all inter-area traffic should go via the
backbone area2 .
To allow for a more flexible area hierarchy, OSPF provides a feature referred to as
virtual links, however, OSPF virtual links are not supported in WeOS v4.21.1.
27.1.1.3
Route redistribution and default route
Route information learnt from other routing protocols (RIP, BGP3 , etc.) can be
redistributed (i.e., imported) into the OSPF domain. The same goes for static
routes, and directly connected networks.
To let a router redistribute routing information into the OSPF domain, the
”redistribute” command is used, e.g., ”redistribute rip” to import routes
learnt via RIP. An OSPF router performing route distribution into the OSPF domain
is referred to as an administrative system border router (ASBR).
Routers can inject a default route (0.0.0.0/0) into the OSPF domain. This is done
using the ”distribute-default [always]” command. Without the ”always”
keyword, the router will only inject the default route if it itself has a default route.
2 The reason for introducing these topology limitations is to avoid the ”counting to infinity” seen
in distance vector protocols (see chapter 28) problem to occur for OSPF inter-area routing.)
3 As of WeOS v4.21.1 BGP is not supported.
© 2017 Westermo Teleindustri AB
663
Westermo OS Management Guide
Version 4.21.1-0
External routes can be added at two levels, type 1 and type 2 external routes:
ˆ Type 1: Type 1 external routes are typically used when importing routes,
that are locally managed, e.g., a static routes inside your domain, or from a
local RIP domain.
The ASBR located in area 0.0.0.2 in fig. 27.4 would preferably redistribute
the routes learnt via RIP as type 1 external routes.
ˆ Type 2: Type 2 external routes are typically used when importing routes
managed by another operator, e.g., routes learnt via BGP.
The ASBRs located in area 0.0.0.0 in fig. 27.4 would preferably redistribute
the routes learnt via BGP as type 2 external routes.
27.1.1.4
Stub areas and totally stubby areas
In some situations one wish to limit the routing information going into an area to
be limited even further, perhaps due to limited resources on the router. For this
situation, OSPF provides a special area type referred to as a stub area.
As with other OSPF routers, routers inside a stub area will have full routing information for networks and routers within their own area and summary routes
to destinations in other areas, but need not keep routing information learnt from
external sources (static routes, or routes learnt via other routing protocols such
as RIP, BGP, etc.). In a stub area, routing to networks outside the OSPF domain
is instead based on default routing towards the ABR(s); i.e., the ABR will filter
out all external routing information and instead inject a default route (pointing to
itself) area.
To create a stub area, all routers in the area (ABRs as well as internal routers)
must declare the area as stub. An example is given below.
Example
router
ospf
router-id 192.168.5.11
network 192.168.5.0/24 area 0.0.0.0
network 192.168.11.0/24 area 0.0.0.1
area 0.0.0.1
stub
end
end
664
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
To reduce the routing information going into a stub area even further, it is possible to prohibit summary routes from other areas to go into a stub area. This
is done by adding the no-summary parameter to the stub command (”stub
no-summary”); this is only needed on the ABR(s) of the stub area.
Such areas are referred to as totally stubby areas.
The cost of the default route being injected into the stub area is by default set to
”1”. The cost value can be configured via the ”default-cost” command within
the area context.
The backbone area cannot be configured as a stub area.
27.1.1.5
Not so stubby areas (NSSAs)
In a stub area, no router can redistribute routing information learnt from external sources (static routes, BGP, etc.). That is, a stub area cannot contain an
autonomous system border router (ASBR).
If you wish to have an ASBR in an area, but limit the amount of routing information
to keep track of as in a stub area, OSPF provides an area type known as not so
stubby area (NSSA).
Fig. 27.4 demonstrates a case where NSSAs can be a useful choice. Here we
assume that area 0.0.0.1 and area 0.0.0.2 are preferably defined as stub areas
to avoid that BGP routes (redistributed by the ASBRs in the backbone area) are
propagated into those areas. But area 0.0.0.2 includes a router connected to
a local RIP network. By defining area 0.0.0.2 as a NSSA, the RIP routes can be
redistributed into the OSPF network.
NSSA are created in the same way as a stub area (see section 27.1.1.4). All
routers in the area must declare the area as NSSA. An example is given below.
Example
router
ospf
router-id 192.168.5.12
network 192.168.5.0/24 area 0.0.0.0
network 192.168.16.0/24 area 0.0.0.2
area 0.0.0.2
nssa
end
end
© 2017 Westermo Teleindustri AB
665
Westermo OS Management Guide
Version 4.21.1-0
Internet
Internet
(BGP)
Area 0.0.0.0
(Backbone)
ASBR
ASBR
R
R
R
ABR
ABR
Area 0.0.0.1
Stub
Area 0.0.0.2
NSSA
R
R
R
R
R
R
ASBR
RIP Network
R
R
Figure 27.4: Topology where NSSA areas are useful.
As with stub areas, NSSAs are able to prohibit inter-area routing information to
be distributed inside the area (use ”nssa no-summary” on the ABRs of the area).
Such areas are called NSSA totally stub areas.
The backbone area cannot be configured as a NSSA area.
666
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
27.1.1.6
Additional Area Specific Settings
ABRs are able to filter and to aggregate routing information before distributing it into another area. This is managed using the ”range <NETWORK/LEN>
[not-advertise]” command.
ˆ Route filtering: With the ”not-advertise” keyword, any route matching the
given range will be filtered out when distributing routing information outside
a certain area.
ˆ Route summarisation: Without the ”not-advertise” keyword, all routes
matching the given range will be summarised (aggregated) as a single destination (of given network and prefix length) outside of a certain area.
Below is and example where an ABR will filter out routes in 192.168.16.0/20 when
distributing routes from area 0.0.0.2. Similarly, all routes inside area 0.0.0.2
matching 172.16.0.0/16 will be summarised to single route, when distributing
routes from area 0.0.0.2.
Example
router
ospf
router-id 192.168.5.12
network 192.168.5.0/24 area 0.0.0.0
network 192.168.16.0/24 area 0.0.0.2
network 192.168.19.0/24 area 0.0.0.2
area 0.0.0.2
range 192.168.16.0/20 not-advertise
range 172.16.0.0/16
end
end
27.1.1.7
Passive Interfaces
In some situations you may wish to include a router’s subnets as part of the
OSPF routing domain without running OSPF on the associated network interface.
To accomplish this the network should be defined in the router ospf context (as
usual), and the related interface should be declared as passive in the interface
ospf context. Below is an example where network 192.168.33.0/24 should be
included in the OSPF domain, but where the associated interface (vlan100) is
declared as passive.
© 2017 Westermo Teleindustri AB
667
Westermo OS Management Guide
Version 4.21.1-0
Example
[frame=single]
iface vlan100 inet static
...
... Skipping lines
...
address 192.168.33.1/24
ospf
passive
end
end
router
ospf
router-id 192.168.15.1
network 192.168.15.0/24 area 0.0.0.0
network 192.168.33.0/24 area 0.0.0.0
end
end
By default, OSPF will run on all interfaces which have an associated network
declared as an OSPF network. If OSPF should not run on such an interface, that
interface should be declared as passive, as described above. However, WeOS
is able to support use cases where the interfaces should be passive by default.
The parameters controlling the behaviour are the ”passive-interface” setting
in router ospf context, and the ”passive” setting in the interface ospf context.
ˆ passive-interface: Use the ”[no] passive-interface” setting in router
ospf context to control whether interfaces should be passive in OSPF by
default or not. Default setting: Active (”no passive-interface”)
ˆ passive: Use the ”[no] passive [auto]” setting in interface ospf context
to control whether a specific interface should be passive (”passive”), active (”no passive”), or to automatically follow (”passive auto”) the global
OSPF setting declared by the ”[no] passive-interface” setting in router
ospf context. Default: Auto (”passive auto”)
Below is an example, with the same result as above, where interfaces are passive
in OSPF by default.
668
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Example
iface vlan110 inet static
...
... Skipping lines
...
address 192.168.15.1/24
ospf
no passive
end
end
router
ospf
router-id 192.168.15.1
passive-interface
network 192.168.15.0/24 area 0.0.0.0
network 192.168.33.0/24 area 0.0.0.0
end
end
27.1.1.8
OSPF security
If an ”external” OSPF router happens to connect to your network (maliciously or
by mistake) the routing inside your domain can be affected severely. E.g., if that
router injects a default route into the OSPF domain, all traffic supposed to go to
your Internet gateway may instead be routed towards this ”foreign” router.
To avoid that this happens, it is good practise to enable authentication of all
OSPF messages inside your network. WeOS provides two forms of authentication
of OSPF messages:
ˆ Plain: Plain text authentication will protect against the situation when careless users attach an OSPF router to your network by mistake. However,
since the password is sent in plain text inside the OSPF messages, it does
not prohibit a deliberate attacker to inject routing information into your network. Plain text secrets are text strings of 4-8 characters.
ˆ MD5: With MD5 authentication each OSPF message will include a cryptographic checksum, i.e., message authentication code (MAC), based on a secret only known by the system administrator. MD5 secrets are text strings
of 4-16 characters.
Authentication of OSPF messages is configured per network interface, and is disabled by default.
Use of MD5 authentication is recommended. When using MD5 authentication,
© 2017 Westermo Teleindustri AB
669
Westermo OS Management Guide
Version 4.21.1-0
an associated key identifier must be specified. The purpose of the key identifier
is to enable use of multiple MD5 keys in parallel when performing key roll-over.
However, as of WeOS version v4.21.1 only a single OSPF secret per interface is
supported.
Warning
Configuring OSPF authentication remotely in an operational network can be
dangerous, since the communication towards that router can be broken if
the neighbour routers do not yet have the corresponding authentication configuration. In this case it is good practice to always have a redundant routing
path to the router you are configuring.
If the you end up in the situation where you can no longer reach a router
due to a change in OSPF authentication configuration, you may be able to
solve the situation by first logging into a ”neighbour” of the ”unreachable
router”, and from that router use SSH (see section 7.3.33) to login to the
”unreachable router”, and then update the configuration appropriately.
27.1.1.9
Finding OSPF Neighbours
OSPF routers will periodically transmit OSPF Hello messages, and routers can
thereby discover new neighbour routers, and also detect if a neighbour router is
down. There two parameter settings related to the OSPF hello messages. These
settings are configured per interface.
ˆ Hello-interval: The interval (in seconds) at which this router is transmitting
Hello messages. Default: 10 seconds
ˆ Dead-interval: The interval (in seconds) after which a neighbour router is
considered down if no Hello message from that router is received4 . Default:
40 seconds
Note
All routers attached to a link must have identical ”hello-interval” and ”deadinterval” settings. That is, an OSPF router will only accept incoming Hello
messages with identical hello and dead interval values as the router itself is
using on that interface.
4 If the interface towards that neighbour goes down (e.g., if (all) the Ethernet port(s) associated
with that interface goes down), the router will react immediately instead of waiting for the deadinterval to expire.
670
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
27.1.1.10
Designated OSPF router
In shared networks, such as Ethernets, there may be several routers attached to
the same LAN. Representing a LAN as a full mesh of links between the attached
routers may grow the OSPF database substantially if the number routers is large.
Instead, link state protocols, such as OSPF, treats a shared link as a logical star,
with a virtual node in the middle representing the shared network, see 27.5. The
router which takes the role of network is referred to as the designated router.
R2
R1
R2
R3
R1
(DR)
R4
R3
R1*
R5
R5
a)
R4
b)
Figure 27.5: Link state protocols such as OSPF logically represent a shared link
(a) as a star (b). One of the attached routers (here R1), will take the role as
designated router and represent the ”network” in the middle.
The designated router (DR), as well as a backup designated router (BDR), are
elected automatically. If no node has been elected as DR or BDR, the router with
the highest configured DR election priority becomes the DR, using the router-id
as tie-breaker when more than one router has highest priority.
OSPF implements a sticky DR election scheme. Once a router has become DR, it
will keep that role even when a router with higher DR priority comes up. However,
a DR will give up its role if it discovers another router, which also consider itself
to be DR, and if that router has higher priority (with router-id as tie). Such a
situation could occur if a segmented LAN becomes connected.
© 2017 Westermo Teleindustri AB
671
Westermo OS Management Guide
Version 4.21.1-0
27.2
OSPF Web
The Web interface provides configuration of OSPF.
Menu path: Configuration ⇒ Routing ⇒ OSPF
When entering the OSPF configuration page the basic settings are presented.
Router ID
OSPF Networks
Click on the
icon to set the OSPF router
identifier. The router ID is given in a dotted
decimal form <a.b.c.d> or as an integer
Enable OSPF on the router interface with the
specified IP subnet (NETWORK/LEN). Click on
the
to edit settings or the
icon to delete
an entry. Press the Add button to add an entry.
To view all settings, click on Show Advanced View (see next page).
672
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Router ID
Click on the
icon to set the OSPF router
identifier. The router ID is given in a dotted
decimal form <a.b.c.d> or as an integer
Continued on next page
© 2017 Westermo Teleindustri AB
673
Westermo OS Management Guide
Version 4.21.1-0
OSPF Networks
Interfaces Default Passive
Distribute Default Route
Redistribute
Neighbor(s)
Area Specific Settings
Protocol Distance
27.2.1
Continued from previous page
Enable OSPF on the router interface with the
specified IP subnet (NETWORK/LEN). Click on
the
to edit settings or the
icon to delete
an entry. Press the Add button to add an entry.
Define whether OSPF should be run on the interfaces defined (implicitly) via the OSPF network settings.
Enable/disabled injection of a default route into
the OSPF domain
Enable/disabled import of external routing information into the OSPF domain
Setup OSPF neighbor routers explicitly
Add specific settings to an area. Click on the
to edit settings or the
icon to delete an
entry. Press the Add button to add an entry.
The administrative distance used when selecting between multiple routes to the same destination.
OSPF Status Page
Menu path: Status ⇒ Routing ⇒ OSPF
674
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Show the status of OSPF.
© 2017 Westermo Teleindustri AB
675
Westermo OS Management Guide
Version 4.21.1-0
27.3
Managing OSPF via the CLI
The table below shows OSPF management features available via the CLI.
Command
Configure General OSPF Settings
router
[no] ospf
[no] router-id <ROUTERID>
[no] network <NETWORK/LEN>
[area <AREAID>]
[no] neighbor <ADDRESSLIST>
[no] passive-interface
[no] distribute-default [always]
[metric-type <1|2>]
[metric <0-16777214>]
[no] redistribute connected
[metric-type <1|2>]
[metric <0-16777214>]
[no] redistribute static [metric-type <1|2>]
[metric <0-16777214>]
[no] redistribute rip [metric-type <1|2>]
[metric <0-16777214>]
[no] distance <1-255>
[no] area <AREAID>
[no] stub [no-summary]
[no] nssa [no-summary]
[no] default-cost <0-16777215>
[no] range <NETWORK/LEN>
[<advertise|not-advertise>]
Configure Interface Specific OSPF Settings
interface <IFACE>
[no] ospf
[no] passive [auto]
[no] cost <1-65535>
676
Default
Section
Disabled
Auto
area 0
Sec.
Sec.
Sec.
Sec.
Disabled
Active
Disabled
Sec. 27.3.4
Sec. 27.3.5
Sec. 27.3.6
Disabled
Sec. 27.3.7
Disabled
Sec. 27.3.7
Disabled
Sec. 27.3.7
110
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
Disabled
Disabled
0
advertise
26.3.1
27.3.1
27.3.2
27.3.3
27.3.8
27.3.9
27.3.10
27.3.11
27.3.12
27.3.13
Sec. 27.3.14
Auto
Sec. 27.3.15
10
Sec. 27.3.16
Continued on next page
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Continued from previous page
Command
Default
Section
[no] hello-interval <1-65535>
10
Sec. 27.3.17
[no] dead-interval <1-65535>
40
Sec. 27.3.18
[no] auth <md5 [KEYID] | plain> <SECRET> Disabled Sec. 27.3.19
[no] priority <0-255>
1
Sec. 27.3.20
View OSPF Status
show ip ospf
show ip ospf route
show ip ospf neighbor [<IFACE | detail>]
show ip ospf database [asbr-summary|external|
network|router|summary>
show ip ospf database max-age
show ip ospf database self-originate
27.3.1
Sec.
Sec.
Sec.
Sec.
27.3.21
27.3.22
27.3.23
27.3.24
Sec. 27.3.24
Sec. 27.3.24
Activate OSPF and Manage General OSPF Settings
Syntax [no] ospf
Context Router Protocol Configuration context
Usage Enter the Router OSPF Configuration context, and activate OSPF with default settings if OSPF is not activated already. Instead of running ”ospf”
from the Router Protocol Configuration context, you can use ”router ospf”
directly from the Global Configuration
Use ”no ospf” to disable OSPF and delete all existing OSPF configuration.
Use ”show ospf” to show a summary of all general OSPF settings. Also
available as ”show” command within the Router OSPF Configuration context.
Default values Disabled (no ospf)
27.3.2
Configure OSPF Router-ID
Syntax [no] router-id <ROUTER-ID>
© 2017 Westermo Teleindustri AB
677
Westermo OS Management Guide
Version 4.21.1-0
Context Router OSPF Configuration context
Usage Set the OSPF router identifier, which must be unique within your OSPF
domain. The router ID is a 32-bit value, and is given in a dotted 1decimal
form <a.b.c.d> (where a-d are numbers in the range 0-255), or as an integer
(0..232 − 1). Commonly the router ID is set equal to one of the router’s IP
addresses.
In Auto mode, the router ID is automatically set to the IP address of one of
the router’s interface (the highest IP address), and stick to that value until
the OSPF process is restarted.
Use ”show router-id” to show the router-ID setting.
Default values Auto (no router-id)
27.3.3
Enable OSPF on an Interface
Syntax [no] network <NETWORK/LEN> [area <AREAID>]
Context Router OSPF Configuration context
Usage Enable OSPF on the router interface with the specified IP subnet (NETWORK/LEN), include that IP subnet in the OSPF routing domain, and determine the associated OSPF area.
The area ID is a 32-bit number, and is entered in dotted decimal form, or as
an integer (0..232 − 1). By default, the backbone area (0.0.0.0) is assumed.
Use ”no network <NETWORK/LEN> [area <AREAID>]” to delete a configured ”network” entry.
Use ”show network” to show the OSPF network settings.
Default values Disabled, i.e., no ”network” entries exist when first activating
OSPF (see section 27.3.2). The backbone area (0.0.0.0) is used as default
area.
27.3.4
Configure Static Neighbour Router
Syntax [no] neighbor <ADDRESSLIST>
Context Router OSPF Configuration context
678
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Usage Manually configure OSPF neighbours. This may be useful when intermediate switches do not propagate IP multicast, or when using OSPF in NBMA
(non-broadcast multiple access) networks.
Use ”neighbor <IPADDR>” to manually add one (or more) OSPF neighbour
router(s). Use ”no neighbor” to remove all manually configured neighbours, or ”no neighbor <IPADDDR>” to remove a specific neighbour.
Use ”show neighbor” to show manually configured OSPF neighbours.
27.3.5
Configure Interface Default Active/Passive Setting
Syntax [no] passive-interface
Context Router OSPF Configuration context
Usage Define whether OSPF should be run on the interfaces defined (implicitly)
via the OSPF ”network” command (see section 27.3.3).
If the setting is ”no passive-interface”, the interfaces associated with
the ”network” command will automatically run OSPF, unless OSPF is explicitly disabled on the interface (see the ”passive” command in section 27.3.15).
Similarly, if the setting is ”passive-interface”, the interfaces associated
with the ”network” command will not run OSPF, unless OSPF is explicitly enabled on the interface (see the ”no passive” command in section 27.3.15).
Use ”show passive-interface” to show the default behaviour of OSPF interfaces (passive or active).
Default values Active (”no passive-interface”)
27.3.6
Configure Distribution of Default Route into OSPF Domain
Syntax [no] distribute-default [always] [metric-type <1|2>]
[metric <0-16777214>]
Context Router OSPF Configuration context
Usage Inject a default route into the OSPF domain, i.e., announce that this router
can reach network 0.0.0.0/0.
© 2017 Westermo Teleindustri AB
679
Westermo OS Management Guide
Version 4.21.1-0
Use the ”always” keyword to make the router always advertise the default
route, regardless if it has one or not. Without the "always" keyword, it will
only advertise if it has one.
Use ”show distribute-default” to show the whether this router is configured to inject a default route into the OSPF domain.
Default values Disabled (”no distribute-default”)
27.3.7
Configure Redistribution of External Route Information
into OSPF Domain
Syntax [no] redistribute <connected|static|rip> [metric-type <1|2>]
[metric <0-16777214>]
Context Router OSPF Configuration context
Usage Import external routing information into the OSPF domain. Redistribution
of connected routes, static routes, and routes learnt via RIP is handled independently, e.g., use ”redistribute rip” to import routes learnt via RIP.
Use ”no redistribute” to remove all redistribution, and ”no redistribute
rip” to remove redistribution of routes learnt via RIP, etc.
Use ”show how redistribute [<connected|static|rip>]” to show the
OSPF redistribution settings. Use ”show redistribute” to show all redistribution settings, or ”show redistribute connected”, etc., to show redistribute settings for specific types of redistribution.
Default values Disabled (”no redistribute”)
27.3.8
Configure Admin Distance for OSPF
Syntax [no] distance <1-255>
Context Router RIP Configuration context
Usage Configure admin distance for all routes learnt via OSPF. If the same route
is learnt via different routing protocols (or as connected or static route), the
route associated with the lowest admin distance will be used. For OSPF the
admin distance defaults to 110. See also sections 21.2.6 and 26.1.3.
Use ”no distance” to reset the OSPF admin distance to its default value.
680
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Use ”show distance” to show the configured OSPF admin distance value.
Default values 110
27.3.9
Manage area specific settings
Syntax [no] area <AREAID>
Context Router OSPF Configuration context
Usage Enter the OSPF Area Configuration context of the specified AREAID to configure area specific settings, such as area type (regular, stub, nssa), interarea route summarisation, etc.
Use ”no area <AREAID>” to remove specific for a single area, and ”no
area” to remove specific settings for all areas.
Use ”show area [<AREAID>]>]” to show a summary of area specific settings. Use ”show area” to show settings for all areas, and ”show area
<AREAID>” to show settings for a specific area. (Also available as ”show”
command within the OSPF Area Configuration context.)
Default values Disabled (”no area”)
27.3.10
Configure an Area as Stub
Syntax [no] stub [no-summary]
Context OSPF Area Configuration context
Usage Configure an area as a stub area. To create a stub area, all routers in
the area (ABRs as well as internal routers) must declare the area as stub.
To configure the area as a totally stubby area, all ABRs in the area should add
the no-summary parameter to the stub command (”stub no-summary”).
Use ”no stub” to let a stub (or nssa) area become a regular area.
Use ”show stub” to show whether this area is configured as stub or not. If
this is a stub area, it will show whether the ”no-summary” keyword is set or
not, i.e., if it is a totally stubby area or just a stub area.
Default values Disabled (i.e., areas are ”regular” OSPF areas by default)
© 2017 Westermo Teleindustri AB
681
Westermo OS Management Guide
Version 4.21.1-0
27.3.11
Configure an Area as NSSA
Syntax [no] nssa [no-summary]
Context OSPF Area Configuration context
Usage Configure an area as a nssa area. To create a nssa area, all routers in
the area (ABRs as well as internal routers) must declare the area as nssa.
To configure the area as a NSSA totally stub area, all ABRs in the area should
add the no-summary parameter to the nssa command (”nssa no-summary”).
Use ”no nssa” to let a nssa (or stub) area become a regular area.
Use ”show nssa” to show whether this area is configured as NSSA or not. If
this is a NSSA area, it will show whether the ”no-summary” keyword is set
or not, i.e., if it is a NSSA totally stub area or just a NSSA area.
Default values Disabled (i.e., areas are ”regular” OSPF areas by default)
27.3.12
Configure default route cost in stub and NSSA areas
Syntax [no] default-cost
Context OSPF Area Configuration context
Usage Configure the cost of the default route injected into a stub area. This
setting only applies to the ABRs of a stub or NSSA area.
Use ”no default-cost” to use the default value for the default cost setting.
Use ”show default-cost” to show the setting of the default-cost, i.e., the
cost of the default route injected by ABRs into a stub or NSSA area.
Default values ”default-cost 0”
27.3.13
Configure inter-area route summarisation and filtering
Syntax [no] range <NETWORK/LEN> [<advertise|not-advertise]
Context OSPF Area Configuration context
Usage Configure inter-area route summarisation or route filtering.
682
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Use the ”range <NETWORK/LEN>” (”range <NETWORK/LEN> advertise” is
equivalent) to aggregate routes (within this area) matching the specified
<NETWORK/LEN> range, before distributing the routes outside this area.
That is, all routes within this range are summarised as a single route, when
advertised outside this area.
Use the ”range <NETWORK/LEN> not-advertise” to prohibit routes (within
this area) matching the specified <NETWORK/LEN> range, to be distributed
outside this area. That is, routes within this range are filtered.
Use ”no range <NETWORK/LEN>” to remove a specific summary/filter setting, or ”no range” to remove all summary/filter settings for this area.
Use ”show default-cost” to show configured route summarisation and
route filtering settings for this area.
Default values Disabled
27.3.14
Manage Interface Specific OSPF Settings
Syntax [no] ospf
Context Interface Configuration context
Usage Enter the Interface OSPF Configuration context, i.e., the context where
Interface specific OSPF settings are configured.
Use ”no ospf” to remove any specific OSPF settings for this interface.
Use ”show ospf” to show a summary of OSPF settings for this interface.
(Also available as ”show” command within the Interface OSPF Configuration
context.)
Default values Disabled (i.e., no interface specific OSPF settings)
27.3.15
Configure Interface OSPF Passive Settings
Syntax [no] passive [auto]
Context Interface OSPF Configuration context
Usage Control whether a specific interface should be passive (”passive”), active (”no passive”), or to automatically follow (”passive auto”) the global
© 2017 Westermo Teleindustri AB
683
Westermo OS Management Guide
Version 4.21.1-0
OSPF setting declared by the ”[no] passive-interface” setting in router
ospf context (see section 27.3.5).
Use ”show passive” to show the OSPF passive interface setting (passive,
active or ”auto”) for this interface.
Default values Auto (”passive auto”)
27.3.16
Configure Interface OSPF Cost Settings
Syntax [no] cost <1-65535>
Context Interface OSPF Configuration context
Usage Configure interface OSPF cost.
Use ”no cost” to return to the default setting.
Note
As of WeOS v4.21.1 only static configuration of the interface OSPF cost
setting is available. Support to let the cost automatically depend on
the interface data rate is planned, but not yet implemented.
Use ”show cost” to show the OSPF cost setting for this interface.
Default values 10 (this may be subject to change in later versions of WeOS.
27.3.17
Configure Interface OSPF Hello Interval Settings
Syntax [no] hello-interval <1-65535>
Context Interface OSPF Configuration context
Usage Configure OSPF hello interval (in seconds) for this interface.
Use ”no hello-interval” to return to the default setting.
Note
The hello interval setting must be the same on neighbour routers.
Use ”show hello-interval” to show the OSPF hello interval setting for this
interface.
684
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Default values 10 (seconds)
27.3.18
Configure Interface OSPF Dead Interval Settings
Syntax [no] dead-interval <1-65535>
Context Interface OSPF Configuration context
Usage Configure OSPF dead interval (in seconds) for this interface.
Use ”no dead-interval” to return to the default setting.
Note
The dead interval setting must be the same on neighbour routers.
Use ”show dead-interval” to sow the OSPF dead interval setting for this
interface.
Default values 40 (seconds)
27.3.19
Configure Authentication of OSPF Messages
Syntax [no] auth <md5 [KEYID] | plain> <SECRET>
Context Interface OSPF Configuration context
Usage Configure authentication of OSPF messages on this interface. Two authentication methods are available:
ˆ MD5: Use ”auth md5 <KEYID> <SECRET>” to use a MD5 cryptographic
authentication. MD5 secrets are text strings of 8-16 characters. A key
identifier (0-255) is associated with MD5 keys. (Both the secret and the
key identifier must be the same on neighbour routers.)
ˆ Plain: Use ”auth plain <SECRET>” to use a clear-text password as authentication. Plain text secrets are text strings of 4-8 characters. (The
secret must be the same on neighbour routers.)
Use ”no auth” to disable authentication of OSPF messages on this interface.
Use ”show auth” to show the OSPF authentication setting for this interface.
© 2017 Westermo Teleindustri AB
685
Westermo OS Management Guide
Version 4.21.1-0
Default values Disabled
27.3.20
Configure OSPF Designated Router Priority
Syntax [no] priority <0-255>
Context Interface OSPF Configuration context
Usage Configure the OSPF designated router priority, which affects the chance
to become designated router on a broadcast network. A higher value increases the chance to become designated router.
Use ”priority 0” to state that this router is not eligible as designated
router on this interface/”IP subnet”.
Use ”no priority” to return to the default setting.
Use ”show priority” to show the OSPF designated router election priority
setting for this interface.
Default values 1 (”priority 1”)
27.3.21
Show General OSPF Status
Syntax show ip ospf
Context Admin Exec context.
Usage Show general OSPF status information.
Default values Not applicable
27.3.22
Show OSPF Routes
Syntax show ip ospf route
Context Admin Exec context.
Usage Show the current least-cost routes learnt via OSPF. See also the command ”show ip route” (section 21.7.26), which displays the full forwarding/routing table.
Default values Not applicable
686
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
27.3.23
Show OSPF Neighbours
Syntax show ip ospf neighbor [<IFACE | detail>]
Context Admin Exec context.
Usage Show current list of OSPF neighbours. Use ”show ip ospf neighbor
IFACE” to list OSPF neighbours for a specific interface, or the keyword ”detail”
to receive a more detailed listing.
Default values By default, neighbours on all interfaces are listed.
27.3.24
Show OSPF Database
Syntax
show ip ospf database [asbr-summary|external|network|router|summary>],
show ip ospf database max-age,
show ip ospf database self-originate
Context Admin Exec context.
Usage Use ”show ip ospf database” to list the current OSPF database. Various keywords can be added to view specific parts of the database.
Default values By default, the full database is listed.
© 2017 Westermo Teleindustri AB
687
Westermo OS Management Guide
Version 4.21.1-0
Chapter 28
Dynamic Routing with RIP
This chapter describes WeOS support for the Routing Information Protocol (RIP.)
WeOS supports dynamic routing via RIP version 1 (RIPv1) and version 2 (RIPv2).
RIP is relatively simple to setup, but does not handle topology changes as rapidly
as the OSPF dynamic routing protocol (support for OSPF is described in chapter 27). Therefore, OSPF is generally preferred over RIP when it is possible to
select dynamic routing protocol.
28.1
Overview of RIP Features
Table 28.1 summarises RIP support in WeOS.
28.1.1
Introduction to RIP
RIP is an example of a distance vector routing protocol, and historically it has
been one of the most widely used intra-domain unicast routing protocol within
the Internet.
RIP is quite simple to configure; commonly you only have to enable RIP and
define which interfaces to run RIP on. The router will automatically discover its
neighbours and start to exchange routing information.
To enable RIP on all interfaces on R1 in fig. 28.1, configuration shown below would
suffice.
688
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Feature
General RIP settings
RIP version
RIP Timers
Passive Interface Default
RIP Networks/Interfaces
RIP Neighbour
Redistribution (static, connected, OSPF)
Distribute Default Route
RIP Admin Distance
Authentication (MD5, plain)
Passive interface
Split Horizon
Send RIP version
Receive RIP version
Web
CLI
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
General Description
Section 28.1.1
Section 28.1.4
Section 28.1.1
-”Section 28.1.2
-”Section 28.1.3
Section 28.1.4
Table 28.1: Summary of RIP features.
Example
router
rip
network 10.0.1.0/24
network 10.0.2.0/24
network 10.0.3.0/24
end
end
The command ”network 10.0.1.0/24” will enable RIP on all interfaces included
within the given range; in this example it states that RIP should be activated
on the ”upper interface” (i.e., the interface with address 10.0.1.3/24). It is also
possible to specify the interfaces explicitly; assuming the three interfaces of R1
are called vlan1, vlan2, and vlan3, the following configuration would give the
same result:
Example
router
rip
network vlan1
network vlan2
network vlan3
end
end
© 2017 Westermo Teleindustri AB
689
Westermo OS Management Guide
Version 4.21.1-0
.1
.2
10.0.1.0/24
.3
R1
.1
10.0.2.0/24
.2
.1
10.0.3.0/24
.2
Figure 28.1: A router (R1) connected to other routers via three interfaces.
Both RIPv1[10] and RIPv2[26] are supported, and RIPv2 is used by default when
RIP is enabled. The major difference between RIPv1 and RIPv2 is that RIPv2
supports flexible subnet masks (CIDR - classless inter-domain routing), while
RIPv1 assumes that IP subnet masks follow the (deprecated) classful addressing
scheme (class A, B and C). In addition, RIPv2 supports message authentication
(section 28.1.3), and can therefore offer protection in situations when ”foreign
RIP routers” are connected (by mistake or as a deliberate attack) to a network
and inject RIP routing messages. Thus, use of RIPv2 is preferred over RIPv1,
except for cases where legacy equipment require the use of RIPv1.
RIPv2 routers exchange routing information using IP multicast (IP address 224.0.0.9)1 .
In case a neighbour router is unable to handle IP multicast, the ”neighbor” command enables the exchange of RIP messages using regular IP unicast.
28.1.2
Redistribution and Injection of Default Route
It is possible to redistribute routing information learnt externally (OSPF, connected routes or static routes) inside the RIP routing domain, using the ”redistribute”
command.
1 While
690
RIPv2 use IP multicast, RIPv1 exchange routing information using broadcast.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
You can also let a RIP router inject a default route (0.0.0.0/0) into your RIP domain,
using the ”distribute-default”.
28.1.3
Authentication
To avoid that false routing information is injected into your network (deliberately
or by mistake) it is possible to authenticate RIPv2 messages. Two authentication
alternatives are available:
ˆ Plain: Plain text authentication will protect against the situation when careless users attach a RIP router to your network by mistake. However, since
the password is sent in plain text inside the RIP messages, it does not prohibit a deliberate attacker to inject routing information into your network.
Plain text secrets are text strings of 4-16 characters.
ˆ MD5: With MD5 authentication each RIP message will include a cryptographic checksum, i.e., message authentication code (MAC), based on a
secret only known by the system administrator. MD5 secrets are text strings
of 4-32 characters.
Authentication of RIP messages is configured per network interface, and is disabled by default.
Use of MD5 authentication is recommended. When using MD5 authentication,
an associated key identifier must be specified. The purpose of the key identifier
is to enable use of multiple MD5 keys in parallel when performing key roll-over.
However, as of WeOS version v4.21.1 only a single RIP secret per interface is
supported.
28.1.4
Passive interface
In some situations you may wish to include a router’s subnets as part of the RIP
routing domain without running RIP on the associated network interface. To accomplish this the network should be defined in the router rip context (as usual),
and the related interface should be declared as passive in the interface rip context. Below is an example where network 10.0.3.0/24 should be included in the
RIP domain, but where the associated interface (vlan3) is declared as passive.
© 2017 Westermo Teleindustri AB
691
Westermo OS Management Guide
Version 4.21.1-0
Example
iface vlan3 inet static
...
... Skipping lines
...
address 10.0.3.1/24
rip
passive
end
end
router
rip
network 10.0.1.0/24
network 10.0.2.0/24
network 10.0.3.0/24
end
end
By default, RIP will run on all interfaces which have an associated network declared as a RIP network. If RIP should not run on such an interface, that interface
should be declared as passive, as described above. However, WeOS is able to
support use cases where the interfaces should be passive by default. The parameters controlling the behaviour are the ”passive-interface” setting in router
rip context, and the ”passive” setting in the interface rip context.
ˆ passive-interface: Use the ”[no] passive-interface” setting in router rip
context to control whether interfaces should be passive in RIP by default or
not. Default setting: Active (”no passive-interface”)
ˆ passive: Use the ”[no] passive [auto]” setting in interface rip context to
control whether a specific interface should be passive (”passive”), active
(”no passive”), or to automatically follow (”passive auto”) the global RIP
setting declared by the ”[no] passive-interface” setting in router rip
context. Default: Auto (”passive auto”)
Below is an example, with the same result as above, where interfaces are passive
in RIP by default.
692
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Example
iface vlan1 inet static
...
... Skipping lines
...
address 10.0.1.3/24
rip
no passive
end
end
iface vlan2 inet static
...
... Skipping lines
...
address 10.0.2.1/24
rip
no passive
end
end
router
rip
passive-interface
network 10.0.1.0/24
network 10.0.2.0/24
network 10.0.3.0/24
end
end
© 2017 Westermo Teleindustri AB
693
Westermo OS Management Guide
Version 4.21.1-0
28.2
RIP Web
The Web interface provides configuration of RIP.
Menu path: Configuration ⇒ Routing ⇒ RIP
When entering the RIP configuration page the basic settings are presented.
Version
RIP Networks/Interfaces
Select what RIP version (1 or 2) to use by default
Enable RIP on the specified router Network/Interface
Click this icon to delete a RIP Network or RIP
Interface.
To view all settings, click on Show Advanced View (see next page).
694
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Version
RIP Networks/Interfaces
Interfaces Default Passive
Select what RIP version (1 or 2) to use by default
Enable RIP on the specified router Network/Interface
Define whether RIP should be run on the interfaces defined (implicitly) via the RIP
Continued on next page
© 2017 Westermo Teleindustri AB
695
Westermo OS Management Guide
Version 4.21.1-0
Distribute Default
Redistribute
Timers
Neighbor(s)
Protocol Distance
28.2.1
Continued from previous page
Enable/disabled injection of a default route into
the RIP domain
Enable/disabled import of external routing information into the RIP domain
Setup timers of the RIP protocol
Setup RIP neighbor routers explicitly
Click this icon to delete a RIP Network or RIP
Interface.
The administrative distance used when selecting between multiple routes to the same destination.
Rip Status Page
Menu path: Status ⇒ Routing ⇒ RIP
Show the status of RIP.
696
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
28.3
Managing RIP via the CLI
The table below shows RIP management features available via the CLI.
Command
Configure General RIP Settings
router
[no] rip
[no] version <1|2>
[no] timers [update <SEC>]
[invalid <SEC>]
[flush <SEC>]
[no] network <NETWORK | IFACE>
[no] neighbor <ADDRESSLIST>
[no] passive-interface
[no] distribute-default
[no] redistribute connected
[no] redistribute static
[no] redistribute ospf
[no] distance <1-255>
Configure Interface Specific RIP Settings
interface <IFACE>
[no] rip
[no] passive [auto]
[no] split-horizon [poisoned-reverse]
[no] send-version <1,2>
[no] receive-version <1,2>
[no] auth <md5 [keyid] | plain> <SECRET>
View RIP Status
show ip rip
© 2017 Westermo Teleindustri AB
Default
Section
Sec.
Sec.
Sec.
Sec.
26.3.1
28.3.1
28.3.2
28.3.3
Active
Disabled
Disabled
Disabled
Disabled
120
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
28.3.4
28.3.5
28.3.6
28.3.7
28.3.8
28.3.8
28.3.8
28.3.9
Auto
Enabled
Auto
Auto
Disabled
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
21.6.1
28.3.10
28.3.11
28.3.12
28.3.13
28.3.14
28.3.15
Disabled
version 2
update 30
invalid 180
flush 240
Sec. 28.3.16
697
Westermo OS Management Guide
Version 4.21.1-0
28.3.1
Activate RIP and Manage General RIP Settings
Syntax [no] rip
Context Router Protocol Configuration context
Usage Enter the Router RIP Configuration context, and activate RIP with default
settings if RIP is not activated already. Instead of running ”rip” from the
Router context, you can use ”router rip” directly from the Global Configuration
Use ”no rip” to disable RIP and delete all existing RIP configuration.
Use ”show rip” to show a summary of all general RIP settings. Also available as ”show” command within the Router RIP Configurationcontext.
Default values Disabled (no rip)
28.3.2
Configure Default RIP Version
Syntax [no] version <1|2>
Context Router RIP Configuration context
Usage Select what RIP version (1 or 2) to use by default, both with respect to
sending and receiving of RIP messages. The setting can be overridden per
interface using the ”receive-version” (section 28.3.14) and ”send-version”
(section 28.3.14) respectively.
Use ”no version” to return to the default setting.
Use ”show version” to show the default RIP version setting.
Default values RIPv2 (version 2)
28.3.3
Configure RIP Protocol Timers
Syntax [no] timers [update <SEC>] [invalid <SEC>] [flush <SEC>]
Context Router RIP Configuration context
Usage Several timers of the RIP protocol can be changed using the timers command. All timers take a value between <5-2147483647> seconds.
698
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
ˆ The update timer controls the interval between sending unsolicited Response Messages to all neighboring routers.
ˆ The invalid timer controls the time before a route is expired and removed from the kernel routing table. It is kept for ƒ sh − nd seconds in the internal RIP routing table to notify neighbors that a route
has been dropped.
ˆ The flush timer should be longer than the invalid timer. It controls the
time when a route is finally cleared from the routing table.
Important
All routers should have the same timings setup.
Use ”show timers” to show the configured RIP protocol timers.
Default values Use ”no timers” to return to the default timers:
update 30 sec
invalid 180 sec
flush 240 sec
Example
timers update 5 invalid 15 flush 30
This sends out updates every five seconds, invalidates a route if a router is not
heard from in 15 seconds and flushes the route after an additional 15 seconds.
28.3.4
Enable RIP on an Interface
Syntax [no] network <NETWORK/LEN | IFACE>
Context Router RIP Configuration context
Usage Enable RIP on the specified router interface. The interface can be specified either explicitly (”network <IFACE>”) or implicitly giving the IP subnet
associated with the interface (”network <NETWORK/LEN>”).
Use ”no network <IFACE>” and ”no network <NETWORK/LEN>” to remove
an existing ”network” entry.
© 2017 Westermo Teleindustri AB
699
Westermo OS Management Guide
Version 4.21.1-0
Use ”show network” to show the RIP network settings, i.e., which interfaces/subnets that are included in the RIP routing domain.
Default values Disabled, i.e., when first activating RIP (section 28.3.1), RIP will
not be enabled on any interface.
28.3.5
Configure Unicast Neighbor
Syntax [no] neighbor <ADDRESSLIST>
Context Router RIP Configuration context
Usage Configure one or more RIP neighbor routers explicitly. This is useful in
case the neighbor router is unable to handle IP multicast. An ”ADDRESSLIST”
is a comma-separated list of IPv4 address, e.g,
”neighbor 192.168.1.1,192.168.3.2”. Calling the ”neighbor” command
twice (with arguments ”192.168.1.1” and ”192.168.3.2” respectively) would
be equivalent.
Use ”no neighbor” to remove all configured neighbours, and ”no neighbor
<ADDRESSLIST>” to remove a specific neighbour settings.
Use ”show neighbor” to show the configured RIP Unicast Neighbours.
Default values Disabled (No neighbours defined)
28.3.6
Configure Interface Default Active/Passive Setting
Syntax [no] passive-interface
Context Router RIP Configuration context
Usage Define whether RIP should be run on the interfaces defined (implicitly)
via the RIP ”network” command (see section 28.3.4).
If the setting is ”no passive-interface”, the interfaces associated with
the ”network” command will automatically run RIP, unless RIP is explicitly
disabled on the interface (see the ”passive” command in section 28.3.11).
Similarly, if the setting is ”passive-interface”, the interfaces associated
with the ”network” command will not run RIP, unless RIP is explicitly enabled on the interface (see the ”no passive” command in section 28.3.11).
700
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Use ”show passive-interface” to show the default behaviour of RIP interfaces (passive or active).
Default values Active (”no passive-interface”)
28.3.7
Configure Distribution of Default Route into RIP Domain
Syntax [no] distribute-default
Context Router RIP Configuration context
Usage Inject a default route into the RIP domain, i.e., announce that this router
can reach network 0.0.0.0/0.
Use ”[no distribute-default” to stop this router from injecting a default
route into the RIP domain.
Use ”show distribute-default” to show the RIP redistribution settings.
Use ”show redistribute” to show all redistribution settings, or ”show redistribute
connected”, etc., to show redistribute settings for specific types of redistribution.
Default values Disabled (”no distribute-default”)
28.3.8
Configure Redistribution of External Route Information
into RIP Domain
Syntax [no] redistribute <connected|static|ospf>
Context Router RIP Configuration context
Usage Import external routing information into the RIP domain. Redistribution
of connected routes, static routes, and routes learnt via OSPF is handled
independently, e.g., use ”redistribute ospf” to import routes learnt via
OSPF.
Use ”no redistribute” to remove all redistribution, and ”no redistribute
ospf” to remove redistribution of routes learnt via OSPF, etc.
Use ”show redistribute [<connected|static|rip>]” to show the RIP
redistribution settings. Use ”show redistribute” to show all redistribution
settings, or ”show redistribute connected”, etc., to show redistribute
settings for specific types of redistribution.
© 2017 Westermo Teleindustri AB
701
Westermo OS Management Guide
Version 4.21.1-0
Default values Disabled (”no redistribute”)
28.3.9
Configure Admin Distance for RIP
Syntax [no] distance <1-255>
Context Router RIP Configuration context
Usage Configure admin distance for all routes learnt via RIP. If the same route
is learnt via different routing protocols (or as connected or static route), the
route associated with the lowest admin distance will be used. For RIP the
admin distance defaults to 120. See also sections 21.2.6 and 26.1.3.
Use ”no distance” to reset the RIP admin distance to its default value.
Use ”show distance” to show the configured RIP admin distance value.
Default values 120
28.3.10
Manage Interface Specific RIP Settings
Syntax [no] rip
Context Interface Configuration context
Usage Enter the Interface RIP configuration context, i.e., the context where Interface specific RIP settings are configured.
Use ”no rip” to remove any specific RIP settings for this interface.
Use ”show rip” to show a summary of RIP settings for this interface.
Default values Disabled (i.e., no interface specific RIP settings)
28.3.11
Configure Interface RIP Passive Settings
Syntax [no] passive [auto]
Context Interface RIP Configuration context
Usage Control whether a specific interface should be passive (”passive”), active (”no passive”), or to automatically follow (”passive auto”) the global
702
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
RIP setting declared by the ”[no] passive-interface” setting in router rip
context (see section 28.3.6).
Use ”show passive” to show the RIP passive interface setting (passive, active or ”auto”) for this interface.
Default values Auto (”passive auto”)
28.3.12
Configure Split Horizon Setting
Syntax [no] split-horizon [poisoned-reverse]
Context Interface RIP Configuration context
Usage Enable or disable split horizon on this interface, with optional poison reverse. Split horizon is a RIP mechanism to mitigate the counting to infinity
issue appearing in distance vector protocols such as RIP. Poisoned reverse
is a variant where the router actively advertises routes as unreachable over
the interface which they were learned. The effect of such an announcement
is to immediately remove most looping routes before they can propagate
through the network.
Use ”show split-horizon” to show whether split horizon is enabled on this
interface or not. If the optional poisoned reverse setting is enabled, that is
also stated.
Default values Enabled (”split-horizon”), with poison reverse disabled.
28.3.13
Configure RIP Version for Sending on this Interface
Syntax [no] send-version <1,2>
Context Interface RIP Configuration context
Usage Control whether this interface should use the global RIP version setting (section 28.3.2) when sending RIP messages on this interface (”no
send-version”), or to override the global setting by sending RIPv1 (”send-version
1”), RIPv2 (”send-version 2”), or both RIPv1 and RIPv2 (”send-version
1,2”).
Use ”no send-version” to remove override settings and return to auto
setting. (Override can also be removed for individual versions, e.g., ”no
send-version 1” to remove version 1 as override setting.)
© 2017 Westermo Teleindustri AB
703
Westermo OS Management Guide
Version 4.21.1-0
Use ”show send-version” to show RIP version override settings when accepting incoming RIP messages on this interface.
Default values Auto (”no send-version”)
28.3.14
Configure RIP Version for Receiving on this Interface
Syntax [no] receive-version <1,2>
Context Interface RIP Configuration context
Usage Control whether this interface should use the global RIP version setting (section 28.3.2) when accepting incoming RIP messages on this interface (”no receive-version”), or to override the global setting by accepting RIPv1 (”receive-version 1”), RIPv2 (”receive-version 2”), or both
RIPv1 and RIPv2 (”receive-version 1,2”).
Use ”no receive-version” to remove override settings and return to auto
setting. (Override can also be removed for individual versions, e.g., ”no
receive-version 1” to remove version 1 as override setting.)
Use ”show receive-version” to show RIP version override settings when
accepting incoming RIP messages on this interface.
Default values Auto (”no receive-version”)
28.3.15
Configure Authentication of RIP Messages
Syntax [no] auth <md5 [KEYID] | plain> <SECRET>
Context Interface RIP Configuration context
Usage Configure authentication of RIP messages on this interface. Two authentication methods are available:
ˆ MD5: Use ”auth md5 <KEYID> <SECRET>” to use a MD5 cryptographic
authentication. MD5 secrets are text strings of 4-32 characters. A key
identifier (0-255) is associated with MD5 keys. (Both the secret and the
key identifier must be the same on neighbour routers.)
ˆ Plain: Use ”auth plain <SECRET>” to use a clear-text password as authentication. Plain text secrets are text strings of 4-16 characters. (The
secret must be the same on neighbour routers.)
704
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Use ”no auth” to disable authentication of RIP messages on this interface.
Use ”show auth” to show the RIP authentication setting for this interface.
Default values Disabled
28.3.16
Show RIP Status Information
Syntax show ip rip (or simply ”show rip”)
Context Admin Exec context.
Usage Show RIP status information, e.g., active interfaces, discovered RIP neighbours, etc.
Default values Not applicable
© 2017 Westermo Teleindustri AB
705
Westermo OS Management Guide
Version 4.21.1-0
Chapter 29
IP Multicast Routing
This chapter describes the mechanisms involved in IP multicast routing and how
to setup and debug static multicast routing in WeOS.
29.1
Summary of WeOS Multicast Routing Features
Feature
Enable IP Forwarding
Enable IP Multicast Forwarding
Configure Static Multicast Routes
Multicast Routing Statistics
Related Settings
Layer-2 multicast forwarding
IGMP Snooping
Static Multicast Router Ports
Static MAC FDB entries
Block local ping responses
VRRP control of IP Multicast
706
Web
X
X
X
X
CLI
X
X
X
X
X
X
X
X
X
X
X
X
X
General Description
Section 29.1.1
-”-”-”-
Section 29.1.3
-”-”Section 29.1.4
Section 30.1.6
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
29.1.1
Overview of IP multicast
Multicast is an efficient data distribution mechanism for purposes of reaching
more than one receiver. IP multicast applications, such as a camera, need only
send one packet to reach a group of receivers. The network infrastructure,
switches and routers, send a copy of the packet to each subscriber of the group.
A multicast group is an IP address. In IPv4 the entire 224.0.0.0/4 block is reserved,
i.e., 224.0.0.0 – 239.255.255.255. However, not all address are available to the
end-user and some use-cases may not provide the most optimal distribution in
switched (layer-2) networks.
The 224.0.0.0/24 subnet (224.0.0.*) is reserved for control protocols, e.g., IGMP,
RIPv2 and OSPF.
Like regular IP addresses IP multicast groups must be translated to Ethernet (LAN)
MAC addresses. However, the range of reserved MAC multicast addresses is too
small, see RFC1112[6] for details.
The lack of reserved multicast MAC addresses may be a problem in switched
networks where the switch fabric often only supports IGMP Snooping (Sec. 20.1),
i.e., filtering, per MAC address. E.g., subscribers of group 224.1.2.3 will also
receive all traffic sent to group 225.1.2.3.
This is due to the mapping to MAC addresses, in our case
ˆ 224.1.2.3 maps to 01:00:5e:01:02:03
ˆ 225.1.2.3 maps to 01:00:5e:01:02:03
ˆ etc.
On a per LAN basis (layer-2) IP multicast is managed by IGMP (routers) and IGMP
Snooping (switches). Managing multicast on this level is important due to its
inherent broadcast nature. Knowledge of this can be very important when debugging multicast (re)distribution and routing.
Routing of IP multicast can be done either dynamically (e.g., DVMRP, PIM) or
statically. WeOS currently only supports the latter.
29.1.2
Static multicast routing
Contrary to static unicast, multicast has a separate routing table and is handled
a little bit differently. To be able to route multicast you need the following:
© 2017 Westermo Teleindustri AB
707
Westermo OS Management Guide
Version 4.21.1-0
ˆ Enable IP forwarding
ˆ Enable IP multicast forwarding
ˆ Setup a multicast route
ˆ Multicast data with a TTL > 1
The two enable flags simply control routing and multicast routing, respectively.
However, if IP forwarding is disabled toggling the multicast forwarding flag will
have no effect.
A static multicast route is made up of a group, an inbound interface, an optional
sender address and one or more outbound interfaces. There can be at most 128
multicast routes with at most eight (8) outbound interfaces per route.
The source, or sender address, is optional in WeOS but the underlying Linux kernel still needs a source address to be able to route the traffic. The multicast
routing daemon in WeOS manages this by adding rules to the kernel on-demand
based on the “source-less” rules specified. For each new multicast stream, from a
given group and inbound interface, the routing daemon checks to see if a matching mroute rule exists and then adds that source specific rule to the kernel. This
may cause some initial delays in activation of such rules.
29.1.3
IP multicast and IGMP Snooping
In LAN networks IGMP Snooping is often employed in switches to limit the distribution of IP multicast. Without subscribers to a certain multicast group, distribution
of a camera’s multicast stream is halted at the first switch. When IGMP Snooping
is disabled, the camera’s multicast stream is instead broadcast to all ports on the
switch, or all ports in the VLAN. For details, see Sec. 20.1 and Sec. 15.1.5.
In currently available network equipment, as well as modern operating systems,
IGMP is a well established protocol that works well. There may however still
exist older networking equipment, e.g., Programmable Logic Controllers (PLCs),
that does not know how to join a multicast group using IGMP. For such devices to
receive multicast it is possible in WeOS to either disable IGMP Snooping per VLAN,
add a specific FDB MAC entry for the multicast group to open up additional ports
in the switch, or use the multicast router port feature to forward all multicast on
a given port.
708
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
29.1.4
Blocking Local Ping Responses
To ensure that the multicast stream actually is received for routing by the CPU,
the WeOS router sends an IGMP join for the multicast group to be routed on the
given inbound interface. This has the odd side-effect that the router now also
responds to local pings to that group. To disable this, see Sec. 21.7.17.
© 2017 Westermo Teleindustri AB
709
Westermo OS Management Guide
Version 4.21.1-0
29.2
Managing Multicast Routing via Web Interface
Menu path: Configuration ⇒ Routing ⇒ Common
The WeOS web interface has full support for managing, configuring and debugging, static IP multicast routing.
To be able to route multicast both the Unicast and Multicast forwarding tick boxes
must be checked. The Unicast tick box is actually the big switch that controls all
IP routing.
Figure 29.1: Enable IP multicast forwarding.
29.2.1
Adding a Static Multicast Route
Menu path: Configuration ⇒ Routing ⇒ Static Multicast
By default no static multicast routes are setup. Click on New to create a new
static multicast route.
Figure 29.2: No multicast routes enabled by default.
710
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Enter the IPv4 multicast group address, the inbound interface and the source of
the sender.
Figure 29.3: Declare multicast group, inbound interface and source of sender.
Add outbound interfaces to your multicast route by selecting them in the drop
down and clicking Add for each one.
Figure 29.4: Select an outbound interface and press Add for each one.
© 2017 Westermo Teleindustri AB
711
Westermo OS Management Guide
Version 4.21.1-0
29.2.2
Adding a Sourceless Static Multicast Route
Menu path: Configuration ⇒ Routing ⇒ Static Multicast
WeOS supports ”source-less” static multicast routes as well, simply leave the
Source Address field empty.
Figure 29.5: Source-less: declare only multicast group, inbound and outbound
interfaces.
712
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
29.2.3
Overview of Configured Multicast Routes
Menu path: Configuration ⇒ Routing ⇒ Static Multicast
Figure 29.6: Overview of configured static multicast routes.
29.2.4
Deleting a Static Multicast Route
Menu path: Configuration ⇒ Routing ⇒ Static Multicast
In the overview, click the trashcan icon for the static multicast routing rule to
delete.
Figure 29.7: Confirm deleting a static multicast route by clicking Yes.
© 2017 Westermo Teleindustri AB
713
Westermo OS Management Guide
Version 4.21.1-0
29.2.5
Show Kernel Multicast Routing Table
Menu path: Status ⇒ Routing ⇒ Multicast Routes
The actual kernel multicast routing table is very useful to inspect for debugging,
e.g., seeing the amount of packets routed or any on-demand added ”source-less”
multicast routes.
Figure 29.8: Kernel multicast routing table, active multicast routes.
714
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
29.3
Managing Multicast Routing via CLI
The following table shows CLI commands relevant for managing, debugging and
querying static multicast routes in WeOS.
Command
Configure IP multicast routing
ip
[no] multicast-forwarding
[no] mroute group <MCADDR> in <IFNAME>
[src <IPADDR>] out <IFNAME-LIST>
Default
Section
Disabled
Section 29.3.1
Section 29.3.2
Show IP multicast routing status
show ip mroute
Section 29.3.3
There are some additional CLI settings which may be of interest when configuring
IP multicast on your unit. The table below lists the most relevant settings.
Command
Default
Related settings (IGMP, MAC FDB, VRRP, etc.)
fdb
[no] mac <MACADDR> port <PORTLIST>
Section
Section 15.4.3
vlan <VID>
[no] igmp
Enabled
Section 15.4.13
Disabled
Enabled
Section 20.3.4
Section 21.7.4
Enabled
Section 21.7.17
ip
[no] mcast-router-ports <PORTLIST>
[no] forwarding
icmp
[no] broadcast-ping
firewall
[no] filter [ARGS . . . ]
[no] nat [ARGS . . . ]
Section 31.3.3
Section 31.3.5
Continued on next page
© 2017 Westermo Teleindustri AB
715
Westermo OS Management Guide
Version 4.21.1-0
Command
iface <IFNAME>
vrrp <INSTANCE>
[no] mroute-ctrl
Continued from previous page
Default Section
Disabled
Related status commands (MAC FDB, IGMP, etc.)
show fdb
show ip igmp
show firewall
29.3.1
Section 30.3.12
Section 15.4.19
Section 20.3.6
Section 31.3.13
Enable/disable IP multicast forwarding
Syntax [no] multicast-forwarding
Context IP Configuration context
Usage Enable/disable IP multicast forwarding (multicast routing). Use command
”multicast-forwarding” to enable IP multicast forwarding, given that IP
forwarding (routing) is enabled (”forwarding”, see section 21.7.4).
”no multicast-forwarding” disables IP multicast forwarding.
Use ”show multicast-forwarding” to show whether IP multicast forwarding is enabled or disabled.
Default values Disabled (”no multicast-forwarding”)
29.3.2
Configure static multicast routes
Syntax [no] mroute group <MCADDR> in <IFNAME>
[src <IPADDR>] out <IFNAME-LIST>
group <MCADDR> IPv4 multicast group to route
in <IFNAME> Inbound interface for multicast stream
src <IPADDR> Optional IPv4 sender address of multicast stream
716
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
out <IFNAME-LIST> Comma separated list of destination/outbound interfaces for multicast stream. MAX:8
Context IP Configuration context
Usage Add/remove a static multicast route.
If the src field is omitted from an mroute rule, any multicast stream matching
the given group and inbound interface will be added on-demand to the kernel multicast routing table. Use the Admin Exec command show ip mroute
to inspect.
Use the ”no”-form of the command to remove rules. The src and out arguments are not needed, e.g., ”no mroute group 225.1.2.3 in vlan1”.
Without any arguments ”no route” will remove all configured static multicast routes.
Use ”show mroute” to list configured static IP multicast routes.
29.3.3
Show IP multicast status and statistics
Syntax show ip mroute
Context Admin Exec context
Usage Show IP Multicast Forwarding table and statistics.
This command is useful to inspect the actual routes setup in the kernel multicast routing table. In particular this command is useful when having setup
”source-less” mroute rules.
Default values Not applicable.
Example Assume you have configured the following mroute rules:
Example
example:/config/ip/#> mroute group 225.1.2.3 src 192.168.2.42 in vlan1 out vlan2,vlan3
example:/config/ip/#> mroute group 225.3.2.1 in vlan1 out vlan2,vlan3
Then the resulting kernel multicast routing table may end up looking like
this:
© 2017 Westermo Teleindustri AB
717
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/#> show ip mroute
Group
Source
Inbound
Packets Bytes
Invalid Outbound
===================================================================================
225.1.2.3
192.168.2.42
vlan1
0
0
0
vlan2, vlan3
225.3.2.1
192.168.2.20
vlan1
0
0
0
vlan2, vlan3
225.3.2.1
192.168.2.21
vlan1
0
0
0
vlan2, vlan3
===================================================================================
The latter two entries have been added on-demand, this happens as soon as
initial multicast data frames from unknown sources are received on interface
vlan1 destined for group 225.3.2.1.
The columns Packets, Bytes and Invalid denote the total number of packets, bytes and number of invalid packets per rule. Please note that when
reconfiguring static multicast rules, or when related interfaces go up/down
the statistics are reset. So do not rely on them for accurate measurements,
they only exist to aid in debugging.
718
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
Chapter 30
Virtual Router Redundancy
(VRRP)
This chapter describes WeOS support for the Virtual Router Redundancy Protocol
version 2 (VRRPv2)[23] and version 3 (VRRPv3)[30].
VRRP is a standard protocol to enable redundancy between a host and its router,
in case the router goes down. VRRP can also be used for load balancing purposes.
VRRP provides router redundancy for regular (unicast) IP traffic by letting multiple
routers share a virtual IP and MAC address. If the (master) router goes down, a
backup router will automatically take over.
WeOS provides an optional feature, where the VRRP state (master or backup)
is used to enable/disable IP multicast routing of incoming IP multicast packets.
With this option enabled, the backup router will prevent the routing of (static) IP
multicast routes in addition to IP unicast routing. See chapter 29 for information
on support for static IP multicast routing in WeOS.
© 2017 Westermo Teleindustri AB
719
Westermo OS Management Guide
Version 4.21.1-0
30.1
Introduction to WeOS VRRP support
The table below summarises VRRP support in WeOS.
Feature
VRRP Instances
Virtual Router IDs (VRIDs)
Virtual Router IP Address
Virtual Router Priority
Static Priority
Dynamic Priority
Preemption control
Version Specific Settings
VRRP versions (v2/v3)
Advertisement Interval
Regular (v2)
Fast (v3)
Message
authentication (v2)
Advanced Features
Synchronisation Groups
Multicast Routing Control
Load balancing
30.1.1
Web
X
X
X
X
X
X
X
CLI
X
X
X
X
X
X
X
General Description
Sections 30.1.1-30.1.2
Sections 30.1.1-30.1.2
Sections 30.1.1-30.1.2
Sections 30.1.1-30.1.2
Sections 30.1.1-30.1.2
Sections 30.1.1-30.1.2
Sections 30.1.1-30.1.2
X
X
X
X
X
X
X
X
Sections
Sections
Sections
Sections
X
X
Section 30.1.4
X
X
X
X
X
X
Section 30.1.5
Section 30.1.6
Section 30.1.7
30.1.2-30.1.3
30.1.2-30.1.3
30.1.2-30.1.3
30.1.2-30.1.3
VRRP Overview
The primary objective of VRRP is to enable redundancy between a host and
its neighbour router, i.e., you can deploy additional routers on an IP subnet as
backup routers, and have one of the backup routers to automatically take over if
the primary router fails. Fig. 30.1 can be used to illustrate the need for VRRP in
such a scenario.
ˆ A host will typically have an IP setting where the default gateway points
to a specific router. An example is given in fig. 30.1a, where the host (H)
will send all traffic towards the Internet via Router 1 (R1) with IP address
720
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
192.168.1.1. If R1 fails, the host will lose Internet connectivity even though
a redundant path (R2) happens to exists.
ˆ VRRP enables routers to share a virtual IP (VIP) address. The router with the
highest priority acts as master for the VIP address, while the other routers
are backups in case the master fails. Fig. 30.1b illustrates the use of VRRP.
R1 and R2 are both responsible for the VIP address (192.168.1.3), with R1
as master since it has higher priority (150>100). If R1 goes down, R2 will
become master of the VIP address and communication can automatically
resume. Note that the default gateway of the host is configured to the VIP
address.
Internet
(or Corporate Intranet)
R1
R2
.1
.2
192.168.1.0/24
Internet
(or Corporate Intranet)
Virtual IP:
192.168.1.3
Priority 150
(Master)
R2
.1
.2
192.168.1.0/24
.78
H
R1
Default GW:
192.168.1.1
(i.e., "R1")
a)
H
Virtual IP:
192.168.1.3
Priority 100
(Backup)
.78
Default GW:
192.168.1.3
(i.e., "VIP")
b)
Figure 30.1: Illustrating the need for VRRP to support redundancy: a) Host (H)
loses connectivity when Router 1 (R1) fails. b) Host (H) can continue to communicate even though Router 1 (R1) fails, since VRRP enables Router 2 (R2) to take
over.
Note
VRRP enables a host to have redundant routers. For redundancy ”router to
router”, dynamic routing protocols such as OSPF (chapter 27) or RIP (chapter 28) can be used.
30.1.2
Common VRRP parameters
Some common VRRP parameters are listed below:
1. VRRP instance: WeOS allows you to configure up to 16 VRRP instances per
unit. Each instance will operate on a (VLAN) interface (e.g., vlan1) and be
© 2017 Westermo Teleindustri AB
721
Westermo OS Management Guide
Version 4.21.1-0
assigned a virtual router identifier (VRID), see item 2 below.
Note
The ”VRRP instance number” is a parameter only used by WeOS for
internal book keeping, e.g., when establishing VRRP synchronisation
groups (section 30.1.5). The VRRP instance number is not exchanged
in any VRRP message.
2. Virtual Router Identifier (VRID): Each instance is assigned a virtual router
instance identifier (VRID) in range 0-255. All routers on a LAN, acting as
virtual routers for a specific virtual IP address, must be configured with the
same VRID. That is, R1 and R2 in fig. 30.1b should have the same VRID, e.g.,
”33”.
Note
As of WeOS v4.21.1, a specific VRID (such as ”33”) can only be used
once per WeOS unit. Using the same VRID in a second VRRP instance
is not possible on a WeOS unit, not even on another LAN.
3. Virtual IP address (VIP): WeOS allows you to configure one VIP address per
VRRP instance. When designing your network there are some restrictions to
consider when selecting the VIP address.
ˆ Select VIP in correct IP subnet: The VIP address should be in the same
IP subnet as the regular IP address assigned to the interface (e.g., the
VIP address in fig. 30.1b is 192.168.1.3, which is in the same subnet as
R1’s and R2’s IP addresses on that subnet).
ˆ Select VIP not ”owned” by any router: Although it is possible to use an
address assigned to (i.e., owned by) a router as the VIP address, it is
recommended that a separate IP address is used.
Consider the example in fig. 30.1b): According to the recommendation,
the chosen VIP address (”192.168.1.3”) is separate from the addresses
assigned to R1 (”192.168.1.1”) and R2 (”192.168.1.2”).
Although discouraged, it would have been possible to chose ”192.168.1.1”
as VIP address. Being the owner of the address, R1 must in that case
be configured with priority 255, with dynamic priority disabled. More
information on VRRP priority is found in item 5 below.
4. Advertisement interval: In VRRP, the master will announce its presence by
sending VRRP Advertisements on a certain interval. For VRRPv2 the inter-
722
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
val can be configured in range 1-255 seconds. VRRPv3 allows sub-second
intervals (in steps of 100 ms) in range 0.1-40 seconds. All VRRP routers
associated with the same VRID must use the same VRRP version (see section 30.1.3), and must have the same advertisement interval setting.
A low VRRP advertisement interval gives faster fail-over (the time to detect
that a master is down is roughly 3 times the advertisement interval).
Default advertisement interval: 1 (second)
5. VRRP Priority: The VRRP priority parameter is used to define which router
should become master of the VIP address when multiple routers are available. (If two routers with the same priority transitions to master state, the
router with the highest IP address will win the election.)
The priority can be configured in range 1-255, where the value ”255” should
be used if (and only if) the router is also the owner of the VIP address (see
the Note in item 3 above). Default priority: 100
WeOS supports dynamic VRRP priority. E.g., if the master router loses its
Internet connection it should lower its priority dynamically (or even decline
to be master), this to allow for a backup router to take over immediately.
For example, if R1 in fig. 30.1b would lose its upstream connection, it could
lower its priority to 30, whereby R2 would could take over if preemption is
enabled.
In WeOS, dynamic VRRP priority is configured by mapping the status of an
event trigger, typically a ping trigger (see section 24.1) to a priority adjustment value.
If a router is the owner of the VIP, it should be configured with priority ”255”,
with dynamic priority disabled.
6. VRRP Preemption: The VRRP master election is not controlled by the priority
setting alone; there is also a preemption parameter, which enables you to
select to have a deterministic master election procedure (highest priority
always becomes master), or a sticky behaviour where the elected master
router would keep its role even when another router with higher priority
later appears on the network. With preemption disabled, the second router
would refrain from taking over as long as the current master continuous to
send advertisements.
The exception to this is if the new router connected to the subnet is the VIP
address owner (priority 255); the VIP owner will always preempt an existing
master.
© 2017 Westermo Teleindustri AB
723
Westermo OS Management Guide
Version 4.21.1-0
When preemption is enabled, an optional preemption delay parameter can
be configured (default 0 seconds), which determines how long the router
should wait until preemption is activated. Default: Disabled
Note
When the instance belongs to a synchronized group, the instance with
the shortest preemption delay will be used.
Note
Preemption only occurs when starting or restarting a higher priority
backup router, e.g. if a link down event occurs preemption will not be
used.
A sample VRRP configuration for R1 in fig. 30.1b is shown below:
Example
router vrrp 1
iface vlan2
address 192.168.1.3
vrid 33
priority 150
end
30.1.3
Selecting VRRP version (VRRPv2 or VRRPv3)
WeOS supports VRRP version 2 and version 3.The additions to version 3 is shorter
advertisement interval (faster failover) and IPv6 support (not supported in WeOS).
Authentication has been removed completely in version 3 since it was considered
to not provide any real security. It is mandatory that the master and the backup
routers uses the same VRRP version. Default: VRRPv2
30.1.4
Authentication (VRRPv2 only)
Warning
Use of VRRP authentication is discouraged[11], as it may cause more harm
than help.
For VRRPv2, WeOS supports a simple form of VRRP message authentication, enabling the inclusion of a plain-text password in the VRRP advertisements[23].
724
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
To avoid that multiple master routers appear on an IP subnet, a WeOS VRRP router
will refrain from becoming master if it hears another router with mismatching
VRRP authentication information.
30.1.5
VRRP Synchronisation Groups
VRRP synchronisation is a function to keep the VRRP role (master vs backup) the
same for different VRRP instances on the same unit, see fig. 30.2.
192.168.55.0/24
.1
.2
VRID: 33
Virtual IP:
192.168.55.3
Priority 100
(Backup)
VRID: 33
Virtual IP:
192.168.55.3
Priority 150
(Master)
R2
R1
VRID: 1
Virtual IP:
192.168.1.3
Priority 150
(Master)
VRID: 1
Virtual IP:
192.168.1.3
Priority 100
(Backup)
.1
.2
192.168.1.0/24
H
.78
Default GW:
192.168.1.3
(i.e., "VIP")
Figure 30.2: Illustrating a topology using synchronised groups. Both instances
on R1 will always remain in master state as long no fault is detected (e.g. link
down). On fault R1 will become backup on both instances and R2 will become
master for both instances.
A synchronisation group consists of two VRRP instances. These two instances
should be active on different VLAN network interfaces, e.g. VRID 1 on interface
vlan1 can be synchronized with VRID 33 on interface vlan2. The VRRP instances
on a unit will only take the master role if it considers itself to have the highest
VRRP priority for both instances. If one of the VRRP instances in the synchronisation group would transistion to backup state (e.g. link down), the other instance
will also change state to backup, i.e. the instances in the synchronisation group
will always have the same state.
© 2017 Westermo Teleindustri AB
725
Westermo OS Management Guide
Version 4.21.1-0
30.1.6
VRRP Control of static IP Multicast Routing
When using static multicast routing and VRRP a problem that can occur is that
the multicast packets will get duplicated. This can be avoided by using the VRRP
multicast routing control. When using this feature, only the master router will
forward incoming multicast traffic from the configured VRRP interface while the
backup router will prevent the packets from being forwarded.
Note
The setting is applied per interface. It is not recommended to configure more
than one instance per interface as this will lead to unpredictable results.
30.1.7
Load sharing
It is possible to use VRRP for load sharing between routers, and still provide
redundancy, by having the routers acting as backup for each other. Fig. 30.3
shows a load sharing example. Here the VIP addresses reside within the same IP
subnet. However, since WeOS supports multi-netting, the VIP addresses could be
on different IP subnets.
Internet
(or Corporate Intranet)
VRID 33:
Virtual IP:
192.168.1.3
Priority 150
(Master)
R1
VRID 44:
Virtual IP:
192.168.1.4
Priority 75
(Backup)
VRID 33:
Virtual IP:
192.168.1.3
Priority 100
(Backup)
R2
.1
VRID 44:
Virtual IP:
192.168.1.4
Priority 200
(Master)
.2
192.168.1.0/24
.22
H
.25
H
.14
H
.90
H
.67
H
.77
H
Default GW: 192.168.1.3
Default GW: 192.168.1.4
(R1 Master, R2 Backup)
(R2 Master, R1 Backup)
Figure 30.3: Example setup where R1 and R2 share the load from IP subnet
192.168.1.0/24, and using VRRP to backup each other.
726
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
30.2
Managing VRRP via the web interface
Menu path: Configuration ⇒ Routing ⇒ VRRP
The main VRRP configuration page lists the currently configured VRRP instances
on all interfaces.
Grouping
Interface
VRID
To work with groups for synchronised fail-over, select two
instances or a group for grouping/ungrouping. A group is
displayed with a [ linking the grouped instances, and common background colour.
The interface on which to listen for VRRP information and
act as gateway. Only VLAN interfaces may be selected.
Virtual Router ID. A unique ID common to those routers that
will provide redundancy..
Edit
Click this icon to edit a VRRP instance.
Delete
Click this icon to remove a VRRP instance. You will be asked
to acknowledge the removal before it is actually executed.
lick this button to create a new VRRP instance.
Continued on next page
Button New
© 2017 Westermo Teleindustri AB
727
Westermo OS Management Guide
Version 4.21.1-0
Button Group
Button
Ungroup
728
Continued from previous page
For synchronised fail-over - first select two ungrouped VRRP
instances and then click this button to group the instances.
For synchronised fail-over - first select one group of VRRP instances and then click this button to ungroup the instances.
They will be left as two individual instances that has to be
removed separately.
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
30.2.1
Create a new VRRP instance using the web interface
Menu path: Configuration ⇒ Routing ⇒ VRRP ⇒ New
Interface
The interface on which to listen for VRRP information and
act as gateway. Only VLAN interfaces may be selected.
Continued on next page
© 2017 Westermo Teleindustri AB
729
Westermo OS Management Guide
Version 4.21.1-0
Continued from previous page
A unique ID common to those routers that will provide redundancy.
A virtual address that the routers will use when providing
the gateway support. The VIP address should be in the same
IP subnet as the regular IP address assigned to the interface
Version
VRRP version to use (v2 or v3).
Advertisement The interval in seconds how often a VRRP advertisement
Interval
message will be sent out. Allowed values: v2: 1-255 seconds v3: 0.1-40 seconds, in 100 msec intervals between
0.1 and 1.0 (default: 1).
Advertisement The interval in seconds how often a VRRP advertisement
Interval
message will be sent out. Allowed values: 1-255 seconds
(default: 1)
Priority
A number used for election of current gateway. A higher
number means a higher chance to become elected. If two
routers has the same priority in an election, the router with
the highest IP address will win. The value 255 should be
used if (and only if) the router is also the owner of the virtual
IP address. Allowed values: 1-255 seconds (default: 100)
Preemption
Enable/disable preemption and, if enabled, set a preemption delay. Preemption allows an elected router to remain
as master for a time period If the new router is the virtual
IP address owner (priority 255), it will always become the
master. Default: Disabled
Multicast
Let VRRP control multicast routing. If checked, multicast
Routing
routing will be disabled automatically for this instance when
Control
entering BACKUP state. Only one VRRP instance per interface may be configured for controlling multicast routing.
The checkbox is disabled if another instance is in control.
Virtual
Router ID
Virtual
Address
For more information on the different settings, see section 30.1.1.
730
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
30.2.1.1
Dynamic Priority
Track Trigger
Priority
Adjustment
If not disabled, the alarm trigger selected will, if triggered,
add the priority adjustment value to the router priority.
A positive or negative number to add to the priority when
the alarm has triggered. Allowed values: -255 to 255.
For more information on the different settings, see section 30.1.1.
30.2.2
Edit VRRP settings using the web interface
Menu path: Configuration ⇒ Routing ⇒ VRRP ⇒
For description of fields, see section 30.2.1.
30.2.3
VRRP Status Page
Menu path: Status ⇒ Routing ⇒ VRRP
Show the status of all configured VRRP instances.
© 2017 Westermo Teleindustri AB
731
Westermo OS Management Guide
Version 4.21.1-0
30.3
Managing VRRP via the CLI
The VRRP CLI syntax has been changed from an approach where VRRP was configured per (VLAN) interface to an approach where VRRP instances are configured as a common router service. Entering the configuration via both methods has been supported since WeOS v4.9.x. When storing the configuration,
WeOS v4.13.x uses the new (router service) style.
Command
Configure VRRP Settings
router
[no] vrrp <INSTANCEID>
[no] iface <IFNAME>
[no] vrid <VRID>
[no] version <2|3>
[no] address <ADDRESS>
[no] interval <INTERVAL> [ms]
[no] priority <1..255>
[no] preempt [delay <0..1000>]
[no] auth <plain> <SECRET>
[no] track trigger <ID> adjust <DELTA>
[no] sync <INSTANCEID>
[no] mroute-ctrl
Default
2
1
100
Disabled
Disabled
Disabled
Diabled
Disabled
View VRRP Status
show vrrp
30.3.1
Section
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
Sec.
26.3.1
30.3.1
30.3.2
30.3.3
30.3.4
30.3.5
30.3.6
30.3.7
30.3.8
30.3.9
30.3.10
30.3.11
30.3.12
Sec. 30.3.13
Create and Manage a VRRP Instance
Syntax [no] vrrp <INSTANCEID>
Context Router Protocol Configuration context
Usage Create, manage, or delete a VRRP instance. Use ”vrrp <INSTANCEID>”
to enter the VRRP Instance Configuration context of the specified VRRP in-
732
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
stance (INSTANCEID can be in the range 1-16). If the instance does not
already exist, it will be created.
Use ”no vrrp <INSTANCEID>” to remove a specific VRRP instance, or ”no
vrrp” to remove all configured VRRP instances.
At most 16 VRRP instances can be created per unit.
Use ”show vrrp [INSTANCE]” to show summary of VRRP settings. Use
”show vrrp” to list settings for all configured VRRP instances, and ”show
vrrp INSTANCE” to list settings for a specific VRRP instance.
Default values Disabled
30.3.2
Configure VRRP interface
Syntax [no] iface <IFNAME>
Context VRRP Instance Configuration context
Usage Configure VRRP interface.
An interface is a mandatory setting (”no iface” is an invalid setting).
Use ”show iface” to show the configured interface for this VRRP instance.
Default values None
30.3.3
Configure Virtual Router ID
Syntax [no] vrid <VRID>
Context VRRP Instance Configuration context
Usage Set the virtual router identifier (VRID) used for the VRRP instance. The
VRID must be unique per switch.
A virtual router identifier is a mandatory setting (”no vrid” is an invalid
setting).
Use ”show vrid” to show the configured virtual router ID (VRID) for this
VRRP instance.
Default values None
© 2017 Westermo Teleindustri AB
733
Westermo OS Management Guide
Version 4.21.1-0
30.3.4
Configure VRRP Version
Syntax [no] version <2|3>
Context VRRP Instance Configuration context
Usage Configure VRRP version to be used.
Use ”no version” to return to the default version setting.
Use ”show version” to show the configured version (2 or 3) for this VRRP
instance.
Default values 2
30.3.5
Configure Virtual Address
Syntax [no] address <ADDRESS>
Context VRRP Instance Configuration context
Usage Set the virtual IP address (VIP address) used for the VRRP instance.
The VIP address should be within the same IP subnet as the regular IP address assigned to the interface (see section 21.6.3).
Only one VIP address can be configured per VRRP instance.
Use ”show address” to show the configured virtual IP (VIP) address for this
VRRP instance.
Default values Disabled
30.3.6
Configure VRRP Advertisement Interval
Syntax [no] interval <1..MAX> | <100..MAX*1000> msec
Context VRRP Instance Configuration context
Usage Configure VRRP advertisement interval in seconds or milliseconds. MAX
(in syntax description) is depending on version and is 255 for version 2 and
40 for version 3.
734
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
For version 2 the allowed interval is <1..255> seconds and for version 3
the allowed interval is <0.1..40> seconds. To configure an interval that is a
fraction of a second one must set the interval in milliseconds.
A small value enables faster fail-over.
Use ”no interval” to return to the default interval setting.
Use ”show interval” to show the configured advertisement interval for this
VRRP instance.
Default values 1 (second)
Example In this example, the interval is set to 500 milliseconds. The setting is
only valid for VRRP version 3.
Example
example:/config/#> router
example:/config/router/#> vrrp 33
example:/config/router/vrrp-33/#> interval 500 msec
example:/config/router/vrrp-33/#> leave
example:/#> copy running start
30.3.7
Configure VRRP Priority
Syntax [no] priority <1..255>
Context VRRP Instance Configuration context
Usage Configure VRRP priority. A high value increases the chance to become
master of the VIP address (see also the ”preempt” command in section 30.3.8).
Priority ”255” should be used if (and only if) this router is the owner of the
IP address used as VIP address, i.e., if the VIP address is assigned as an IP
address to this router’s interface (see section 21.6.3).
Use ”no priority” to return to the default priority setting.
Use ”show priority” to show the configured VRRP priority for this VRRP
instance.
Default values 100
© 2017 Westermo Teleindustri AB
735
Westermo OS Management Guide
Version 4.21.1-0
30.3.8
Enable or Disable VRRP Master Preemption
Syntax [no] preempt [delay <0..1000>]
Context VRRP Instance Configuration context
Usage Enable or disable VRRP master preemption. If enabled, this router will
preempt an existing master if the current master has lower priority. (Note:
The owner of a VIP address will always take over as master irrespective of
the ”preempt” setting.)
When preemption is enabled, the router will wait a time interval depending on the configured advertisement interval and a configurable preemption
delay (seconds) before taking over as master.
Note
Preemption only occurs when starting or restarting a higher priority
backup router, e.g. if a link down event occurs preemption will not be
used.
Note
Note: When the instance belongs to a synchronized group, the instance
with the shortest preemption delay will be used.
Use ”no preempt” to prohibit this router to preempt an existing VRRP master.
Use ”show preempt” to show the configured VRRP master preemption setting for this VRRP instance.
Default values Disabled (”no preempt”) When enabled, the delay defaults to
0 seconds.
30.3.9
Configure VRRP Message Authentication
Syntax [no] auth <plain> <SECRET>
Context VRRP Instance Configuration context
Usage Configure VRRP message authentication. Simple clear-text authentication is supported for VRRP version 2.
736
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
The associated secret can be 4-7 characters. Valid characters are ASCII
characters 33-126, except ’#’ (ASCII 35).
Authentication is not available in VRRP version 3. Authentication will automatically be disabled if version 3 is configured. Use ”no auth” to disable
VRRP message authentication.
Use ”show auth” to show the configured VRRP message authentication setting for this VRRP instance.
Default values Disabled
30.3.10
Configure VRRP Dynamic Priority
Syntax [no] track trigger <ID> adjust <DELTA>
Context VRRP Instance Configuration context
Usage Configure dynamic VRRP priority. The VRRP priority will be adjusted by
the given delta value (-255 to 255) when the associated trigger reports
”alarm” status. E.g., ”track trigger 2 adjust -100” will decrease the
VRRP priority by 100 when there is an alarm condition on trigger 2.
When a router is the owner of the VIP, i.e. configured with priority ”255”,
the dynamic priority has no effect.
Use ”no track” to remove (all) track entries defined for this VRRP instance.
(As of WeOS v4.21.1, at most one ”track” entry can be configured.)
Use ”show track” to show the configured VRRP track entries, i.e., the dynamic VRRP priority setting.
Default values Disabled
Example In this example, this virtual router’s priority is lowered from 150 to
50, if the router cannot reach the host 192.168.3.11 through the (upstream)
interface vlan2.
© 2017 Westermo Teleindustri AB
737
Westermo OS Management Guide
Version 4.21.1-0
Example
example:/config/#> alarm
example:/config/alarm/#> trigger ping
example:/config/alarm/trigger-2/#> peer 192.168.3.11 outbound vlan2
example:/config/alarm/trigger-2/#> end
example:/config/alarm/#> end
example:/config/#> router
example:/config/router/#> vrrp 33
example:/config/router/vrrp-33/#> address 192.168.2.1
example:/config/router/vrrp-33/#> priority 150
example:/config/router/vrrp-33/#> track trigger 2 adjust -100
example:/config/router/vrrp-33/#> leave
example:/#> copy running start
30.3.11
Configure VRRP Synchronisation
Syntax [no] sync <VRRP ID>
Context VRRP Instance Configuration context
Usage Configure synchronization between two VRRP instances. This will specify a state monitoring between two VRRP instances. It guarantees that two
VRRP instances remain in the same state. The synchronized instances monitor each other. Changing this parameter will change the same parameter
on the corresponding instance.
Use ”no sync” to remove synchronization for this instance, this will remove
synchronization for the corresponding instance as well.
Use ”show sync” to show the configured VRRP instance ID this instance is
synchronized with.
Default values Disabled
Example In this example, virtual router instance 33 is synchronized with instance 35.
Example
example:/config/#> router
example:/config/router/#> vrrp 33
example:/config/router/vrrp-33/#> sync 35
example:/config/router/vrrp-33/#> leave
example:/#> copy running start
738
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
30.3.12
Configure VRRP Multicast Routing Control
Syntax [no] mroute-ctrl
Context VRRP Instance Configuration context
Usage Configure whether multicast traffic should be routed on a interface in
BACKUP state. If enabled, muticast traffic will not be routed when VRRP is in
BACKUP state.
Use ”no mroute-ctrl” to remove multicast routing control for this instance.
Use ”show mroute-ctrl” to show the configured VRRP multicast routing
control setting for this instance.
Default values Disabled
30.3.13
Show VRRP Status
Syntax show vrrp
Context Admin Exec context
Usage Show the status of all configured VRRP instances.
Default values Not applicable
© 2017 Westermo Teleindustri AB
739
Westermo OS Management Guide
Version 4.21.1-0
Chapter 31
Firewall Management
When connecting your network to the Internet (or any non-trusted network) a
router with firewall functionality should be used. The firewall will protect against
undesired access to your local servers, or other kinds of network intrusion from
attackers on the Internet.
The WeOS firewall supports the following main features:
ˆ Packet filtering: Packet filters enables you to control what traffic is allowed
to pass through your router/firewall and what packets it should drop. Packet
filter rules can also be specified to control access to services on your router.
ˆ Packet modification: Packet modification makes it possible to modify packets that are routed through the router/firewall.
ˆ Network Address Translation (NAT): The WeOS NAT functionality includes
both network address port translation (NAPT) and 1-TO-1 NAT.
ˆ Port forwarding: Port forwarding is often used together with NAPT, and will
then enable you to access servers in your private network from outside (e.g.,
from the Internet).
The WeOS firewall utilises connection tracking; a rule allowing traffic to pass
through the firewall in one direction, will implicitly allow traffic of established
connections (and traffic of related connections) to also pass in the reverse direction. Application level gateway (ALG) helper functions can be enabled to provide
connection tracking of more complex protocols, such as FTP and SIP.
Section 31.1 describes the firewall functionality available in WeOS. Sections 31.2
and 31.3 cover firewall management via the Web Interface and via the CLI.
740
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
31.1
Overview
Table 31.1 summarises the supported firewall functionality. Sections 31.1.1-31.1.5
provide further information on the WeOS firewall support.
Feature
Enable Firewall
Packet filtering
Enable Packet Filtering
Filtering Rules
Rule Reordering
Activate/Deactivate Rules
Default Forward Policy
Default Input Policy
Stateful Packet Inspection
Packet modification
DSCP
Network Address Translation
NAPT
1-TO-1 NAT
Port Forwarding
ALG Helpers
Logging
View Firewall Configuration
View Firewall Status
Web
X
CLI
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
General Description
Sections 31.1.1-31.1.2
Sections 31.1.1-31.1.2
Sections 31.1.1-31.1.2
Sections 31.1.1-31.1.2
Sections 31.1.1-31.1.2
Sections 31.1.1-31.1.2
Sections 31.1.1-31.1.2
Sections 31.1.1-31.1.2
Sections 31.1.1-31.1.2
Sections 31.1.1, 31.1.3
Section 31.1.3.3
X
X
X
X
X
X
X
X
X
X
Sections 31.1.1, 31.1.4
Sections 31.1.1, 31.1.4
Sections 31.1.1, 31.1.5
Section 31.1.1
Section 31.1.6
X
X
X
Table 31.1: Summary of Firewall functionality in WeOS
31.1.1
Firewall introduction
The WeOS firewall includes support for three related types of functionality:
ˆ Packet Filtering: The packet filtering support is primarily used to control
what traffic is allowed to be routed via the switch (forward filtering), but can
also be used to control accessibility to services on the switch itself (input
filtering).
© 2017 Westermo Teleindustri AB
741
Westermo OS Management Guide
Version 4.21.1-0
The WeOS firewall utilises connection tracking; a filter rule allowing traffic
to pass through the firewall in one direction, will implicitly allow traffic of
established connections (and traffic of related connections) to also pass in
the reverse direction. Connection tracking can configured to handle more
complex protocols by enabling ALG helpers (see below).
WeOS supports up to 1024 filtering rules. The WeOS packet filtering support
is further described in sections 31.1.2 and 31.1.2.3.
ˆ Packet modification: WeOS currently supports one kind of packet modification:
– DSCP: The Differentiated Services Code Point (DSCP) field of the IP
header is used for classifying traffic in some environments. The value
of this field can be modified by WeOS when routing the IP packets.
WeOS supports up to 32 packet modifier rules. The WeOS packet modification support is further described in section 31.1.3.
ˆ Network Address Translation (NAT): WeOS supports two kinds of NAT support:
– NAPT: NAPT is the most common NAT form, where a common (public) IP
address is shared by a set of hosts in a private network. This form of NAT
is sometimes referred to as IP Masquerading or port address translation
(PAT). NAPT is often used together with port forwarding, see below.
– 1-TO-1 NAT: 1-TO-1 NAT enables you to translate a whole range of IP
addresses to another set of addresses.
WeOS supports up to 512 NAT rules. The WeOS NAT support is further described in section 31.1.4.
ˆ Port Forwarding: Port forwarding is commonly used together with NAPT. With
port forwarding a service (such as a Web Server) located in a private network, can be made accessible from the public network, typically from the
Internet.
WeOS supports up to 256 port forwarding rules. The WeOS port forwarding
support is further described in section 31.1.5.
Some network protocols are more complex and therefore more difficult than others to handle by the connection tracking function in a firewall or NAT device. An
example is FTP, which utilises a control connection to exchange information on
TCP port numbers for data connections for the actual file transfers – to enable a
PC to download files through a firewall from an FTP server on the Internet, the
742
© 2017 Westermo Teleindustri AB
Westermo OS Management Guide
Version 4.21.1-0
firewall must inspect the FTP control connection to learn which connections to let
through. To make the firewall handle such protocols correctly, protocol specific
ALG helpers can be enabled. As of WeOS v4.21.1 ALG helpers for FTP, TFTP, SIP,
IRC, H323 and PPTP are supported. ALG helpers have some impact on the unit’s
routing performance, thus are by default disabled.
31.1.2
Packet Filtering
To Switch
(HTTP, SSH, SNMP, ...)
INPUT
FILTERING
Packet
Modification
OUTPUT
FILTERING
FORWARD
FILTERING
FORWARD
MODIFICATION
PREROUTING
From Switch
(HTTP, SSH, SNMP, ...)
Packet
Filtering
Port
Forwarding
NAPT
POSTROUTING
1−1 NAT
NETWORK
NETWORK
Figure 31.1: Overview of Firewall mechanism. Thick lines represent packet flows.
Fig. 31.1 presents an overview of the firewall mechanism, including the components for packet filtering, packet modification, NAT, and port forwarding.
The following sections provide a more in-depth description of the WeOS packet
filtering functions.
ˆ Filtering chains (input, forward, output): Filter rules can apply to
– traffic destined to the switch (input filtering), e.g., HTTP traffic to manage the switch,
– traffic forwarded/routed by the switch (forward filtering), or
– traffic generated by the switch (output filtering).
© 2017 Westermo Teleindustri AB
743
Westermo OS Management Guide
Version 4.21.1-0
The WeOS firewall supports input and forward filtering, but not output filtering. Section 31.1.2.1 gives more details on WeOS handling of filtering
chains.
ˆ Configurable allow/deny filter rules: The user can define filter rules to specify traffic to be allowed or denied, and the order of the configured rules.
Incoming packets are evaluated against the filter rules – the first matching
rule will decide how to treat the packet (allow or deny). Section 31.1.2.2
describes packet matching parameters for filter rules, and section 31.1.2.3
provides more information on filter evaluation order (both for configured
filter rules and implicit filter rules described below).
Default rules to allow ”ping”
When enabling the firewall, the user is offered to add a set of default
rules - these rules allow ICMP packet to pass the input filter, thereby
enabling operators to ping the unit after enabling the firewall. These
rules are treated as any other configured rule, thus can be removed,
etc.
ˆ Implicit filter rules: The WeOS firewall implicitly adds firewall rules for services enabled on the unit, e.g., for DHCP, OSPF or DNS. The primary purpose
of this is to simplify management of those services when the firewall is enab