Red Hat Enterprise Linux 7 Networking Guide - X

Red Hat Enterprise Linux 7
Networking Guide
Configuration and Administration of networking for Red Hat Enterprise
Linux 7
Stephen Wadeley
Christian Huffman
Red Hat Enterprise Linux 7 Networking Guide
Configuration and Administration of networking for Red Hat Enterprise
Linux 7
Stephen Wadeley
Red Hat Engineering Co ntent Services
swadeley@redhat.co m
Christian Huffman
Red Hat Engineering Co ntent Services
chuffman@redhat.co m
Legal Notice
Co pyright © 20 10 –20 14 Red Hat, Inc.
This do cument is licensed by Red Hat under the Creative Co mmo ns Attributio n-ShareAlike 3.0
Unpo rted License. If yo u distribute this do cument, o r a mo dified versio n o f it, yo u must pro vide
attributio n to Red Hat, Inc. and pro vide a link to the o riginal. If the do cument is mo dified, all Red
Hat trademarks must be remo ved.
Red Hat, as the licenso r o f this do cument, waives the right to enfo rce, and agrees no t to assert,
Sectio n 4 d o f CC-BY-SA to the fullest extent permitted by applicable law.
Red Hat, Red Hat Enterprise Linux, the Shado wman lo go , JBo ss, MetaMatrix, Fedo ra, the Infinity
Lo go , and RHCE are trademarks o f Red Hat, Inc., registered in the United States and o ther
co untries.
Linux ® is the registered trademark o f Linus To rvalds in the United States and o ther co untries.
Java ® is a registered trademark o f Oracle and/o r its affiliates.
XFS ® is a trademark o f Silico n Graphics Internatio nal Co rp. o r its subsidiaries in the United
States and/o r o ther co untries.
MySQL ® is a registered trademark o f MySQL AB in the United States, the Euro pean Unio n and
o ther co untries.
No de.js ® is an o fficial trademark o f Jo yent. Red Hat So ftware Co llectio ns is no t fo rmally
related to o r endo rsed by the o fficial Jo yent No de.js o pen so urce o r co mmercial pro ject.
The OpenStack ® Wo rd Mark and OpenStack Lo go are either registered trademarks/service
marks o r trademarks/service marks o f the OpenStack Fo undatio n, in the United States and o ther
co untries and are used with the OpenStack Fo undatio n's permissio n. We are no t affiliated with,
endo rsed o r spo nso red by the OpenStack Fo undatio n, o r the OpenStack co mmunity.
All o ther trademarks are the pro perty o f their respective o wners.
Abstract
The Red Hat Enterprise Linux 7 Netwo rking Guide do cuments relevant info rmatio n regarding the
co nfiguratio n and administratio n o f netwo rk interfaces, netwo rks and netwo rk services in Red
Hat Enterprise Linux 7. It is o riented to wards system administrato rs with a basic understanding
o f Linux and netwo rking. This bo o k is based o n the Red Hat Enterprise Linux 6 Deplo yment
Guide. The chapters related to netwo rking were taken fro m the Deplo yment Guide to fo rm the
fo undatio n fo r this bo o k.
T able of Cont ent s
T able of Contents
. .art
⁠P
. . .I.. IP
. . Net
. . . working
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4. . . . . . . . . .
. .hapt
⁠C
. . . .er
. .1. .. Int
. . .roduct
. . . . . .ion
. . .t.o. Red
. . . . Hat
. . . . Ent
. . . erprise
. . . . . . .Linux
. . . . . Net
. . . working
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5. . . . . . . . . .
⁠1.1. Ho w this Bo o k is Struc tured
5
⁠1.2. IP Netwo rks vers us no n-IP Netwo rks
5
⁠1.3. Intro d uc tio n to Netwo rkManag er
5
⁠1.4. Ins talling Netwo rkManag er
6
⁠1.5. Netwo rk Co nfig uratio n Us ing a Text Us er Interfac e (nmtui)
7
⁠1.6 . Netwo rk Co nfig uratio n Us ing Netwo rkManag er' s CLI (nmc li)
8
⁠1.7. Netwo rk Co nfig uratio n Us ing the Co mmand -Line Interfac e (CLI)
8
⁠1.8 . Netwo rkManag er and the Netwo rk Sc rip ts
9
⁠1.9 . Netwo rk Co nfig uratio n Us ing s ys c o nfig Files
10
⁠1.10 . Ad d itio nal Res o urc es
11
. .hapt
⁠C
. . . .er
. .2. .. Configure
. . . . . . . . . IP
. . Net
. . . .working
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. 3. . . . . . . . . .
⁠2 .1. Static and Dynamic Interfac e Setting s
13
⁠2 .2. Us ing the Text Us er Interfac e, nmtui
14
⁠2 .3. Us ing the Netwo rkManag er Co mmand Line To o l, nmc li
14
⁠2 .4. Us ing the Co mmand Line Interfac e (CLI)
⁠2 .5. Us ing Netwo rkManag er with the G NO ME G rap hic al Us er Interfac e
⁠2 .6 . Ad d itio nal Res o urc es
25
30
53
. .hapt
⁠C
. . . .er
. .3.
. .Configure
. . . . . . . . . Host
. . . . .Names
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
...........
⁠3 .1. Und ers tand ing Ho s t Names
54
⁠3 .2. Co nfig uring Ho s t Names Us ing Text Us er Interfac e, nmtui
54
⁠3 .3. Co nfig uring Ho s t Names Us ing ho s tnamec tl
⁠3 .4. Co nfig uring Ho s t Names Us ing nmc li
55
56
⁠3 .5. Ad d itio nal Res o urc es
57
. .hapt
⁠C
. . . .er
. .4. .. Configure
. . . . . . . . . Net
. . . work
. . . . .Bonding
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
...........
⁠4 .1. Und ers tand ing the Default Behavio r o f Mas ter and Slave Interfac es
58
⁠4 .2. Co nfig ure Bo nd ing Us ing the Text Us er Interfac e, nmtui
⁠4 .3. Us ing the Netwo rkManag er Co mmand Line To o l, nmc li
58
63
⁠4 .4. Us ing the Co mmand Line Interfac e (CLI)
⁠4 .5. Us ing Channel Bo nd ing
64
67
⁠4 .6 . Creating a Bo nd Co nnec tio n Us ing a G UI
⁠4 .7. Ad d itio nal Res o urc es
74
80
. .hapt
⁠C
. . . .er
. .5.
. .Configure
. . . . . . . . . Net
. . . work
. . . . .T. eaming
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .8. 1. . . . . . . . . .
⁠5 .1. Und ers tand ing Netwo rk Teaming
81
⁠5 .2. Und ers tand ing the Default Behavio r o f Mas ter and Slave Interfac es
81
⁠5 .3. Co mp aris o n o f Netwo rk Teaming to Bo nd ing
⁠5 .4. Und ers tand ing the Netwo rk Teaming Daemo n and the " Runners "
⁠5 .5. Ins tall the Netwo rk Teaming Daemo n
⁠5 .6 . Co nverting a Bo nd to a Team
82
83
83
83
⁠5 .7. Selec ting Interfac es to Us e as Po rts fo r a Netwo rk Team
⁠5 .8 . Selec ting Netwo rk Team Co nfig uratio n Metho d s
⁠5 .9 . Co nfig ure a Netwo rk Team Us ing the Text Us er Interfac e, nmtui
⁠5 .10 . Co nfig ure a Netwo rk Team Us ing the Co mmand Line
85
85
85
90
⁠5 .11. Co ntro lling teamd with teamd c tl
⁠5 .12. Co nfig ure teamd Runners
⁠5 .13. Creating a Netwo rk Team Us ing a G UI
⁠5 .14. Ad d itio nal Res o urc es
98
99
10 6
111
1
Red Hat Ent erprise Linux 7 Net working G uide
. .hapt
⁠C
. . . .er
. .6. .. Configure
. . . . . . . . . Net
. . . work
. . . . .Bridging
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1.1. 2. . . . . . . . . .
⁠6 .1. Co nfig ure Brid g ing Us ing the Text Us er Interfac e, nmtui
112
⁠6 .2.
⁠6 .3.
⁠6 .4.
⁠6 .5.
Us ing the Netwo rkManag er Co mmand Line To o l, nmc li
Us ing the Co mmand Line Interfac e (CLI)
Co nfig ure Netwo rk Brid g ing Us ing a G UI
Ad d itio nal Res o urc es
115
117
120
125
. .hapt
⁠C
. . . .er
. .7. .. Configure
. . . . . . . . . 8. 0
. .2..1. Q
. . VLAN
. . . . . .t agging
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1.2. 7. . . . . . . . . .
⁠7 .1. Selec ting VLAN Interfac e Co nfig uratio n Metho d s
⁠7 .2. Co nfig ure 8 0 2.1Q VLAN tag g ing Us ing the Text Us er Interfac e, nmtui
⁠7 .3. Co nfig ure 8 0 2.1Q VLAN Tag g ing Us ing the Co mmand Line To o l, nmc li
⁠7 .4. Co nfig ure 8 0 2.1Q VLAN Tag g ing Us ing the Co mmand Line
127
127
129
132
⁠7 .5. Co nfig ure 8 0 2.1Q VLAN Tag g ing Us ing a G UI
⁠7 .6 . Ad d itio nal Res o urc es
134
136
. .hapt
⁠C
. . . .er
. .8. .. Consist
. . . . . . .ent
. . . Net
. . . .work
. . . . Device
. . . . . . Naming
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1. 37
...........
⁠8 .1. Naming Sc hemes Hierarc hy
137
⁠8 .2. Und ers tand ing the Devic e Renaming Pro c ed ure
137
⁠8 .3. Und ers tand ing the Pred ic tab le Netwo rk Interfac e Devic e Names
⁠8 .4. Naming Sc heme fo r Netwo rk Devic es Availab le fo r Linux o n Sys tem z
138
138
⁠8 .5. Naming Sc heme fo r VLAN Interfac es
139
⁠8 .6 . Co ns is tent Netwo rk Devic e Naming Us ing b io s d evname
⁠8 .7. No tes fo r Ad minis trato rs
139
140
⁠8 .8 . Co ntro lling the Selec tio n o f Netwo rk Devic e Names
⁠8 .9 . Dis ab ling Co ns is tent Netwo rk Devic e Naming
141
141
⁠8 .10 . Tro ub les ho o ting Netwo rk Devic e Naming
142
⁠8 .11. Ad d itio nal Res o urc es
143
. .art
⁠P
. . .II.. .InfiniBand
. . . . . . . . . and
. . . . RDMA
. . . . . .Net
. . . working
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1.4. 4. . . . . . . . . .
. .hapt
⁠C
. . . .er
. .9. .. Configure
. . . . . . . . . InfiniBand
. . . . . . . . . .and
. . . .RDMA
. . . . . .Net
. . .works
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1. 4. 5. . . . . . . . . .
⁠9 .1. Und ers tand ing InfiniBand and RDMA tec hno lo g ies
145
⁠9 .2. InfiniBand and RDMA related s o ftware p ac kag es
⁠9 .3. Co nfig uring the Bas e RDMA Sub s ys tem
146
147
⁠9 .4. Co nfig uring the Sub net Manag er
149
⁠9 .5. Tes ting Early InfiniBand RDMA o p eratio n
⁠9 .6 . Co nfig uring IPo IB
151
154
⁠9 .7. Co nfig ure InfiniBand Us ing the Text Us er Interfac e, nmtui
156
⁠9 .8 . Co nfig ure IPo IB us ing the c o mmand -line to o l, nmc li
⁠9 .9 . Co nfig ure IPo IB Us ing the c o mmand line
158
159
⁠9 .10 . Tes ting an RDMA netwo rk after IPo IB is c o nfig ured
⁠9 .11. Co nfig ure IPo IB Us ing a G UI
16 1
16 1
⁠9 .12. Ad d itio nal Res o urc es
16 3
. .art
⁠P
. . .III.
. . Servers
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1.6. 4. . . . . . . . . .
. .hapt
⁠C
. . . .er
. .1. 0. .. DHCP
. . . . . .Servers
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1. 6. 5. . . . . . . . . .
⁠10 .1. Why Us e DHCP?
⁠10 .2. Co nfig uring a DHCP Server
16 5
16 5
⁠10 .3. DHCP Relay Ag ent
172
⁠10 .4. Co nfig uring a Multiho med DHCP Server
⁠10 .5. DHCP fo r IPv6 (DHCPv6 )
173
176
⁠10 .6 . Ad d itio nal Res o urc es
176
. .hapt
⁠C
. . . .er
. .1. 1. .. DNS
. . . . .Servers
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1.7. 8. . . . . . . . . .
2
T able of Cont ent s
. .hapt
⁠C
. . . .er
. .1. 1. .. DNS
. . . . .Servers
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1.7. 8. . . . . . . . . .
⁠11.1. Intro d uc tio n to DNS
178
⁠11.2. BIND
179
. . . . . . . . .Hist
Revision
. . . ory
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2. 0. 5. . . . . . . . . .
⁠A .1. Ac kno wled g ments
20 5
⁠I.ndex
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2. 0. 5. . . . . . . . . .
3
Red Hat Ent erprise Linux 7 Net working G uide
⁠Part I. IP Networking
This part describes how to configure the network on Red Hat Enterprise Linux.
4
⁠Chapt er 1 . Int roduct ion t o Red Hat Ent erprise Linux Net working
Chapter 1. Introduction to Red Hat Enterprise Linux Networking
1.1. How t his Book is St ruct ured
All new material in this book has been written and arranged in such a way as to clearly separate
introductory material, such as explanations of concepts and use cases, from configuration tasks.
Red Hat Engineering Content Services hope that you can quickly find configuration instructions you
need, while still providing some relevant explanations and conceptual material to help you
understand and decide on the appropriate tasks relevant to your needs. Where material has been
reused from the Red Hat Enterprise Linux 6 Deployment Guide, it has been reviewed and changed,
where possible, to fit this idea of separating concepts from tasks.
The material is grouped according to the goal rather than the method. Instructions on how to achieve
a specific task using different methods are grouped together. This is intended to make it easier for
you to find the information on how to achieve a particular task or goal, and at the same time allow
you to quickly see the different methods available.
In each chapter, the configuration methods will be presented in the following order:
the text user interface tool, n mt u i,
N et wo rkMan ag er's command-line tool n mcli,
other command-line methods and the use of configuration files,
a graphical user interface (GUI) method, such as the use of n m- co n n ect io n - ed it o r or
co n t ro l- n et wo rk to direct Netwo rkManag er.
The command line can be used to issue commands, hence the term command-line interface (CLI)
however the command line can also start an editor, to compose or edit configuration files. Therefore
the use of ip commands and configuration files, such as i fcfg files, will be documented together.
1.2. IP Net works versus non-IP Net works
Most modern networks fall into one of two very broad categories: IP based networks. These are all
networks that communicate via Internet Protocol addresses, which is the standard for the Internet and
for most internal networks today. This generally includes Ethernet, Cable Modems, D SL Modems, dial
up modems, Wi-Fi, VPN connections and more.
Then there are non-IP based networks. These are usually very specific niche networks, but one in
particular has grown in usage enough to warrant mention here and that is InfiniBand. Because
InfiniBand is not an IP network, many features and configurations normally used on IP networks are
not applicable to InfiniBand. Chapter 9, Configure InfiniBand and RDMA Networks in this guide covers
the specific requirements of configuring and administrating an InfiniBand network and also the
broader class of RD MA capable devices.
1.3. Int roduct ion t o Net workManager
In Red Hat Enterprise Linux 7, the default networking service is provided by Netwo rkManag er, which
is a dynamic network control and configuration daemon that attempts to keep network devices and
connections up and active when they are available. The traditional i fcfg type configuration files
are still supported. See Section 1.8, “ NetworkManager and the Network Scripts” for more information.
T ab le 1.1. A Su mmary o f N et wo rkin g T o o ls an d Ap p licat io n s
5
Red Hat Ent erprise Linux 7 Net working G uide
Ap p licat io n o r T o o l
D escrip t io n
N et wo rkMan ag er
n mt u i
n mcli
The default networking daemon
A simple curses-based text user interface (TUI) for N et wo rkMan ag er
A command-line tool provided to allow users and scripts to interact
with N et wo rkMan ag er
A graphical user interface tool provided by the GNOME Shell
A GTK+ 3 application available for certain tasks not yet handled by
co n t ro l- cen t er
co n t ro l- cen t er
n m- co n n ect io n - ed it o r
Netwo rkManag er can be used with the following types of connections: Ethernet, VLANs, Bridges,
Bonds, Teams, Wi-Fi, mobile broadband (such as cellular 3G), and IP-over-InfiniBand. For these
connection types, Netwo rkManag er can configure network aliases, IP addresses, static routes, D NS
information, and VPN connections, as well as many connection-specific parameters. Finally,
Netwo rkManag er provides an API via D -Bus which allows applications to query and control
network configuration and state.
1.4 . Inst alling Net workManager
N et wo rkMan ag er is installed by default on Red Hat Enterprise Linux. If necessary, to ensure that it
is, run the following command as the ro o t user:
~]# yum i nstal l Netwo rkManag er
For information on user privileges and gaining privileges, see the Red Hat Enterprise Linux 7 System
Administrator's Guide.
1.4 .1. T he Net workManager Daemon
The N et wo rkMan ag er daemon runs with root privileges and is, by default, configured to start up at
boot time. You can determine whether the N et wo rkMan ag er daemon is running by entering this
command:
~]$ systemctl status Netwo rkManag er
NetworkManager.service - Network Manager
Loaded: loaded (/lib/systemd/system/NetworkManager.service; enabled)
Active: active (running) since Fri, 08 Mar 2013 12:50:04 +0100; 3 days
ago
The systemctl status command will report N et wo rkMan ag er as Acti ve: i nacti ve (d ead )
if the N et wo rkMan ag er service is not running. To start it for the current session run the following
command as the root user:
~]# systemctl start Netwo rkManag er
Run the systemctl enabl e command to ensure that N et wo rkMan ag er starts up every time the
system boots:
~]# systemctl enabl e Netwo rkManag er
For more information on starting, stopping and managing services, see the Red Hat Enterprise Linux 7
System Administrator's Guide.
1.4 .2. Int eract ing wit h Net workManager
6
⁠Chapt er 1 . Int roduct ion t o Red Hat Ent erprise Linux Net working
1.4 .2. Int eract ing wit h Net workManager
Users do not interact with the N et wo rkMan ag er system service directly. Instead, users perform
network configuration tasks via graphical and command-line user interface tools. The following tools
are available in Red Hat Enterprise Linux 7:
1. A simple curses-based text user interface (TUI) for N et wo rkMan ag er, n mt u i, is available.
2. A command-line tool, n mcli, is provided to allow users and scripts to interact with
N et wo rkMan ag er. Note that n mcli can be used on GUI-less systems like servers to control
all aspects of N et wo rkMan ag er. It is on an equal footing with the GUI tools.
3. The GNOME Shell also provides a network icon in its Notification Area representing network
connection states as reported by N et wo rkMan ag er. The icon has multiple states that serve
as visual indicators for the type of connection you are currently using.
4. A graphical user interface tool called co n t ro l- cen t er, provided by the GNOME Shell, is
available for desktop users. It incorporates a N et wo rk settings tool. To start it, press the
Super key to enter the Activities Overview, type co ntro l netwo rk and then press Enter.
The Super key appears in a variety of guises, depending on the keyboard and other
hardware, but often as either the Windows or Command key, and typically to the left of the
Spacebar.
5. A graphical user interface tool, n m- co n n ect io n - ed it o r, is available for certain tasks not yet
handled by co n t ro l- cen t er. To start it, press the Super key to enter the Activities Overview,
type netwo rk co nnecti o ns or nm-co nnecti o n-ed i to r and then press Enter.
Fig u re 1.1. N et wo rk co n n ect io n ico n st at es
1.5. Net work Configurat ion Using a T ext User Int erface (nmt ui)
The N et wo rkMan ag er text user interface (TUI) tool, n mt u i, provides a text interface to configure
networking by controlling N et wo rkMan ag er. The tool is contained in the subpackage
NetworkManager-tui. At time of writing, it is not installed along with N et wo rkMan ag er by default. To
install NetworkManager-tui, issue the following command as ro o t:
~]# yum i nstal l Netwo rkManag er-tui
If required, for details on how to verify that N et wo rkMan ag er is running, see Section 1.4.1, “ The
NetworkManager D aemon” .
To start n mt u i, issue a command as follows:
~]$ nmtui
The text user interface appears. To navigate, use the arrow keys or press T ab to step forwards and
press Shi ft+T ab to step back through the options. Press Enter to select an option. The Space bar
toggles the status of a check box.
The following commands are available:
7
Red Hat Ent erprise Linux 7 Net working G uide
nmtui ed i t connection-name
If no connection name is supplied, the selection menu appears. If the connection name is
supplied and correctly identified, the relevant Ed i t co nnecti o n screen appears.
nmtui co nnect connection-name
If no connection name is supplied, the selection menu appears. If the connection name is
supplied and correctly identified, the relevant connection is activated. Any invalid command prints
a usage message.
At time of writing, n mt u i does not support all types of connections. In particular, you cannot edit
VPNs, Wi-Fi connections using WPA Enterprise, or Ethernet connections using 80 2. 1X.
1.6. Net work Configurat ion Using Net workManager's CLI (nmcli)
The N et wo rkMan ag er command-line tool, n mcli, provides a command line way to configure
networking by controlling N et wo rkMan ag er. It is installed, along with N et wo rkMan ag er, by
default. If required, for details on how to verify that N et wo rkMan ag er is running, see Section 1.4.1,
“ The NetworkManager D aemon” .
Examples of using the n mcli tool for each task will be included where possible, before explaining the
use of other command-line methods and graphical user interfaces. See Section 2.3, “ Using the
NetworkManager Command Line Tool, nmcli” for an introduction to n mcli.
1.7. Net work Configurat ion Using t he Command-Line Int erface (CLI)
The commands for the ip utility, sometimes referred to as iproute2 after the upstream package name,
are documented in the man i p(8) page. The package name in Red Hat Enterprise Linux 7 is iproute.
If necessary, you can check that the ip utility is installed by checking its version number as follows:
~]$ i p -V
ip utility, iproute2-ss130716
The ip commands can be used to add and remove addresses and routes to interfaces in parallel with
N et wo rkMan ag er, which will preserve them and recognize them in n mcli, n mt u i, co n t ro l- cen t er,
and the D -Bus API.
Note that the ip utility replaces the i fco nfi g utility because the net-tools package (which provides
i fco nfi g ) does not support InfiniBand addresses. The command i p hel p prints a usage
message. Specific help is available for OBJECTS, for example: i p l i nk hel p and i p ad d r
hel p.
Note
ip commands given on the command line will not persist after a system restart. Where
persistence is required, make use of configuration files (i fcfg files) or add the commands to
a script.
8
⁠Chapt er 1 . Int roduct ion t o Red Hat Ent erprise Linux Net working
Examples of using the command line and configuration files for each task are included after n mt u i
and n mcli examples but before explaining the use of one of the graphical user interfaces to
N et wo rkMan ag er, namely, co n t ro l- cen t er and n m- co n n ect io n - ed it o r.
1.8. Net workManager and t he Net work Script s
In previous Red Hat Enterprise Linux releases, the default way to configure networking was using
network scripts. The term network scripts is commonly used for the script /etc/i ni t. d /netwo rk and
any other installed scripts it calls. The user supplied files are typically viewed as configuration, but
can also be interpreted as an amendment to the scripts.
Although N et wo rkMan ag er provides the default networking service, Red Hat developers have
worked hard to ensure that scripts and N et wo rkMan ag er cooperate with each other. Administrators
who are used to the scripts can certainly continue to use them. We expect both systems to be able to
run in parallel and work well together. It is expected that most user shell scripts from previous
releases will still work. Red Hat recommends that you test them first.
Running Net work Script
Run the script o n ly with the syst emct l utility which will clear any existing environment variables and
ensure clean execution. The command takes the following form:
systemctl start| sto p| restart| status netwo rk
D o n o t ru n any service by calling /etc/i ni t. d /servicename start| sto p| restart| status
directly.
Note that in Red Hat Enterprise Linux 7, N et wo rkMan ag er is started first, and
/etc/i ni t. d /netwo rk checks with N et wo rkMan ag er to avoid tampering with
N et wo rkMan ag er's connections. N et wo rkMan ag er is intended to be the primary application
using sysconfig configuration files and /etc/i ni t. d /netwo rk is intended to be secondary,
playing a fallback role.
The /etc/i ni t. d /netwo rk script is not event-driven, it runs either:
1. manually (by one of the systemctl commands start| sto p| restart netwo rk),
2. on boot and shutdown if the network service is enabled (as a result of the command
systemctl enabl e netwo rk).
It is a manual process and does not react to events that happen after boot. Users can also call the
scripts i fup and i fd o wn manually.
Cust om Commands and t he Net work Script s
Custom commands in the scripts /sbi n/i fup-l o cal , i fd o wn-pre-l o cal , and i fd o wnl o cal are only executed when those devices are controlled by the /etc/i ni t. d /netwo rk service.
If you modified the initscripts themselves (for example, /etc/sysco nfi g /netwo rkscri pts/i fup-eth) then those changes would be overwritten by an initscripts package update.
Therefore it is recommend that you avoid modifying the initscripts directly and make use of the
/sbi n/i f*l o cal scripts, so that your custom changes will survive package updates. The
initscripts just check for the presence of the relevant /sbi n/i f*l o cal and run them if they exist.
The initscripts do not place anything in the /sbi n/i f*l o cal scripts, nor does the initscripts RPM
(or any package) own or modify those files.
There are ways to perform custom tasks when network connections go up and down, both with the
9
Red Hat Ent erprise Linux 7 Net working G uide
old network scripts and with N et wo rkMan ag er. When N et wo rkMan ag er is enabled, the i fup and
i fd o wn script will ask N et wo rkMan ag er whether N et wo rkMan ag er manages the interface in
question, which is found from the “ D EVICE=” line in the i fcfg file. If N et wo rkMan ag er does
manage that device, and the device is not already connected, then i fup will ask N et wo rkMan ag er
to start the connection.
If the device is managed by N et wo rkMan ag er and it is already connected, nothing is done.
If the device is not managed by N et wo rkMan ag er, then the scripts will start the connection using
the older, non-N et wo rkMan ag er mechanisms that they have used since the time before
N et wo rkMan ag er existed.
If you are calling " i fd o wn" and the device is managed by N et wo rkMan ag er, then i fd o wn will ask
N et wo rkMan ag er to terminate the connection.
The scripts dynamically check N et wo rkMan ag er, so if N et wo rkMan ag er is not running, the
scripts will fall back to the old, pre-N et wo rkMan ag er script-based mechanisms.
1.9. Net work Configurat ion Using sysconfig Files
The /etc/sysco nfi g / directory is a location for configuration files and scripts. Most network
configuration information is stored there, with the exception of VPN, mobile broadband and PPPoE
configuration, which are stored in /etc/Netwo rkManag er/ subdirectories. Interface specific
information for example, is stored in i fcfg files in the /etc/sysco nfi g /netwo rk-scri pts/
directory.
The file /etc/sysco nfi g /netwo rk is for global settings. Information for VPNs, mobile broadband
and PPPoE connections is stored in /etc/Netwo rkManag er/system-co nnecti o ns/.
In Red Hat Enterprise Linux 7 when you edit an i fcfg file, N et wo rkMan ag er is not automatically
aware of the change and has to be prompted to notice the change. If you use one of the tools to
update N et wo rkMan ag er profile settings, then N et wo rkMan ag er does not implement those
changes until you reconnect using that profile. For example, if configuration files have been changed
using an editor, N et wo rkMan ag er must be told to read the configuration files again. To do that,
issue the following command as ro o t:
~]# nmcl i co nnecti o n rel o ad
The above command reads all connection profiles. Alternatively, to reload only one changed file,
i fcfg -ifname, issue a command as follows:
~]# nmcl i co n l o ad /etc/sysco nfi g /netwo rk-scri pts/i fcfg -ifname
The command accepts multiple file names. These commands require ro o t privileges. For more
information on user privileges and gaining privileges, see the Red Hat Enterprise Linux 7 System
Administrator's Guide and the su(1) and sud o (8) man pages.
Changes made using tools such as n mcli do not require a reload but do require the associated
interface to be put down and then up again. That can be done by using commands in the following
format:
nmcli dev disconnect interface-name
Followed by:
10
⁠Chapt er 1 . Int roduct ion t o Red Hat Ent erprise Linux Net working
nmcli con up interface-name
N et wo rkMan ag er does not trigger any of the network scripts, though the network scripts will try to
trigger N et wo rkMan ag er if it is running when i fup commands are used. See Section 1.8,
“ NetworkManager and the Network Scripts” for an explanation of the network scripts.
The i fup script is a generic script which does a few things and then calls interface-specific scripts
like i fup-ethX, i fup-wi rel ess, i fup-ppp, and so on. When a user runs i fup eth0 manually,
the following occurs:
1. i fup looks for a file called /etc/sysco nfi g /netwo rk-scri pts/i fcfg -eth0 ;
2. if the i fcfg file exists, i fup looks for the T Y P E key in that file to determine which typespecific script to call;
3. i fup calls i fup-wi rel ess or i fup-eth or i fup-XXX based on T Y P E;
4. the type-specific scripts do type-specific setup;
5. and then the type-specific scripts let common functions perform IP -related tasks like D HC P or
static setup.
On bootup, /etc/i ni t. d /netwo rk reads through all the i fcfg files and for each one that has
O NBO O T = yes, it checks whether N et wo rkMan ag er is already starting the D EVICE from that i fcfg
file. If N et wo rkMan ag er is starting that device or has already started it, nothing more is done for
that file, and the next O NBO O T = yes file is checked. If N et wo rkMan ag er is not yet starting that
device, the initscripts will continue with their traditional behavior and call i fup for that i fcfg file.
The end result is that any i fcfg file that has O NBO O T = yes is expected to be started on system
bootup, either by N et wo rkMan ag er or by the initscripts. This ensures that some legacy network
types which N et wo rkMan ag er does not handle (such as ISD N or analog dialup modems) as well
as any new application not yet supported by N et wo rkMan ag er are still correctly started by the
initscripts even though N et wo rkMan ag er is unable to handle them.
Note
It is recommended not to store backup i fcfg files in the same location as the live ones. The
script literally does i fcfg -* with an exclude only for these extensions: . o l d , . o ri g ,
. rpmnew, . rpmo ri g , and . rpmsave. The best way is not to store backup files anywhere
within the /etc/ directory.
1.10. Addit ional Resources
The following sources of information provide additional resources regarding networking for Red Hat
Enterprise Linux 7.
1.10.1. Inst alled Document at ion
man(1) man page — D escribes man pages and how to find them.
Netwo rkManag er(8) man page — D escribes the network management daemon.
Netwo rkManag er. co nf(5) man page — D escribes the Netwo rkManag er configuration file.
11
Red Hat Ent erprise Linux 7 Net working G uide
/usr/share/d o c/i ni tscri pts-version/sysco nfi g . txt — D escribes configuration files
and their directives.
12
⁠Chapt er 2 . Configure IP Net working
Chapter 2. Configure IP Networking
2.1. St at ic and Dynamic Int erface Set t ings
When to use static addressing and when to use dynamic addressing? These decisions are
subjective, they depend on your accessed needs, your specific requirements. Having a policy,
documenting it, and applying it consistently are usually more important than the specific decisions
you make. In a traditional company LAN, this is an easier decision to make as you typically have
fewer servers than other hosts. Provisioning and installation tools make providing static
configurations to new hosts easy and using such tools will change your work flow and requirements.
The following two sections are intended to provide just basic guidance to those who have not
already been through this decision-making process. Experienced system administrators will likely
have their own set of rules and requirements that may differ from what is discussed here. For more
information on automated configuration and management, see the O p en LMI section in the Red Hat
Enterprise Linux 7 System Administrators Guide. The Red Hat Enterprise Linux 7 Installation Guide
documents the use of kickst art which can also be used for automating the assignment of network
settings.
2.1.1. When t o Use St at ic Net work Int erface Set t ings
Use static IP addressing on those servers and devices whose network availability you want to
ensure when automatic assignment methods, such as D HC P , fail. D HC P , D NS, and authentication
servers are typical examples. Interfaces for out-of-band management devices are also worth
configuring with static settings as these devices are supposed to work, as far as is possible,
independently of other network infrastructure.
For hosts which are not considered vital, but for which static IP addressing is still considered
desirable, use an automated provisioning method when possible. For example, D HC P servers can be
configured to provide the same IP address to the same host every time. This method could be used
for communal printers for example.
All the configuration tools listed in Section 2.1.3, “ Selecting Network Configuration Methods” allow
assigning static IP addresses manually. The n mcli tool is also suitable for use with scripted
assignment of network configuration.
2.1.2. When t o Use Dynamic Int erface Set t ings
Enable and use dynamic assignment of IP addresses and other network information whenever there
is no compelling reason not to. The time saved in planning and documenting manual settings can be
better spent elsewhere. The dynamic host control protocol (D HCP) is a traditional method of
dynamically assigning network configurations to hosts. See Section 10.1, “ Why Use D HCP?” for
more information on this subject.
N et wo rkMan ag er will by default call the D HC P client, d h clien t , when a profile has been set to
obtain addresses automatically, or when an interface configuration file has BOOTPROTO set to
d hcp. Where D HC P is required, an instance of d hcl i ent is started for every Internet protocol, IP v4
and IP v6 , on an interface. Where N et wo rkMan ag er is not running, or not managing an interface,
then the legacy network service will call instances of d hcl i ent as required.
2.1.3. Select ing Net work Configurat ion Met hods
T o co n f ig u re an in t erf ace u sin g N et wo rkMan ag er's text user interface tool, n mt u i,
proceed to Section 2.2, “ Using the Text User Interface, nmtui”
13
Red Hat Ent erprise Linux 7 Net working G uide
T o co n f ig u re an in t erf ace u sin g N et wo rkMan ag er's command-line tool, n mcli, proceed to
Section 2.3, “ Using the NetworkManager Command Line Tool, nmcli”
T o co n f ig u re a n et wo rk in t erf ace man u ally, see Section 2.4, “ Using the Command Line
Interface (CLI)” .
T o co n f ig u re a n et wo rk u sin g g rap h ical u ser in t erf ace t o o ls, proceed to Section 2.5,
“ Using NetworkManager with the GNOME Graphical User Interface”
2.2. Using t he T ext User Int erface, nmt ui
The text user interface tool n mt u i can be used to configure an interface in a terminal window. Issue
the following command to start the tool:
~]$ nmtui
The text user interface appears. Any invalid command prints a usage message.
Fig u re 2.1. T h e N et wo rkMan ag er T ext U ser In t erf ace st art in g men u
To navigate, use the arrow keys or press T ab to step forwards and press Shi ft+T ab to step back
through the options. Press Enter to select an option. The Space bar toggles the status of a check
box.
See Section 1.5, “ Network Configuration Using a Text User Interface (nmtui)” for information on
installing n mt u i.
2.3. Using t he Net workManager Command Line T ool, nmcli
The command‐line tool n mcli can be used by both users and scripts for controlling
N et wo rkMan ag er. The basic format of a command is as follows:
nmcli O P T IO NS OBJECT { C O MMAND | help }
14
⁠Chapt er 2 . Configure IP Net working
where OBJECT can be one of g eneral , netwo rki ng , rad i o , co nnecti o n, or d evi ce. The most
used options are: -t, --terse for use in scripts, the -p, --pretty option for users, and the -h,
--hel p option. Command completion has been implemented for n mcli, so remember to press T ab
whenever you are unsure of the command options available. See the nmcl i (1) man page for a
complete list of the options and commands.
The n mcli tool has some built-in context-sensitive help. For example, issue the following two
commands and notice the difference:
~]$ nmcl i hel p
Usage: nmcli [OPTIONS] OBJECT { COMMAND | help }
OPTIONS
-t[erse]
-p[retty]
-m[ode] tabular|multiline
-f[ields] <field1,field2,...>|all|common
-e[scape] yes|no
values
-n[ocheck]
NetworkManager versions
-a[sk]
-w[ait] <seconds>
finishing operations
-v[ersion]
-h[elp]
OBJECT
g[eneral]
n[etworking]
r[adio]
c[onnection]
d[evice]
terse output
pretty output
output mode
specify fields to output
escape columns separators in
don't check nmcli and
ask for missing parameters
set timeout waiting for
show program version
print this help
NetworkManager's general status and operations
overall networking control
NetworkManager radio switches
NetworkManager's connections
devices managed by NetworkManager
~]$ nmcl i g eneral hel p
Usage: nmcli general { COMMAND | help }
COMMAND := { status | hostname | permissions | logging }
status
hostname [<hostname>]
permissions
logging [level <log level>] [domains <log domains>]
In the second example above the help is related to the object g eneral .
The nmcl i -exampl es(5) man page has many useful examples. A brief selection is shown here:
To show the overall status of N et wo rkMan ag er:
nmcli general status
15
Red Hat Ent erprise Linux 7 Net working G uide
To control N et wo rkMan ag er logging:
nmcli general logging
To show all connections:
nmcli connection show
To show only currently active connections, add the -a, --acti ve option as follows:
nmcli connection show --active
To show devices recognized by N et wo rkMan ag er and their state:
nmcli device status
Commands can be shortened and some options omitted. For example the command:
nmcli connection modify id 'MyCafe' 802-11-wireless.mtu 1350
Can be reduced to the following command:
nmcli con mod MyCafe 802-11-wireless.mtu 1350
The i d option can been omitted because the connection ID (name) is unambiguous for n mcli in this
case. As you become familiar with the commands, further abbreviations can be made. For example:
nmcli connection add type ethernet
can be reduced to:
nmcli c a type eth
Note
Remember to use tab completion when in doubt.
St art ing and St opping an Int erface Using nmcli
The n mcli tool can be used to start and stop any network interface, including masters. For example:
nmcli
nmcli
nmcli
nmcli
16
con
con
dev
dev
up id bond0
up id port0
disconnect iface bond0
disconnect iface ens3
⁠Chapt er 2 . Configure IP Net working
Note
It is recommended to use nmcl i d ev d i sco nnect i face iface-name rather than
nmcl i co n d o wn i d id-string because disconnection places the interface into a
“ manual” mode, in which no automatic connection will be started until the user tells
N et wo rkMan ag er to start a connection or until an external event like a carrier change,
hibernate, or sleep, occurs.
T he nmcli Int eract ive Connect ion Edit or
The n mcli tool has an interactive connection editor. To use it, enter the following command:
~]$ nmcl i co n ed i t
You will be prompted to enter a valid connection type from the list displayed. After entering a
connection type you will be placed at the n mcli prompt. If you are familiar with the connection types
you can add a valid connection type option to the nmcl i co n ed i t command and be taken
straight to the n mcli prompt. The format is as follows for editing an existing connection profile:
nmcli con edit [id | uuid | path] ID
For adding and editing a new connection profile, the following format applies:
nmcli con edit [type new-connection-type] [con-name new-connection-name]
Type hel p at the n mcli prompt to see a list of valid commands. Use the d escri be command to get
a description of settings and their properties. The format is as follows:
describe setting.property
For example:
nmcli> d escri be team. co nfi g
2.3.1. Underst anding t he nmcli Opt ions
Many of the n mcli commands are self-explanatory, however a few command options are worth a
moments study:
type — T h e co n n ect io n t yp e.
Allowed values are: ad sl , bo nd , bo nd -sl ave, bri d g e, bri d g e-sl ave, bl ueto o th,
cd ma, ethernet, g sm, i nfi ni band , o l pc-mesh, team, team-sl ave, vl an, wi fi ,
wi max.
Each connection type has type-specific command options. Press T ab to see a list of them
or see the T Y P E_SP EC IFIC _O P T IO NS list in the nmcl i (1) man page. The type option
is applicable after the following: nmcl i co nnecti o n ad d and nmcl i co nnecti o n
ed i t.
co n-name — T h e n ame assig n ed t o a co n n ect io n p ro f ile.
17
Red Hat Ent erprise Linux 7 Net working G uide
If you do not specify a connection name, one will be generated as follows:
type-ifname[-number]
The connection name is the name of a connection profile and should not be confused with
the interface name that denotes a device (wlan0, ens3, em1, and so on). Users can however
name the connections after interfaces, but they are not the same thing. There can be
multiple connection profiles available for a device. This is particularly useful for mobile
devices or when switching a network cable back and forth between different devices. Rather
than edit the configuration, create different profiles and apply them to the interface as
needed. The i d option also refers to the connection profile name.
i d — An id en t if icat io n st rin g assig n ed b y t h e u ser t o a co n n ect io n p ro f ile.
The ID can be used in nmcl i co nnecti o n commands to identify a connection. The
NAME field in the output always denotes the connection ID (name). It refers to the same
connection profile name that the co n-name does.
uui d — A u n iq u e id en t if icat io n st rin g assig n ed b y t h e syst em t o a co n n ect io n
p ro f ile.
The UUID can be used in nmcl i co nnecti o n commands to identify a connection.
2.3.2. Connect ing t o a Net work Using nmcli
To list the currently available network connections, issue a command as follows:
~]$ nmcl i co n sho w
NAME
UUID
TYPE
DEVICE
Auto Ethernet
9b7f2511-5432-40ae-b091-af2457dfd988 802-3-ethernet ens3
fb157a65-ad32-47ed-858c-102a48e064a2 802-3-ethernet
ens3
MyWiFi
91451385-4eb8-4080-8b82-720aab8328dd 802-11-wireless
wlan0
Note that the NAME field in the output always denotes the connection ID (name). It is not the interface
name even though it might look the same. In the example above on the second line ens3 in the NAME
field is the connection ID given by the user to the profile applied to the interface ens3. In the last line
the user has assigned the connection ID MyWi Fi to the interface wlan0.
Adding an Ethernet connection means creating a configuration profile which is then assigned to a
device. Before creating a new profile, review the available devices as follows:
~]$ nmcl i d ev status
DEVICE TYPE
STATE
ens3
ethernet disconnected
ens9
ethernet disconnected
lo
loopback unmanaged
CONNECTION
----
Adding a Dynam ic Et he rne t Co nne ct io n
To add an Ethernet configuration profile with dynamic IP configuration, allowing D HC P to assign the
network configuration, a command in the following format can be used:
18
⁠Chapt er 2 . Configure IP Net working
nmcli connection add type ethernet con-name connection-name ifname
interface-name
For example, to create a dynamic connection profile named my-office, issue a command as follows:
~]$ nmcl i co n ad d type ethernet co n-name my-office i fname ens3
Connection 'my-office' (fb157a65-ad32-47ed-858c-102a48e064a2) successfully
added.
N et wo rkMan ag er will set its internal parameter co nnecti o n. auto co nnect to yes.
N et wo rkMan ag er will also write out settings to /etc/sysco nfi g /netwo rk-scri pts/i fcfg my-o ffi ce where the ONBOOT directive will be set to yes.
Note that manual changes to the ifcfg file will not be noticed by N et wo rkMan ag er until the interface
is next brought up. See Section 1.9, “ Network Configuration Using sysconfig Files” for more
information on using configuration files.
To bring up the Ethernet connection, issue a command as follows:
~]$ nmcl i co n up my-office
Connection successfully activated (D-Bus active path:
/org/freedesktop/NetworkManager/ActiveConnection/5)
Review the status of the devices and connections:
~]$ nmcl i d evi ce
DEVICE TYPE
ens3
ethernet
ens9
ethernet
lo
loopback
status
STATE
connected
disconnected
unmanaged
CONNECTION
my-office
---
To change the host name sent by a host to a D HC P server, modify the d hcp-ho stname property as
follows:
~]$ nmcl i co n mo d i fy my-o ffi ce my-office i pv4 . d hcp-ho stname host-name
i pv6 . d hcp-ho stname host-name
To change the IP v4 client ID sent by a host to a D HC P server, modify the d hcp-cl i ent-i d
property as follows:
~]$ nmcl i co n mo d i fy my-o ffi ce my-office i pv4 . d hcp-cl i ent-i d clientID-string
There is no d hcp-cl i ent-i d property for IP v6 , d h clien t creates an identifier for IP v6 . See the
d hcl i ent(8) man page for details.
To ignore the D NS servers sent to a host by a D HC P server, modify the i g no re-auto -d ns property
as follows:
~]$ nmcl i co n mo d i fy my-o ffi ce my-office i pv4 . i g no re-auto -d ns yes
i pv6 . i g no re-auto -d ns yes
See the nm-setti ng s(5) man page for more information on properties and their settings.
19
Red Hat Ent erprise Linux 7 Net working G uide
Examp le 2.1. C o n f ig u rin g a D yn amic Et h ern et C o n n ect io n U sin g t h e In t eract ive
Ed it o r
To configure a dynamic Ethernet connection using the interactive editor, issue commands as
follows:
~]$ nmcl i co n ed i t type ethernet co n-name ens3
===| nmcli interactive connection editor |===
Adding a new '802-3-ethernet' connection
Type 'help' or '?' for available commands.
Type 'describe [<setting>.<prop>]' for detailed property description.
You may edit the following settings: connection, 802-3-ethernet
(ethernet), 802-1x, ipv4, ipv6, dcb
nmcli> describe ipv4.method
=== [method] ===
[NM property description]
IPv4 configuration method. If 'auto' is specified then the appropriate
automatic method (DHCP, PPP, etc) is used for the interface and most
other properties can be left unset. If 'link-local' is specified, then
a link-local address in the 169.254/16 range will be assigned to the
interface. If 'manual' is specified, static IP addressing is used and
at least one IP address must be given in the 'addresses' property. If
'shared' is specified (indicating that this connection will provide
network access to other computers) then the interface is assigned an
address in the 10.42.x.1/24 range and a DHCP and forwarding DNS server
are started, and the interface is NAT-ed to the current default network
connection. 'disabled' means IPv4 will not be used on this connection.
This property must be set.
nmcli> set ipv4.method auto
nmcli> save
Saving the connection with 'autoconnect=yes'. That might result in an
immediate activation of the connection.
Do you still want to save? [yes] yes
Connection 'ens3' (090b61f7-540f-4dd6-bf1f-a905831fc287) successfully
saved.
nmcli> quit
~]$
The default action is to save the connection profile as persistent. If required, the profile can be held
in memory only, until the next restart, by means of the save tempo rary command.
Adding a St at ic Et he rne t Co nne ct io n
To add an Ethernet connection with static IP v4 configuration, a command in the following format
can be used:
nmcli connection add type ethernet con-name connection-name ifname
interface-name ip4 address gw4 address
20
⁠Chapt er 2 . Configure IP Net working
IP v6 address and gateway information can be added using the i p6 and g w6 options.
For example, a command to create a static Ethernet connection with only IP v4 address and gateway
is as follows:
~]$ nmcl i co n ad d type ethernet co n-name test-lab i fname ens9 i p4
10 . 10 . 10 . 10 /24 \
g w4 10 . 10 . 10 . 254
Optionally, at the same time specify IP v6 address and gateway for the device as follows:
~]$ nmcl i co n ad d type ethernet co n-name test-lab i fname ens9 i p4
10 . 10 . 10 . 10 /24 \
g w4 10 . 10 . 10 . 254 i p6 abbe: : cafe g w6 20 0 1: d b8: : 1
Connection 'test-lab' (05abfd5e-324e-4461-844e-8501ba704773) successfully
added.
N et wo rkMan ag er will set its internal parameter i pv4 . metho d to manual and
co nnecti o n. auto co nnect to yes. N et wo rkMan ag er will also write out settings to
/etc/sysco nfi g /netwo rk-scri pts/i fcfg -my-o ffi ce where the corresponding
BOOTPROTO will be set to no ne and ONBOOT will be set to yes.
Note that manual changes to the ifcfg file will not be noticed by N et wo rkMan g er until the interface is
next brought up. See Section 1.9, “ Network Configuration Using sysconfig Files” for more information
on using configuration files.
To set two IP v4 D NS server addresses:
~]$ nmcl i co n mo d test-lab i pv4 . d ns "8. 8. 8. 8 8. 8. 4 . 4 "
Note that this will replace any previously set D NS servers. To set two IP v6 D NS server addresses:
~]$ nmcl i co n mo d test-lab i pv6 . d ns "20 0 1: 4 86 0 : 4 86 0 : : 8888
20 0 1: 4 86 0 : 4 86 0 : : 884 4 "
Note that this will replace any previously set D NS servers. Alternatively, to add additional D NS servers
to any previously set, use the + prefix as follows:
~]$ nmcl i co n mo d test-lab + i pv4 . d ns "8. 8. 8. 8 8. 8. 4 . 4 "
~]$ nmcl i co n mo d test-lab + i pv6 . d ns "20 0 1: 4 86 0 : 4 86 0 : : 8888
20 0 1: 4 86 0 : 4 86 0 : : 884 4 "
To bring up the new Ethernet connection, issue a command as follows:
~]$ nmcl i co n up test-lab i fname ens9
Connection successfully activated (D-Bus active path:
/org/freedesktop/NetworkManager/ActiveConnection/6)
Review the status of the devices and connections:
~]$ nmcl i d evi ce status
DEVICE TYPE
STATE
CONNECTION
21
Red Hat Ent erprise Linux 7 Net working G uide
ens3
ens9
lo
ethernet
ethernet
loopback
connected
connected
unmanaged
my-office
test-lab
--
To view detailed information about the newly configured connection, issue a command as follows:
~]$ nmcl i -p co n sho w test-lab
========================================================================
=======
Connection profile details (test-lab)
========================================================================
=======
connection.id:
test-lab
connection.uuid:
05abfd5e-324e-4461-844e8501ba704773
connection.interface-name:
ens9
connection.type:
802-3-ethernet
connection.autoconnect:
yes
connection.timestamp:
1410428968
connection.read-only:
no
connection.permissions:
connection.zone:
-connection.master:
-connection.slave-type:
-connection.secondaries:
connection.gateway-ping-timeout:
0
[output truncated]
The use of the -p, --pretty option adds a title banner and section breaks to the output.
Examp le 2.2. C o n f ig u rin g a St at ic Et h ern et C o n n ect io n U sin g t h e In t eract ive Ed it o r
To configure a static Ethernet connection using the interactive editor, issue commands as follows:
~]$ nmcl i co n ed i t type ethernet co n-name ens3
===| nmcli interactive connection editor |===
Adding a new '802-3-ethernet' connection
Type 'help' or '?' for available commands.
Type 'describe [>setting<.>prop<]' for detailed property description.
You may edit the following settings: connection, 802-3-ethernet
(ethernet), 802-1x, ipv4, ipv6, dcb
nmcli> set ipv4.addresses 192.168.122.88/24
Do you also want to set 'ipv4.method' to 'manual'? [yes]: yes
nmcli>
nmcli> save temporary
Saving the connection with 'autoconnect=yes'. That might result in an
immediate activation of the connection.
Do you still want to save? [yes] no
nmcli> save
Saving the connection with 'autoconnect=yes'. That might result in an
immediate activation of the connection.
22
⁠Chapt er 2 . Configure IP Net working
Do you still want to save? [yes] yes
Connection 'ens3' (704a5666-8cbd-4d89-b5f9-fa65a3dbc916) successfully
saved.
nmcli> quit
~]$
The default action is to save the connection profile as persistent. If required, the profile can be held
in memory only, until the next restart, by means of the save tempo rary command.
Lo cking a Pro file t o a Spe cific De vice
To lock a profile to a specific interface device, the commands used in the examples above include the
interface name. For example:
nmcli connection add type ethernet con-name connection-name ifname
interface-name
To make a profile usable for all compatible Ethernet interfaces, issue a command as follows:
nmcli connection add type ethernet con-name connection-name ifname "*"
Note that you have to use the i fname argument even if you do not want to set a specific interface.
Use the wildcard character * to specify that the profile can be used with any compatible device.
To lock a profile to a specific MAC address, use a command in the following format:
nmcli connection add type ethernet con-name "connection-name" ifname "*"
mac 00:00:5E:00:53:00
Adding a Wi-Fi Co nne ct io n
To view the available Wi-Fi access points, issue a command as follows:
~]$ nmcl i d ev wi fi l i st
SSID
MODE CHAN RATE
SIGNAL
FedoraTest
Infra 11
54 MB/s 98
Red Hat Guest Infra 6
54 MB/s 97
Red Hat
Infra 6
54 MB/s 77
* Red Hat
Infra 40
54 MB/s 66
VoIP
Infra 1
54 MB/s 32
MyCafe
Infra 11
54 MB/s 39
BARS SECURITY
▂▄▆█ WPA1
▂▄▆█ WPA2
▂▄▆_ WPA2 802.1X
▂▄▆_ WPA2 802.1X
▂▄__ WEP
▂▄__ WPA2
To create a Wi-Fi connection profile with static IP configuration, but allowing automatic D NS address
assignment, issue a command as follows:
~]$ nmcl i co n ad d co n-name MyCafe i fname wl an0 type wi fi ssi d MyC afe \
i p4 19 2. 16 8. 10 0 . 10 1/24 g w4 19 2. 16 8. 10 0 . 1
To set a WPA2 password, for example “ caffeine” , issue commands as follows:
~]$ nmcl i co n mo d i fy MyCafe wi fi -sec. key-mg mt wpa-psk
~]$ nmcl i co n mo d i fy MyCafe wi fi -sec. psk caffei ne
23
Red Hat Ent erprise Linux 7 Net working G uide
See the Red Hat Enterprise Linux 7 Security Guide for information on password security.
To change Wi-Fi state, issue a command in the following format:
~]$ nmcl i rad i o wi fi [o n | o ff ]
Changing a Spe cific Pro pe rt y
To check a specific property, for example mtu, issue a command as follows:
~]$ nmcl i co nnecti o n sho w i d ' MyCafe' | g rep mtu
802-11-wireless.mtu:
auto
To change the property of a setting, issue a command as follows:
~]$ nmcl i co nnecti o n mo d i fy i d ' MyCafe' 80 2-11-wi rel ess. mtu 1350
To verify the change, issue a command as follows:
~]$ nmcl i co nnecti o n sho w i d ' MyCafe' | g rep mtu
802-11-wireless.mtu:
1350
Note that N et wo rkMan ag er refers to parameters such as 80 2-3-ethernet and 80 2-11wi rel ess as the setting, and mtu as a property of the setting. See the nm-setti ng s(5) man page
for more information on properties and their settings.
2.3.3. Configuring St at ic Rout es Using nmcli
To configure static routes using the n mcli tool, the interactive editor mode must be used.
Examp le 2.3. C o n f ig u rin g St at ic R o u t es U sin g n mcli Ed it o r
To configure a static route for an Ethernet connection using the interactive editor, issue commands
as follows:
~]$ nmcl i co n ed i t type ethernet co n-name ens3
===| nmcli interactive connection editor |===
Adding a new '802-3-ethernet' connection
Type 'help' or '?' for available commands.
Type 'describe [>setting<.>prop<]' for detailed property description.
You may edit the following settings: connection, 802-3-ethernet
(ethernet), 802-1x, ipv4, ipv6, dcb
nmcli> set ipv4.routes 192.168.122.0/24 10.10.10.1
nmcli>
nmcli> save persistent
Saving the connection with 'autoconnect=yes'. That might result in an
immediate activation of the connection.
Do you still want to save? [yes] yes
24
⁠Chapt er 2 . Configure IP Net working
Connection 'ens3' (704a5666-8cbd-4d89-b5f9-fa65a3dbc916) successfully
saved.
nmcli> quit
~]$
2.4 . Using t he Command Line Int erface (CLI)
2.4 .1. Configuring a Net work Int erface Using ifcfg Files
Interface configuration files control the software interfaces for individual network devices. As the
system boots, it uses these files to determine what interfaces to bring up and how to configure them.
These files are usually named i fcfg -name, where the suffix name refers to the name of the device
that the configuration file controls. By convention, the i fcfg file's suffix is the same as the string
given by the D EVIC E directive in the configuration file itself.
St at ic Ne t wo rk Se t t ings
To configure an interface with static network settings using i fcfg files, for an interface with the
name eth0, create a file with name i fcfg -eth0 in the /etc/sysco nfi g /netwo rk-scri pts/
directory as follows:
DEVICE=eth0
BOOTPROTO=none
ONBOOT=yes
PREFIX=24
IPADDR=10.0.1.27
Optionally specify the hardware or MAC address using the HWAD D R directive. Note that this may
influence the device naming procedure as explained in Chapter 8, Consistent Network Device Naming.
You do not need to specify the network or broadcast address as this is calculated automatically by
ip calc.
Dynam ic Ne t wo rk Se t t ings
To configure an interface with dynamic network settings using i fcfg files, for an interface with name
em1, create a file with name i fcfg -em1 in the /etc/sysco nfi g /netwo rk-scri pts/ directory as
follows:
DEVICE=em1
BOOTPROTO=dhcp
ONBOOT=yes
Optionally specify the hardware or MAC address using the HWAD D R directive. Note that this may
influence the device naming procedure as explained in Chapter 8, Consistent Network Device Naming.
To configure an interface to send a different host name to the D HC P server, add the following line to
the i fcfg file.
DHCP_HOSTNAME=hostname
To configure an interface to ignore routes sent by a D HC P server, add the following line to the i fcfg
file.
25
Red Hat Ent erprise Linux 7 Net working G uide
PEERDNS=no
This will prevent network service from updating /etc/reso l v. co nf with the D NS servers received
from a D HC P server.
To configure an interface to use particular D NS servers, set P EER D NS= no as described above and
add lines as follows to the i fcfg file:
DNS1=ip-address
DNS2=ip-address
where ip-address is the address of a D NS server. This will cause the network service to update
/etc/reso l v. co nf with the D NS servers specified.
N et wo rkMan ag er will by default call the D HC P client, d h clien t , when a profile has been set to
obtain addresses automatically, or when an interface configuration file has BOOTPROTO set to
d hcp. Where D HC P is required, an instance of d hcl i ent is started for every Internet protocol, IP v4
and IP v6 , on an interface. Where N et wo rkMan ag er is not running, or not managing an interface,
then the legacy network service will call instances of d hcl i ent as required.
Co nfiguring a DHCP Clie nt
2.4 .2. Configuring a Net work Int erface Using ip Commands
The ip utility can be used to assign IP addresses to an interface. The command takes the following
form:
ip addr [ add | del ] address dev ifname
Assigning a St at ic Addre ss Using ip Co m m ands
To assign an IP address to an interface, issue a command as ro o t as follows:
~]# i p ad d ress ad d 10 . 0 . 0 . 3/24 d ev eth0
The address assignment of a specific device can be viewed as follows:
~]# i p ad d r sho w d ev eth0
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast
state UP qlen 1000
link/ether f0:de:f1:7b:6e:5f brd ff:ff:ff:ff:ff:ff
inet 10.0.0.3/24 brd 10.0.0.255 scope global global eth0
valid_lft 58682sec preferred_lft 58682sec
inet6 fe80::f2de:f1ff:fe7b:6e5f/64 scope link
valid_lft forever preferred_lft forever
Further examples and command options can be found in the i p-ad d ress(8) manual page.
Co nfiguring Mult iple Addre sse s Using ip Co m m ands
As the ip utility supports assigning multiple addresses to the same interface it is no longer necessary
to use the alias interface method of binding multiple addresses to the same interface. The ip
command to assign an address can be repeated multiple times in order to assign multiple address.
For example:
26
⁠Chapt er 2 . Configure IP Net working
~]# i p ad d ress ad d 19 2. 16 8. 2. 223/24 d ev eth1
~]# i p ad d ress ad d 19 2. 16 8. 4 . 223/24 d ev eth1
~]# i p ad d r
3: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast
state UP qlen 1000
link/ether 52:54:00:fb:77:9e brd ff:ff:ff:ff:ff:ff
inet 192.168.2.223/24 scope global eth1
inet 192.168.4 .223/24 scope global eth1
The commands for the ip utility are documented in the i p(8) manual page.
Note
ip commands given on the command line will not persist after a system restart.
2.4 .3. St at ic Rout es and t he Default Gat eway
Static routes are for traffic that must not, or should not, go through the default gateway. Routing is
often handled by devices on the network dedicated to routing (although any device can be
configured to perform routing). Therefore, it is often not necessary to configure static routes on
Red Hat Enterprise Linux servers or clients. Exceptions include traffic that must pass through an
encrypted VPN tunnel or traffic that should take a specific route for reasons of cost or security. The
default gateway is for any and all traffic which is not destined for the local network and for which no
preferred route is specified in the routing table. The default gateway is traditionally a dedicated
network router.
Co nfiguring St at ic Ro ut e s Using t he Co m m and Line
If static routes are required, they can be added to the routing table by means of the i p ro ute ad d
command and removed using the i p ro ute d el command. The more frequently used i p ro ute
commands take the following form:
ip route [ add | del | change | append | replace ] destination-address
See the i p-ro ute(8) man page for more details on the options and formats.
Use the i p ro ute command without options to display the IP routing table. For example:
~]$ ip route
default via 192.168.122.1 dev ens9 proto static metric 1024
192.168.122.0/24 dev ens9 proto kernel scope link src 192.168.122.107
192.168.122.0/24 dev eth0 proto kernel scope link src 192.168.122.126
To add a static route to a host address, in other words to a single IP address, issue a command as
ro o t:
ip route add 192.0.2.1 via 10.0.0.1 [d ev ifname]
Where 192.0.2.1 is the IP address of the host in dotted decimal notation, 10.0.0.1 is the next hop
address and ifname is the exit interface leading to the next hop.
27
Red Hat Ent erprise Linux 7 Net working G uide
To add a static route to a network, in other words to an IP address representing a range of IP
addresses, issue the following command as ro o t:
ip route add 192.0.2.0/24 via 10.0.0.1 [d ev ifname]
where 192.0.2.0 is the IP address of the destination network in dotted decimal notation and /24 is the
network prefix. The network prefix is the number of enabled bits in the subnet mask. This format of
network address slash network prefix length is sometimes referred to as classless inter-domain routing
(CID R) notation.
Static route configuration can be stored per-interface in a /etc/sysco nfi g /netwo rkscri pts/ro ute-interface file. For example, static routes for the eth0 interface would be stored in
the /etc/sysco nfi g /netwo rk-scri pts/ro ute-eth0 file. The ro ute-interface file has two
formats: ip command arguments and network/netmask directives. These are described below.
See the i p-ro ute(8) man page for more information on the i p ro ute command.
Co nfiguring T he De fault Gat e way
The default gateway is determined by the network scripts which parse the
/etc/sysco nfi g /netwo rk file first and then the network interface i fcfg files for interfaces that
are “ up” . The i fcfg files are parsed in numerically ascending order, and the last GATEWAY
directive to be read is used to compose a default route in the routing table.
The default route can thus be indicated by means of the GATEWAY directive and can be specified
either globally or in interface-specific configuration files. Specifying the gateway globally has certain
advantages in static networking environments, especially if more than one network interface is
present. It can make fault finding simpler if applied consistently.
In dynamic network environments, where mobile hosts are managed by N et wo rkMan ag er, gateway
information is likely to be interface specific and is best left to be assigned by D HC P . In special cases
where it is necessary to influence N et wo rkMan ag er's selection of the exit interface to be used to
reach a gateway, make use of the D EFR O UT E= no command in the i fcfg files for those interfaces
which do not lead to the default gateway.
Global default gateway configuration is stored in the /etc/sysco nfi g /netwo rk file. This file
specifies gateway and host information for all network interfaces. .
2.4 .4 . Configuring St at ic Rout es in ifcfg files
Static routes set using ip commands at the command prompt will be lost if the system is shutdown or
restarted. To configure static routes to be persistent after a system restart, they must be placed in perinterface configuration files in the /etc/sysco nfi g /netwo rk-scri pts/ directory. The file name
should be of the format ro ute-ifname. There are two types of commands to use in the configuration
files; ip commands as explained in Section 2.4.4.1, “ Static Routes Using the IP Command Arguments
Format” and the Network/Netmask format as explained in Section 2.4.4.2, “ Network/Netmask
D irectives Format” .
2 .4 .4 .1 . St at ic Ro ut e s Using t he IP Co m m and Argum e nt s Fo rm at
If required in a per-interface configuration file, for example /etc/sysco nfi g /netwo rkscri pts/ro ute-eth0 , define a route to a default gateway on the first line. This is only required if
the gateway is not set via D HC P and is not set globally in the /etc/sysco nfi g /netwo rk file:
default via 192.168.1.1 d ev interface
28
⁠Chapt er 2 . Configure IP Net working
where 192.168.1.1 is the IP address of the default gateway. The interface is the interface that is
connected to, or can reach, the default gateway. The d ev option can be omitted, it is optional. Note
that this setting takes precedence over a setting in the /etc/sysco nfi g /netwo rk file.
If a route to a remote network is required, a static route can be specified as follows. Each line is
parsed as an individual route:
10.10.10.0/24 via 192.168.1.1 [d ev interface]
where 10.10.10.0/24 is the network address and prefix length of the remote or destination network. The
address 192.168.1.1 is the IP address leading to the remote network. It is preferably the next hop
address but the address of the exit interface will work. The “ next hop” means the remote end of a link,
for example a gateway or router. The d ev option can be used to specify the exit interface interface but
it is not required. Add as many static routes as required.
The following is an example of a ro ute-interface file using the ip command arguments format.
The default gateway is 19 2. 16 8. 0 . 1, interface eth0 and a leased line or WAN connection is
available at 19 2. 16 8. 0 . 10 . The two static routes are for reaching the 10 . 10 . 10 . 0 /24 network
and the 172. 16 . 1. 10 /32 host:
default via 192.168.0.1 dev eth0
10.10.10.0/24 via 192.168.0.10 dev eth0
172.16.1.10/32 via 192.168.0.10 dev eth0
In the above example, packets going to the local 19 2. 16 8. 0 . 0 /24 network will be directed out the
interface attached to that network. Packets going to the 10 . 10 . 10 . 0 /24 network and
172. 16 . 1. 10 /32 host will be directed to 19 2. 16 8. 0 . 10 . Packets to unknown, remote, networks
will use the default gateway therefore static routes should only be configured for remote networks or
hosts if the default route is not suitable. Remote in this context means any networks or hosts that are
not directly attached to the system.
Specifying an exit interface is optional. It can be useful if you want to force traffic out of a specific
interface. For example, in the case of a VPN, you can force traffic to a remote network to pass through
a tun0 interface even when the interface is in a different subnet to the destination network.
Important
If the default gateway is already assigned by D HC P and if the same gateway with the same
metric is specified in a configuration file, an error during start-up, or when bringing up an
interface, will occur. The follow error message may be shown: " RTNETLINK answers: File
exists" . This error may be ignored.
2 .4 .4 .2 . Ne t wo rk/Ne t m ask Dire ct ive s Fo rm at
You can also use the network/netmask directives format for ro ute-interface files. The following is
a template for the network/netmask format, with instructions following afterwards:
ADDRESS0=10.10.10.0
NETMASK0=255.255.255.0
GATEWAY0=192.168.1.1
AD D R ESS0 = 10.10.10.0 is the network address of the remote network or host to be reached.
29
Red Hat Ent erprise Linux 7 Net working G uide
NET MASK0 = 255.255.255.0 is the netmask for the network address defined with
AD D R ESS0 = 10.10.10.0.
G AT EWAY 0 = 192.168.1.1 is the default gateway, or an IP address that can be used to reach
AD D R ESS0 = 10.10.10.0
The following is an example of a ro ute-interface file using the network/netmask directives format.
The default gateway is 19 2. 16 8. 0 . 1 but a leased line or WAN connection is available at
19 2. 16 8. 0 . 10 . The two static routes are for reaching the 10 . 10 . 10 . 0 /24 and
172. 16 . 1. 0 /24 networks:
ADDRESS0=10.10.10.0
NETMASK0=255.255.255.0
GATEWAY0=192.168.0.10
ADDRESS1=172.16.1.10
NETMASK1=255.255.255.0
GATEWAY1=192.168.0.10
Subsequent static routes must be numbered sequentially, and must not skip any values. For
example, AD D R ESS0 , AD D R ESS1, AD D R ESS2, and so on.
2.4 .5. Configuring a VPN
IPsec, provided by Lib reswan , is the preferred method for creating a VPN in Red Hat
Enterprise Linux 7. Configuring an IPsec VPN using the command line is documented in the Red Hat
Enterprise Linux 7 Security Guide.
2.5. Using Net workManager wit h t he GNOME Graphical User Int erface
In Red Hat Enterprise Linux 7, N et wo rkMan ag er does not have its own graphical user interface
(GUI). The network connection icon on the top right of the desktop is provided as part of the GNOME
Shell and the Netwo rk settings configuration tool is provided as part of the new GNOME co n t ro lcen t er GUI. The old n m- co n n ect io n - ed it o r GUI is still available for certain tasks.
2.5.1. Connect ing t o a Net work Using a GUI
There are two ways to access the Netwo rk settings window of the co n t ro l- cen t er application:
Press the Super key to enter the Activities Overview, type co ntro l netwo rk, as seen in
Figure 2.2, “ The Network utility being selected in GNOME” and then press Enter. The Netwo rk
settings tool appears. Proceed to Section 2.5.2, “ Configuring New and Editing Existing
Connections” .
30
⁠Chapt er 2 . Configure IP Net working
Fig u re 2.2. T h e N et wo rk u t ilit y b ein g select ed in G N O ME
Click on the GNOME Shell network connection icon in the top right-hand corner of the screen to
open its menu.
When you click on the GNOME Shell network connection icon, you are presented with:
a list of categorized networks you are currently connected to (such as Wi red and Wi -Fi );
a list of all Avai l abl e Netwo rks that N et wo rkMan ag er has detected;
options for connecting to any configured Virtual Private Networks (VPNs); and,
an option for selecting the Netwo rk Setti ng s menu entry.
If you are connected to a network, this is indicated by the symbolic O N button. Clicking anywhere on
the level of the button will toggle the state of the button. If you change the button from ON to OFF you
will disconnect that network connection.
Click Netwo rk Setti ng s. The Netwo rk settings tool appears. Proceed to Section 2.5.2,
“ Configuring New and Editing Existing Connections” .
31
Red Hat Ent erprise Linux 7 Net working G uide
Fig u re 2.3. T h e G N O ME n et wo rk men u , sh o win g all availab le an d co n n ect ed - t o
n et wo rks
2.5.2. Configuring New and Edit ing Exist ing Connect ions
The Netwo rk settings window shows the connection status, its type and interface, its IP address
and routing details, and so on.
Fig u re 2.4 . C o n f ig u re N et wo rks U sin g t h e N et wo rk Set t in g s Win d o w
The Netwo rk settings window has a menu on the left-hand side showing the available network
devices or interfaces. This includes software interfaces such as for VLANs, bridges, bonds, and
32
⁠Chapt er 2 . Configure IP Net working
teams. On the right-hand side, the connection profiles are shown for the selected network device or
interface. A profile is a named collection of settings that can be applied to an interface. Below that is
a plus and a minus button for adding and deleting new network connections, and on the right a gear
wheel icon will appear for editing the connection details of the selected network device or VPN
connection. To add a new connection, click the plus symbol to open the Ad d Netwo rk
C o nnecti o n window and proceed to Section 2.5.2.1, “ Configuring a New Connection” .
Edit ing an Exist ing Co nne ct io n
Clicking on the gear wheel icon of an existing connection profile in the Netwo rk settings window
opens the Netwo rk details window, from where you can perform most network configuration tasks
such as IP addressing, D NS, and routing configuration.
Fig u re 2.5. C o n f ig u re N et wo rks U sin g t h e N et wo rk C o n n ect io n D et ails Win d o w
2 .5 .2 .1 . Co nfiguring a Ne w Co nne ct io n
In the Netwo rk settings window, click the plus sign below the menu to open the Ad d Netwo rk
C o nnecti o n window. This displays a list of connection types that can be added.
Then, to configure:
VPN co n n ect io n s, click the VP N entry and proceed to Section 2.5.7, “ Establishing a VPN
Connection” ;
B o n d co n n ect io n s, click the Bo nd entry and proceed to Section 4.6.1, “ Establishing a Bond
Connection” ;
B rid g e co n n ect io n s, click the Bri d g e entry and proceed to Section 6.4.1, “ Establishing a
Bridge Connection” ;
VLAN co n n ect io n s, click the VLAN entry and proceed to Section 7.5.1, “ Establishing a VLAN
Connection” ; or,
T eam co n n ect io n s, click the T eam entry and proceed to Section 5.13, “ Creating a Network
Team Using a GUI” .
33
Red Hat Ent erprise Linux 7 Net working G uide
2.5.3. Connect ing t o a Net work Aut omat ically
For any connection type you add or configure, you can choose whether you want
N et wo rkMan ag er to try to connect to that network automatically when it is available.
Pro ced u re 2.1. C o n f ig u rin g N et wo rkMan ag er t o C o n n ect t o a N et wo rk Au t o mat ically
Wh en D et ect ed
1. Press the Super key to enter the Activities Overview, type co ntro l netwo rk and then press
Enter. The Netwo rk settings tool appears.
2. Select the network interface from the left-hand-side menu.
3. Click on the gear wheel icon of a connection profile on the right-hand side menu. If you have
only one profile associated with the selected interface the gear wheel icon will be in the lower
right-hand-side corner. The Netwo rk details window appears.
4. Select the Id enti ty menu entry on the left. The Netwo rk window changes to the identity
view.
5. Select C o nnect auto mati cal l y to cause N et wo rkMan ag er to auto-connect to the
connection whenever N et wo rkMan ag er detects that it is available. Clear the check box if
you do not want N et wo rkMan ag er to connect automatically. If the check box is clear, you
will have to select that connection manually in the network connection icon's menu to cause it
to connect.
2.5.4 . Syst em-wide and Privat e Connect ion Profiles
N et wo rkMan ag er stores all connection profiles. A profile is a named collection of settings that can be
applied to an interface. N et wo rkMan ag er stores these connection profiles for system-wide use
(system connections), as well as all user connection profiles. Access to the connection profiles is
controlled by permissions which are stored by N et wo rkMan ag er. See the nm-setti ng s(5) man
page for more information on the co nnecti o n settings permi ssi o ns property. The permissions
correspond to the USER S directive in the i fcfg files. If the USER S directive is not present, the
network profile will be available to all users. As an example, the following command in an i fcfg file
will make the connection available only to the users listed:
USERS="joe bob alice"
This can also be set using graphical user interface tools. In n m- co n n ect io n - ed it o r, there is the
corresponding Al l users may co nnect to thi s netwo rk check box on the G eneral tab,
and in the GNOME co n t ro l- cen t er Network settings Identity window, there is the Make avai l abl e
to o ther users check box.
N et wo rkMan ag er's default policy is to allow all users to create and modify system-wide
connections. Profiles that should be available at boot time cannot be private because they will not be
visible until the user logs in. For example, if user user creates a connection profile user-em2 with
the C o nnect Auto mati cal l y check box selected but with the Make avai l abl e to o ther
users not selected, then the connection will not be available at boot time.
To restrict connections and networking, there are two options which can be used alone or in
combination:
Clear the Make avai l abl e to o ther users check box, which changes the connection to be
modifiable and usable only by the user doing the changing.
34
⁠Chapt er 2 . Configure IP Net working
Use the p o lkit framework to restrict permissions of general network operations on a per-user
basis.
The combination of these two options provides fine-grained security and control over networking.
See the po l ki t(8) man page for more information on p o lkit .
Note that VPN connections are always created as private-per-user, since they are assumed to be
more private than a Wi-Fi or Ethernet connection.
Pro ced u re 2.2. C h an g in g a C o n n ect io n t o B e U ser- sp ecif ic In st ead o f Syst em- Wid e, o r
Vice Versa
D epending on the system's policy, you may need root privileges on the system in order to change
whether a connection is user-specific or system-wide.
1. Press the Super key to enter the Activities Overview, type co ntro l netwo rk and then press
Enter. The Netwo rk settings tool appears.
2. Select the network interface from the left-hand-side menu.
3. Click on the gear wheel icon of a connection profile on the right-hand side menu. If you have
only one profile associated with the selected interface the gear wheel icon will be in the lower
right-hand-side corner. The Netwo rk details window appears.
4. Select the Id enti ty menu entry on the left. The Netwo rk window changes to the identity
view.
5. Select the Make avai l abl e to o ther users check box to cause N et wo rkMan ag er to
make the connection available system-wide.
Conversely, clear the Make avai l abl e to o ther users check box to make the
connection user-specific.
2.5.5. Configuring a Wired (Et hernet ) Connect ion
To configure a wired network connection, press the Super key to enter the Activities Overview, type
co ntro l netwo rk and then press Enter. The Netwo rk settings tool appears.
Select the Wi red network interface from the left-hand-side menu if it is not already highlighted.
The system creates and configures a single wired connection profile called Wi red by default. A profile
is a named collection of settings that can be applied to an interface. More than one profile can be
created for an interface and applied as needed. The default profile cannot be deleted but its settings
can be changed. You can edit the default Wi red profile by clicking the gear wheel icon. You can
create a new wired connection profile by clicking the Ad d P ro fi l e button. Connection profiles
associated with a selected interface are shown on the right-hand side menu.
When you add a new connection by clicking the Ad d P ro fi l e button, N et wo rkMan ag er creates
a new configuration file for that connection and then opens the same dialog that is used for editing
an existing connection. The difference between these dialogs is that an existing connection profile
has a D etai l s and R eset menu entry. In effect, you are always editing a connection profile; the
difference only lies in whether that connection previously existed or was just created by
N et wo rkMan ag er when you clicked Ad d P ro fi l e.
2 .5 .5 .1 . Co nfiguring t he Co nne ct io n Nam e , Aut o -Co nne ct Be havio r, and
Availabilit y Se t t ings
35
Red Hat Ent erprise Linux 7 Net working G uide
Many settings in the Ed i ti ng dialog are common to all connection types, see the Id enti ty view
(or the G eneral tab if using n m- co n n ect io n - ed it o r):
Name — Enter a descriptive name for your network connection. This name will be used to list this
connection in the menu of the Netwo rk window.
MAC Ad d ress — Select the MAC address of the interface this profile must be applied to.
C l o ned Ad d ress — If required, enter a different MAC address to use.
MT U — If required, enter a specific maximum transmission unit (MTU) to use. The MTU value
represents the size in bytes of the largest packet that the link-layer will transmit. This value
defaults to 150 0 and does not generally need to be specified or changed.
Fi rewal l Zo ne — If required, select a different firewall zone to apply. See the Red Hat
Enterprise Linux 7 Security Guide for more information on firewall zones.
C o nnect Auto mati cal l y — Select this box if you want N et wo rkMan ag er to auto-connect to
this connection when it is available. See Section 2.5.3, “ Connecting to a Network Automatically”
for more information.
Make avai l abl e to o ther users — Select this box to create a connection available to all
users on the system. Changing this setting may require root privileges. See Section 2.5.4,
“ System-wide and Private Connection Profiles” for details.
Auto mati cal l y co nnect to VP N when usi ng thi s co nnecti o n — Select this box if
you want N et wo rkMan ag er to auto-connect to the selected VPN connection when this
connection profile is connected. Select the VPN from the drop-down menu.
Savin g Yo u r N ew ( o r Mo d if ied ) C o n n ect io n an d Makin g Fu rt h er C o n f ig u rat io n s
Once you have finished editing your wired connection, click the Appl y button to save your
customized configuration. If the profile was in use while being edited, power cycle the connection to
make N et wo rkMan ag er apply the changes. If the profile is OFF, set it to ON or select it in the
network connection icon's menu. See Section 2.5.1, “ Connecting to a Network Using a GUI” for
information on using your new or altered connection.
You can further configure an existing connection by selecting it in the Netwo rk window and clicking
the gear wheel icon to return to the editing dialog.
Then, to configure:
p o rt - b ased N et wo rk Access C o n t ro l ( PN AC ) , click the 80 2. 1X Securi ty tab and proceed
to Section 2.5.10.1, “ Configuring 802.1X Security” ;
IP v4 settings for the connection, click the IP v4 Setti ng s tab and proceed to Section 2.5.10.4,
“ Configuring IPv4 Settings” ; or,
IP v6 settings for the connection, click the IP v6 Setti ng s tab and proceed to Section 2.5.10.5,
“ Configuring IPv6 Settings” .
2.5.6. Configuring a Wi-Fi Connect ion
This section explains how to use N et wo rkMan ag er to configure a Wi-Fi (also known as wireless or
802.11a/b/g/n) connection to an Access Point.
To configure a mobile broadband (such as 3G) connection, see Section 2.5.8, “ Establishing a
Mobile Broadband Connection” .
36
⁠Chapt er 2 . Configure IP Net working
Quickly Co nne ct ing t o an Available Acce ss Po int
The easiest way to connect to an available access point is to click on the network connection icon to
activate the network connection icon's menu, locate the Service Set Identifier (SSID ) of the access
point in the list of Wi -Fi networks, and click on it. A padlock symbol indicates the access point
requires authentication. If the access point is secured, a dialog prompts you for an authentication
key or password.
N et wo rkMan ag er tries to auto-detect the type of security used by the access point. If there are
multiple possibilities, N et wo rkMan ag er guesses the security type and presents it in the Wi -Fi
securi ty drop-down menu. For WPA-PSK security (WPA with a passphrase) no choice is
necessary. For WPA Enterprise (802.1X) you have to specifically select the security, because that
cannot be auto-detected. If you are unsure, try connecting to each type in turn. Finally, enter the key
or passphrase in the P asswo rd field. Certain password types, such as a 40-bit WEP or 128-bit WPA
key, are invalid unless they are of a requisite length. The C o nnect button will remain inactive until
you enter a key of the length required for the selected security type. To learn more about wireless
security, see Section 2.5.10.2, “ Configuring Wi-Fi Security” .
If N et wo rkMan ag er connects to the access point successfully, the network connection icon will
change into a graphical indicator of the wireless connection's signal strength.
You can also edit the settings for one of these auto-created access point connections just as if you
had added it yourself. The Wi -Fi page of the Netwo rk window has a Hi sto ry button. Clicking it
reveals a list of all the connections you have ever tried to connect to. See Section 2.5.6.2, “ Editing a
Connection, or Creating a Completely New One”
2 .5 .6 .1 . Co nne ct ing t o a Hidde n Wi-Fi Ne t wo rk
All access points have a Service Set Identifier (SSID ) to identify them. However, an access point may
be configured not to broadcast its SSID , in which case it is hidden, and will not show up in
N et wo rkMan ag er's list of Avai l abl e networks. You can still connect to a wireless access point
that is hiding its SSID as long as you know its SSID , authentication method, and secrets.
To connect to a hidden wireless network, press the Super key to enter the Activities Overview, type
co ntro l netwo rk and then press Enter. The Netwo rk window appears. Select Wi -Fi from the
menu and then select C o nnect to Hi d d en Netwo rk to cause a dialog to appear. If you have
connected to the hidden network before, use the C o nnecti o n drop-down to select it, and click
C o nnect. If you have not, leave the C o nnecti o n drop-down as N ew, enter the SSID of the hidden
network, select its Wi -Fi securi ty method, enter the correct authentication secrets, and click
C o nnect.
For more information on wireless security settings, see Section 2.5.10.2, “ Configuring Wi-Fi Security” .
2 .5 .6 .2 . Edit ing a Co nne ct io n, o r Cre at ing a Co m ple t e ly Ne w One
You can edit an existing connection that you have tried or succeeded in connecting to in the past by
opening the Wi -Fi page of the Netwo rk dialog and selecting the gear wheel icon to the right of the
Wi-Fi connection name. If the network is not currently in range, click Hi sto ry to display past
connections. When you click the gear wheel icon the editing connection dialog appears. The
D etai l s window shows the connection details.
To configure a new connection whose SSID is in range, first attempt to connect to it by opening the
Netwo rk window, selecting the Wi -Fi menu entry, and clicking the connection name (by default, the
same as the SSID ). If the SSID is not in range, see Section 2.5.6.1, “ Connecting to a Hidden Wi-Fi
Network” . If the SSID is in range, the procedure is as follows:
37
Red Hat Ent erprise Linux 7 Net working G uide
1. Press the Super key to enter the Activities Overview, type co ntro l netwo rk and then press
Enter. The Netwo rk settings tool appears.
2. Select the Wi -Fi interface from the left-hand-side menu entry.
3. Click the Wi-Fi connection profile on the right-hand side menu you want to connect to. A
padlock symbol indicates a key or password is required.
4. If requested, enter the authentication details.
C o n f ig u rin g t h e SSID , Au t o - C o n n ect B eh avio r, an d Availab ilit y Set t in g s
To edit a Wi-Fi connection's settings, select Wi -Fi in the Netwo rk page and then select the gear
wheel icon to the right of the Wi-Fi connection name. Select Id enti ty. The following settings are
available:
SSID
The Service Set Identifier (SSID ) of the access point (AP).
BSSID
The Basic Service Set Identifier (BSSID ) is the MAC address, also known as a hardware
address, of the specific wireless access point you are connecting to when in
Infrastructure mode. This field is blank by default, and you are able to connect to a
wireless access point by SSID without having to specify its BSSID . If the BSSID is
specified, it will force the system to associate to a specific access point only.
For ad-hoc networks, the BSSID is generated randomly by the mac80211 subsystem when
the ad-hoc network is created. It is not displayed by N et wo rkMan ag er
MAC ad d ress
Select the MAC address, also known as a hardware address, of the Wi-Fi interface to use.
A single system could have one or more wireless network adapters connected to it. The MAC
ad d ress field therefore allows you to associate a specific wireless adapter with a specific
connection (or connections).
C l o ned Ad d ress
A cloned MAC address to use in place of the real hardware address. Leave blank unless
required.
The following settings are common to all connection profiles:
C o nnect auto mati cal l y — Select this box if you want N et wo rkMan ag er to auto-connect to
this connection when it is available. See Section 2.5.3, “ Connecting to a Network Automatically”
for more information.
Make avai l abl e to o ther users — Select this box to create a connection available to all
users on the system. Changing this setting may require root privileges. See Section 2.5.4,
“ System-wide and Private Connection Profiles” for details.
Savin g Yo u r N ew ( o r Mo d if ied ) C o n n ect io n an d Makin g Fu rt h er C o n f ig u rat io n s
Once you have finished editing the wireless connection, click the Appl y button to save your
configuration. Given a correct configuration, you can connect to your modified connection by
selecting it from the network connection icon's menu. See Section 2.5.1, “ Connecting to a Network
Using a GUI” for details on selecting and connecting to a network.
38
⁠Chapt er 2 . Configure IP Net working
You can further configure an existing connection by selecting it in the Netwo rk window and clicking
the gear wheel icon to reveal the connection details.
Then, to configure:
secu rit y au t h en t icat io n for the wireless connection, click Securi ty and proceed to
Section 2.5.10.2, “ Configuring Wi-Fi Security” ;
IP v4 settings for the connection, click IP v4 and proceed to Section 2.5.10.4, “ Configuring IPv4
Settings” ; or,
IP v6 settings for the connection, click IP v6 and proceed to Section 2.5.10.5, “ Configuring IPv6
Settings” .
2.5.7. Est ablishing a VPN Connect ion
IPsec, provided by Lib reswan , is the preferred method for creating a VPN in Red Hat
Enterprise Linux 7. The GNOME graphical user interface tool described below requires the
NetworkManager-libreswan-gnome package. If required, to ensure this package is installed issue the
following command as ro o t:
~]# yum i nstal l Netwo rkManag er-l i breswan-g no me
See Red Hat Enterprise Linux 7 System Administrator's Guide for more information on how to install new
packages in Red Hat Enterprise Linux 7.
Establishing a Virtual Private Network (VPN) enables communication between your Local Area
Network (LAN), and another, remote LAN. This is done by setting up a tunnel across an intermediate
network such as the Internet. The VPN tunnel that is set up typically uses authentication and
encryption. After successfully establishing a VPN connection using a secure tunnel, a VPN router or
gateway performs the following actions upon the packets you transmit:
1. it adds an Authentication Header for routing and authentication purposes;
2. it encrypts the packet data; and,
3. it encloses the data in packets according to the Encapsulating Security Payload (ESP)
protocol, which constitutes the decryption and handling instructions.
The receiving VPN router strips the header information, decrypts the data, and routes it to its intended
destination (either a workstation or other node on a network). Using a network-to-network
connection, the receiving node on the local network receives the packets already decrypted and
ready for processing. The encryption and decryption process in a network-to-network VPN
connection is therefore transparent to clients.
Because they employ several layers of authentication and encryption, VPNs are a secure and
effective means of connecting multiple remote nodes to act as a unified intranet.
Pro ced u re 2.3. Ad d in g a N ew VPN C o n n ect io n
You can configure a new VPN connection by opening the Netwo rk window and selecting the plus
symbol below the menu.
1. Press the Super key to enter the Activities Overview, type co ntro l netwo rk and then press
Enter. The Netwo rk settings tool appears.
2. Select the plus symbol below the menu. The Ad d Netwo rk C o nnecti o n window appears.
39
Red Hat Ent erprise Linux 7 Net working G uide
3. Select the VP N menu entry. The view now changes to offer configuring a VPN manually, or
importing a VPN configuration file.
The appropriate N et wo rkMan ag er VPN plug-in for the VPN type you want to configure must
be installed. See Section 2.5.7, “ Establishing a VPN Connection” .
4. Click the Ad d button to open the C ho o se a VP N C o nnecti o n T ype assistant.
5. Select the VPN protocol for the gateway you are connecting to from the menu. The VPN
protocols available for selection in the menu correspond to the N et wo rkMan ag er VPN plugins installed. See Section 2.5.7, “ Establishing a VPN Connection” .
6. The Ad d Netwo rk C o nnecti o n window changes to present the settings customized for the
type of VPN connection you selected in the previous step.
Pro ced u re 2.4 . Ed it in g an Exist in g VPN C o n n ect io n
You can configure an existing VPN connection by opening the Netwo rk window and selecting the
name of the connection from the list. Then click the Ed i t button.
1. Press the Super key to enter the Activities Overview, type co ntro l netwo rk and then press
Enter. The Netwo rk settings tool appears.
2. Select the VP N connection you want to edit from the left hand menu.
3. Click the C o nfi g ure button.
Co nfiguring t he Co nne ct io n Nam e , Aut o -Co nne ct Be havio r, and Availabilit y
Se t t ings
Five settings in the Ed i ti ng dialog are common to all connection types, see the G eneral tab:
C o nnecti o n name — Enter a descriptive name for your network connection. This name will be
used to list this connection in the menu of the Netwo rk window.
Auto mati cal l y co nnect to thi s netwo rk when i t i s avai l abl e — Select this box if
you want N et wo rkMan ag er to auto-connect to this connection when it is available. See
Section 2.5.3, “ Connecting to a Network Automatically” for more information.
Al l users may co nnect to thi s netwo rk — Select this box to create a connection
available to all users on the system. Changing this setting may require root privileges. See
Section 2.5.4, “ System-wide and Private Connection Profiles” for details.
Auto mati cal l y co nnect to VP N when usi ng thi s co nnecti o n — Select this box if
you want N et wo rkMan ag er to auto-connect to a VPN connection when it is available. Select the
VPN from the drop-down menu.
Fi rewal l Zo ne — Select the Firewall Z one from the drop-down menu. See the Red Hat
Enterprise Linux 7 Security Guide for more information on Firewall Z ones.
Co nfiguring t he VPN T ab
G ateway
The name or IP address of the remote VPN gateway.
G ro up name
The name of a VPN group configured on the remote gateway.
40
⁠Chapt er 2 . Configure IP Net working
User passwo rd
If required, enter the password used to authenticate with the VPN.
G ro up passwo rd
If required, enter the password used to authenticate with the VPN.
User name
If required, enter the user name used to authenticate with the VPN.
P hase1 Al g o ri thms
If required, enter the algorithms to be used to authenticate and set up an encrypted
channel.
P hase2 Al g o ri thms
If required, enter the algorithms to be used for the IPsec negotiations.
D o mai n
If required, enter the D omain Name.
Saving Yo ur Ne w (o r Mo difie d) Co nne ct io n and Making Furt he r Co nfigurat io ns
Once you have finished editing your new VPN connection, click the Save button to save your
customized configuration. If the profile was in use while being edited, power cycle the connection to
make N et wo rkMan ag er apply the changes. If the profile is OFF, set it to ON or select it in the
network connection icon's menu. See Section 2.5.1, “ Connecting to a Network Using a GUI” for
information on using your new or altered connection.
You can further configure an existing connection by selecting it in the Netwo rk window and clicking
C o nfi g ure to return to the Ed i ti ng dialog.
Then, to configure:
IP v4 settings for the connection, click the IP v4 Setti ng s tab and proceed to Section 2.5.10.4,
“ Configuring IPv4 Settings” .
2.5.8. Est ablishing a Mobile Broadband Connect ion
You can use N et wo rkMan ag er's mobile broadband connection abilities to connect to the following
2G and 3G services:
2G — GPRS (General Packet Radio Service), EDGE (Enhanced Data Rates for GSM Evolution), or
CD MA (Code D ivision Multiple Access).
3G — UMTS (Universal Mobile Telecommunications System), HSPA (High Speed Packet Access), or
EVD O (EVolution D ata-Only).
Your computer must have a mobile broadband device (modem), which the system has discovered
and recognized, in order to create the connection. Such a device may be built into your computer (as
is the case on many notebooks and netbooks), or may be provided separately as internal or external
hardware. Examples include PC card, USB Modem or D ongle, mobile or cellular telephone capable
of acting as a modem.
Pro ced u re 2.5. Ad d in g a N ew Mo b ile B ro ad b an d C o n n ect io n
41
Red Hat Ent erprise Linux 7 Net working G uide
You can configure a mobile broadband connection by opening the Netwo rk C o nnecti o ns tool
and selecting the Mo bi l e Bro ad band tab.
1. Press the Super key to enter the Activities Overview, type nm-co nnecti o n-ed i to r and
then press Enter. The Netwo rk C o nnecti o ns tool appears.
2. Click the Ad d button. The C ho o se a C o nnecti o n T ype menu opens.
3. Select the Mo b ile B ro ad b an d menu entry.
4. Click C reate to open the Set up a Mo bi l e Bro ad band C o nnecti o n assistant.
5. Under C reate a co nnecti o n fo r thi s mo bi l e bro ad band d evi ce, choose the
2G- or 3G-capable device you want to use with the connection. If the drop-down menu is
inactive, this indicates that the system was unable to detect a device capable of mobile
broadband. In this case, click C ancel , ensure that you do have a mobile broadbandcapable device attached and recognized by the computer and then retry this procedure. Click
the C o nti nue button.
6. Select the country where your service provider is located from the list and click the C o nti nue
button.
7. Select your provider from the list or enter it manually. Click the C o nti nue button.
8. Select your payment plan from the drop-down menu and confirm the Access Point Name (APN)
is correct. Click the C o nti nue button.
9. Review and confirm the settings and then click the Appl y button.
10. Edit the mobile broadband-specific settings by referring to Section 2.5.8.1, “ Configuring the
Mobile Broadband Tab” .
Pro ced u re 2.6 . Ed it in g an Exist in g Mo b ile B ro ad b an d C o n n ect io n
Follow these steps to edit an existing mobile broadband connection.
1. Press the Super key to enter the Activities Overview, type nm-co nnecti o n-ed i to r and
then press Enter. The Netwo rk C o nnecti o ns tool appears.
2. Select the Mo bi l e Bro ad band tab.
3. Select the connection you want to edit and click the Ed i t button.
4. Configure the connection name, auto-connect behavior, and availability settings.
Five settings in the Ed i ti ng dialog are common to all connection types, see the G eneral
tab:
C o nnecti o n name — Enter a descriptive name for your network connection. This name
will be used to list this connection in the menu of the Netwo rk window.
Auto mati cal l y co nnect to thi s netwo rk when i t i s avai l abl e — Select
this box if you want N et wo rkMan ag er to auto-connect to this connection when it is
available. See Section 2.5.3, “ Connecting to a Network Automatically” for more
information.
Al l users may co nnect to thi s netwo rk — Select this box to create a connection
available to all users on the system. Changing this setting may require root privileges. See
Section 2.5.4, “ System-wide and Private Connection Profiles” for details.
42
⁠Chapt er 2 . Configure IP Net working
Auto mati cal l y co nnect to VP N when usi ng thi s co nnecti o n — Select this
box if you want N et wo rkMan ag er to auto-connect to a VPN connection when it is
available. Select the VPN from the drop-down menu.
Fi rewal l Zo ne — Select the Firewall Z one from the drop-down menu. See the Red Hat
Enterprise Linux 7 Security Guide for more information on Firewall Z ones.
5. Edit the mobile broadband-specific settings by referring to Section 2.5.8.1, “ Configuring the
Mobile Broadband Tab” .
Saving Yo ur Ne w (o r Mo difie d) Co nne ct io n and Making Furt he r Co nfigurat io ns
Once you have finished editing your mobile broadband connection, click the Appl y button to save
your customized configuration. If the profile was in use while being edited, power cycle the
connection to make N et wo rkMan ag er apply the changes. If the profile is OFF, set it to ON or select
it in the network connection icon's menu. See Section 2.5.1, “ Connecting to a Network Using a GUI”
for information on using your new or altered connection.
You can further configure an existing connection by selecting it in the Netwo rk C o nnecti o ns
window and clicking Ed i t to return to the Ed i ti ng dialog.
Then, to configure:
Po in t - t o - p o in t settings for the connection, click the P P P Setti ng s tab and proceed to
Section 2.5.10.3, “ Configuring PPP (Point-to-Point) Settings” ;
IP v4 settings for the connection, click the IP v4 Setti ng s tab and proceed to Section 2.5.10.4,
“ Configuring IPv4 Settings” ; or,
IP v6 settings for the connection, click the IP v6 Setti ng s tab and proceed to Section 2.5.10.5,
“ Configuring IPv6 Settings” .
2 .5 .8 .1 . Co nfiguring t he Mo bile Bro adband T ab
If you have already added a new mobile broadband connection using the assistant (see
Procedure 2.5, “ Adding a New Mobile Broadband Connection” for instructions), you can edit the
Mo bi l e Bro ad band tab to disable roaming if home network is not available, assign a network ID ,
or instruct N et wo rkMan ag er to prefer a certain technology (such as 3G or 2G) when using the
connection.
Number
The number that is dialed to establish a PPP connection with the GSM-based mobile
broadband network. This field may be automatically populated during the initial installation
of the broadband device. You can usually leave this field blank and enter the AP N instead.
Username
Enter the user name used to authenticate with the network. Some providers do not provide a
user name, or accept any user name when connecting to the network.
P asswo rd
Enter the password used to authenticate with the network. Some providers do not provide a
password, or accept any password.
AP N
43
Red Hat Ent erprise Linux 7 Net working G uide
Enter the Access Point Name (APN) used to establish a connection with the GSM-based
network. Entering the correct APN for a connection is important because it often determines:
how the user is billed for their network usage; and/or
whether the user has access to the Internet, an intranet, or a subnetwork.
Netwo rk ID
Entering a Netwo rk ID causes N et wo rkMan ag er to force the device to register only to a
specific network. This can be used to ensure the connection does not roam when it is not
possible to control roaming directly.
T ype
Any — The default value of Any leaves the modem to select the fastest network.
3G (UMT S/HSP A) — Force the connection to use only 3G network technologies.
2G (G P R S/ED G E) — Force the connection to use only 2G network technologies.
P refer 3G (UMT S/HSP A) — First attempt to connect using a 3G technology such as
HSPA or UMTS, and fall back to GPRS or ED GE only upon failure.
P refer 2G (G P R S/ED G E) — First attempt to connect using a 2G technology such as
GPRS or ED GE, and fall back to HSPA or UMTS only upon failure.
Al l o w ro ami ng i f ho me netwo rk i s no t avai l abl e
Uncheck this box if you want N et wo rkMan ag er to terminate the connection rather than
transition from the home network to a roaming one, thereby avoiding possible roaming
charges. If the box is checked, N et wo rkMan ag er will attempt to maintain a good
connection by transitioning from the home network to a roaming one, and vice versa.
P IN
If your device's SIM (Subscriber Identity Module) is locked with a PIN (Personal Identification
Number), enter the PIN so that N et wo rkMan ag er can unlock the device.
N et wo rkMan ag er must unlock the SIM if a PIN is required in order to use the device for
any purpose.
CD MA and EVD O have fewer options. They do not have the AP N, Netwo rk ID , or T ype options.
2.5.9. Est ablishing a DSL Connect ion
This section is intended for those installations which have a D SL card fitted within a host rather than
the external combined D SL modem router combinations typical of private consumer or SOHO
installations.
Pro ced u re 2.7. Ad d in g a N ew D SL C o n n ect io n
You can configure a new D SL connection by opening the Netwo rk C o nnecti o ns window, clicking
the Ad d button and selecting D SL from the Hard ware section of the new connection list.
1. Press the Super key to enter the Activities Overview, type nm-co nnecti o n-ed i to r and
then press Enter. The Netwo rk C o nnecti o ns tool appears.
2. Click the Ad d button.
3. The C ho o se a C o nnecti o n T ype list appears.
44
⁠Chapt er 2 . Configure IP Net working
4. Select D SL and press the C reate button.
5. The Ed i ti ng D SL C o nnecti o n 1 window appears.
Pro ced u re 2.8. Ed it in g an Exist in g D SL C o n n ect io n
You can configure an existing D SL connection by opening the Netwo rk C o nnecti o ns window
and selecting the name of the connection from the list. Then click the Ed i t button.
1. Press the Super key to enter the Activities Overview, type nm-co nnecti o n-ed i to r and
then press Enter. The Netwo rk C o nnecti o ns tool appears.
2. Select the connection you want to edit and click the Ed i t button.
Co nfiguring t he Co nne ct io n Nam e , Aut o -Co nne ct Be havio r, and Availabilit y
Se t t ings
Five settings in the Ed i ti ng dialog are common to all connection types, see the G eneral tab:
C o nnecti o n name — Enter a descriptive name for your network connection. This name will be
used to list this connection in the menu of the Netwo rk window.
Auto mati cal l y co nnect to thi s netwo rk when i t i s avai l abl e — Select this box if
you want N et wo rkMan ag er to auto-connect to this connection when it is available. See
Section 2.5.3, “ Connecting to a Network Automatically” for more information.
Al l users may co nnect to thi s netwo rk — Select this box to create a connection
available to all users on the system. Changing this setting may require root privileges. See
Section 2.5.4, “ System-wide and Private Connection Profiles” for details.
Auto mati cal l y co nnect to VP N when usi ng thi s co nnecti o n — Select this box if
you want N et wo rkMan ag er to auto-connect to a VPN connection when it is available. Select the
VPN from the drop-down menu.
Fi rewal l Zo ne — Select the Firewall Z one from the drop-down menu. See the Red Hat
Enterprise Linux 7 Security Guide for more information on Firewall Z ones.
Co nfiguring t he DSL T ab
Username
Enter the user name used to authenticate with the service provider.
Servi ce
Leave blank unless otherwise directed by your service provider.
P asswo rd
Enter the password supplied by the service provider.
Saving Yo ur Ne w (o r Mo difie d) Co nne ct io n and Making Furt he r Co nfigurat io ns
Once you have finished editing your D SL connection, click the Appl y button to save your
customized configuration. If the profile was in use while being edited, power cycle the connection to
make N et wo rkMan ag er apply the changes. If the profile is OFF, set it to ON or select it in the
network connection icon's menu. See Section 2.5.1, “ Connecting to a Network Using a GUI” for
information on using your new or altered connection.
45
Red Hat Ent erprise Linux 7 Net working G uide
You can further configure an existing connection by selecting it in the Netwo rk C o nnecti o ns
window and clicking Ed i t to return to the Ed i ti ng dialog.
Then, to configure:
T h e MAC ad d ress an d MT U settings, click the Wi red tab and proceed to Section 2.5.5.1,
“ Configuring the Connection Name, Auto-Connect Behavior, and Availability Settings” ;
Po in t - t o - p o in t settings for the connection, click the P P P Setti ng s tab and proceed to
Section 2.5.10.3, “ Configuring PPP (Point-to-Point) Settings” ;
IP v4 settings for the connection, click the IP v4 Setti ng s tab and proceed to Section 2.5.10.4,
“ Configuring IPv4 Settings” .
2.5.10. Configuring Connect ion Set t ings
2 .5 .1 0 .1 . Co nfiguring 8 0 2 .1 X Se curit y
802.1X security is the name of the IEEE standard for port-based Network Access Control (PNAC). It is
also called WPA Enterprise. Simply put, 802.1X security is a way of controlling access to a logical
network from a physical one. All clients who want to join the logical network must authenticate with
the server (a router, for example) using the correct 802.1X authentication method.
802.1X security is most often associated with securing wireless networks (WLANs), but can also be
used to prevent intruders with physical access to the network (LAN) from gaining entry. In the past,
D HC P servers were configured not to lease IP addresses to unauthorized users, but for various
reasons this practice is both impractical and insecure, and thus is no longer recommended. Instead,
802.1X security is used to ensure a logically-secure network through port-based authentication.
802.1X provides a framework for WLAN and LAN access control and serves as an envelope for
carrying one of the Extensible Authentication Protocol (EAP) types. An EAP type is a protocol that
defines how security is achieved on the network.
You can configure 802.1X security for a wired or wireless connection type by opening the Netwo rk
window (see Section 2.5.1, “ Connecting to a Network Using a GUI” ) and following the applicable
procedure below. Press the Super key to enter the Activities Overview, type co ntro l netwo rk and
then press Enter. The Netwo rk settings tool appears. Proceed to Procedure 2.9, “ For a Wired
Connection” or Procedure 2.10, “ For a Wireless Connection” :
Pro ced u re 2.9 . Fo r a Wired C o n n ect io n
1. Select a Wi red network interface from the left-hand-side menu.
2. Either click on Ad d P ro fi l e to add a new network connection profile for which you want to
configure 802.1X security, or select an existing connection profile and click the gear wheel
icon.
3. Then select Securi ty and set the symbolic power button to O N to enable settings
configuration.
4. Proceed to Section 2.5.10.1.1, “ Configuring TLS (Transport Layer Security) Settings”
Pro ced u re 2.10. Fo r a Wireless C o n n ect io n
1. Select a Wi rel ess network interface from the left-hand-side menu. If necessary, set the
symbolic power button to O N and check that your hardware switch is on.
46
⁠Chapt er 2 . Configure IP Net working
2. Either select the connection name of a new connection, or click the gear wheel icon of an
existing connection profile, for which you want to configure 802.1X security. In the case of a
new connection, complete any authentication steps to complete the connection and then click
the gear wheel icon.
3. Select Securi ty.
4. From the drop-down menu select one of the following security methods: LEAP, D yn amic
WEP ( 802.1X) , or WPA & WPA2 En t erp rise.
5. Refer to Section 2.5.10.1.1, “ Configuring TLS (Transport Layer Security) Settings” for
descriptions of which extensible authentication protocol (EAP) types correspond to your
selection in the Securi ty drop-down menu.
2.5.10.1.1. C o n f ig u rin g T LS ( T ran sp o rt Layer Secu rit y) Set t in g s
With Transport Layer Security, the client and server mutually authenticate using the TLS protocol.
The server demonstrates that it holds a digital certificate, the client proves its own identity using its
client-side certificate, and key information is exchanged. Once authentication is complete, the TLS
tunnel is no longer used. Instead, the client and server use the exchanged keys to encrypt data using
AES, TKIP or WEP.
The fact that certificates must be distributed to all clients who want to authenticate means that the
EAP-TLS authentication method is very strong, but also more complicated to set up. Using TLS
security requires the overhead of a public key infrastructure (PKI) to manage certificates. The benefit
of using TLS security is that a compromised password does not allow access to the (W)LAN: an
intruder must also have access to the authenticating client's private key.
N et wo rkMan ag er does not determine the version of TLS supported. N et wo rkMan ag er gathers the
parameters entered by the user and passes them to the daemon, wp a_su p p lican t , that handles the
procedure. It in turn uses OpenSSL to establish the TLS tunnel. OpenSSL itself negotiates the
SSL/TLS protocol version. It uses the highest version both ends support.
Select in g an Au t h en t icat io n Met h o d
Select from one of following authentication methods:
Select T LS for Transport Layer Security and proceed to Section 2.5.10.1.2, “ Configuring TLS
Settings” ;
Select FAST for Flexible Authentication via Secure Tunneling and proceed to Section 2.5.10.1.4,
“ Configuring Tunneled TLS Settings” ;
Select T unnel ed T LS for Tunneled Transport Layer Security, otherwise known as TTLS, or EAPTTLS and proceed to Section 2.5.10.1.4, “ Configuring Tunneled TLS Settings” ;
Select P ro tected EAP (P EAP ) for Protected Extensible Authentication Protocol and proceed to
Section 2.5.10.1.5, “ Configuring Protected EAP (PEAP) Settings” .
2.5.10.1.2. C o n f ig u rin g T LS Set t in g s
Id enti ty
Provide the identity of this server.
User certi fi cate
Click to browse for, and select, a personal X.509 certificate file encoded with Distinguished
Encoding Rules (D ER) or Privacy Enhanced Mail (PEM).
47
Red Hat Ent erprise Linux 7 Net working G uide
C A certi fi cate
Click to browse for, and select, an X.509 certificate authority certificate file encoded with
Distinguished Encoding Rules (D ER) or Privacy Enhanced Mail (PEM).
P ri vate key
Click to browse for, and select, a private key file encoded with Distinguished Encoding Rules
(D ER), Privacy Enhanced Mail (PEM), or the Personal Information Exchange Syntax Standard
(PKCS #12).
P ri vate key passwo rd
Enter the password for the private key in the P ri vate key field. Select Sho w passwo rd to
make the password visible as you type it.
2.5.10.1.3. C o n f ig u rin g FAST Set t in g s
Ano nymo us Id enti ty
Provide the identity of this server.
P AC pro vi si o ni ng
Select the check box to enable and then select from An o n ymo u s, Au t h en t icat ed , and
Both.
P AC fi l e
Click to browse for, and select, a protected access credential (PAC) file.
Inner authenti cati o n
G T C — Generic Token Card.
MSC H APv2 — Microsoft Challenge Handshake Authentication Protocol version 2.
Username
Enter the user name to be used in the authentication process.
P asswo rd
Enter the password to be used in the authentication process.
2.5.10.1.4 . C o n f ig u rin g T u n n eled T LS Set t in g s
Ano nymo us i d enti ty
This value is used as the unencrypted identity.
C A certi fi cate
Click to browse for, and select, a Certificate Authority's certificate.
Inner authenti cati o n
PAP — Password Authentication Protocol.
MSC H AP — Challenge Handshake Authentication Protocol.
MSC H APv2 — Microsoft Challenge Handshake Authentication Protocol version 2.
48
⁠Chapt er 2 . Configure IP Net working
C H AP — Challenge Handshake Authentication Protocol.
Username
Enter the user name to be used in the authentication process.
P asswo rd
Enter the password to be used in the authentication process.
2.5.10.1.5. C o n f ig u rin g Pro t ect ed EAP ( PEAP) Set t in g s
Ano nymo us Id enti ty
This value is used as the unencrypted identity.
C A certi fi cate
Click to browse for, and select, a Certificate Authority's certificate.
P EAP versi o n
The version of Protected EAP to use. Automatic, 0 or 1.
Inner authenti cati o n
MSC H APv2 — Microsoft Challenge Handshake Authentication Protocol version 2.
MD 5 — Message D igest 5, a cryptographic hash function.
G T C — Generic Token Card.
Username
Enter the user name to be used in the authentication process.
P asswo rd
Enter the password to be used in the authentication process.
2 .5 .1 0 .2 . Co nfiguring Wi-Fi Se curit y
Securi ty
N o n e — D o not encrypt the Wi-Fi connection.
WEP 4 0/128- b it K ey — Wired Equivalent Privacy (WEP), from the IEEE 802.11 standard.
Uses a single pre-shared key (PSK).
WEP 128- b it Passp h rase — An MD 5 hash of the passphrase will be used to derive a
WEP key.
LEAP — Lightweight Extensible Authentication Protocol, from Cisco Systems.
D yn amic WEP ( 802.1X) — WEP keys are changed dynamically. Use with
Section 2.5.10.1.1, “ Configuring TLS (Transport Layer Security) Settings”
WPA & WPA2 Perso n al — Wi-Fi Protected Access (WPA), from the draft IEEE 802.11i
standard. A replacement for WEP. Wi-Fi Protected Access II (WPA2), from the 802.11i-2004
standard. Personal mode uses a pre-shared key (WPA-PSK).
49
Red Hat Ent erprise Linux 7 Net working G uide
WPA & WPA2 En t erp rise — WPA for use with a RAD IUS authentication server to provide
IEEE 802.1X network access control. Use with Section 2.5.10.1.1, “ Configuring TLS
(Transport Layer Security) Settings”
Passwo rd
Enter the password to be used in the authentication process.
2 .5 .1 0 .3. Co nfiguring PPP (Po int -t o -Po int ) Se t t ings
C o nfi g ure Metho d s
Use po i nt-to -po i nt encrypti o n (MP P E)
Microsoft Point-To-Point Encryption protocol (RFC 3078).
Al l o w BSD d ata co mpressi o n
PPP BSD Compression Protocol (RFC 1977).
Al l o w D efl ate d ata co mpressi o n
PPP D eflate Protocol (RFC 1979).
Use T C P head er co mpressi o n
Compressing TCP/IP Headers for Low-Speed Serial Links (RFC 1144).
Send P P P echo packets
LCP Echo-Request and Echo-Reply Codes for loopback tests (RFC 1661).
2 .5 .1 0 .4 . Co nfiguring IPv4 Se t t ings
The IP v4 Setti ng s tab allows you to configure the method used to connect to a network, to enter
IP address, route, and D NS information as required. The IP v4 Setti ng s tab is available when
you create and modify one of the following connection types: wired, wireless, mobile broadband, VPN
or D SL. If you need to configure IP v6 addresses, see Section 2.5.10.5, “ Configuring IPv6 Settings” .
If you need to configure static routes, click the R o utes button and proceed to Section 2.5.10.6,
“ Configuring Routes” .
If you are using D HC P to obtain a dynamic IP address from a D HC P server, you can simply set
Metho d to Au t o mat ic ( D H C P) .
Set t in g t h e Met h o d
Availab le IPv4 Met h o d s b y C o n n ect io n T yp e
When you click the Metho d drop-down menu, depending on the type of connection you are
configuring, you are able to select one of the following IP v4 connection methods. All of the methods
are listed here according to which connection type, or types, they are associated with:
Metho d
Au t o mat ic ( D H C P) — Choose this option if the network you are connecting to uses a
D HC P server to assign IP addresses. You do not need to fill in the D HC P cl i ent ID field.
50
⁠Chapt er 2 . Configure IP Net working
Au t o mat ic ( D H C P) ad d resses o n ly — Choose this option if the network you are
connecting to uses a D HC P server to assign IP addresses but you want to assign D NS
servers manually.
Lin k- Lo cal O n ly — Choose this option if the network you are connecting to does not have
a D HC P server and you do not want to assign IP addresses manually. Random addresses
will be assigned as per RFC 3927 with prefix 16 9 . 254 /16 .
Sh ared t o o t h er co mp u t ers — Choose this option if the interface you are configuring is
for sharing an Internet or WAN connection. The interface is assigned an address in the
10 . 4 2. x. 1/24 range, a D HC P server and D NS server are started, and the interface is
connected to the default network connection on the system with network address translation
(NAT).
D isab led — IP v4 is disabled for this connection.
Wired , Wireless an d D SL C o n n ect io n Met h o d s
Man u al — Choose this option if you want to assign IP addresses manually.
Mo b ile B ro ad b an d C o n n ect io n Met h o d s
Au t o mat ic ( PPP) — Choose this option if the network you are connecting to assigns your
IP address and D NS servers automatically.
Au t o mat ic ( PPP) ad d resses o n ly — Choose this option if the network you are
connecting to assigns your IP address automatically, but you want to manually specify
D NS servers.
VPN C o n n ect io n Met h o d s
Au t o mat ic ( VPN ) — Choose this option if the network you are connecting to assigns your
IP address and D NS servers automatically.
Au t o mat ic ( VPN ) ad d resses o n ly — Choose this option if the network you are
connecting to assigns your IP address automatically, but you want to manually specify
D NS servers.
D SL C o n n ect io n Met h o d s
Au t o mat ic ( PPPo E) — Choose this option if the network you are connecting to assigns
your IP address and D NS servers automatically.
Au t o mat ic ( PPPo E) ad d resses o n ly — Choose this option if the network you are
connecting to assigns your IP address automatically, but you want to manually specify
D NS servers.
For information on configuring static routes for the network connection, go to Section 2.5.10.6,
“ Configuring Routes” .
2 .5 .1 0 .5 . Co nfiguring IPv6 Se t t ings
Metho d
Ig n o re — Choose this option if you want to ignore IP v6 settings for this connection.
Au t o mat ic — Choose this option to use SLAAC to create an automatic, stateless
configuration based on the hardware address and router advertisements (RA).
51
Red Hat Ent erprise Linux 7 Net working G uide
Au t o mat ic, ad d resses o n ly — Choose this option if the network you are connecting to
uses router advertisements (RA) to create an automatic, stateless configuration, but you want
to assign D NS servers manually.
Au t o mat ic, D H C P o n ly — Choose this option to not use RA, but request information from
D HC P v6 directly to create a stateful configuration.
Man u al — Choose this option if you want to assign IP addresses manually.
Lin k- Lo cal O n ly — Choose this option if the network you are connecting to does not have
a D HC P server and you do not want to assign IP addresses manually. Random addresses
will be assigned as per RFC 4862 with prefix FE80 : : 0 .
Ad d resses
D N S servers — Enter a comma separated list of D NS servers.
Search d o main s — Enter a comma separated list of domain controllers.
For information on configuring static routes for the network connection, go to Section 2.5.10.6,
“ Configuring Routes” .
2 .5 .1 0 .6 . Co nfiguring Ro ut e s
A host's routing table will be automatically populated with routes to directly connected networks. The
routes are learned by examining the network interfaces when they are “ up” . This section describes
entering static routes to networks or hosts which can be reached by traversing an intermediate
network or connection, such as a VPN tunnel or leased line. In order to reach a remote network or
host, the system is given the address of a gateway to which traffic should be sent.
When a host's interface is configured by D HC P , an address of a gateway that leads to an upstream
network or the Internet is usually assigned. This gateway is usually referred to as the default gateway
as it is the gateway to use if no better route is known to the system (and present in the routing table).
Network administrators often use the first or last host IP address in the network as the gateway
address; for example, 19 2. 16 8. 10 . 1 or 19 2. 16 8. 10 . 254 . Not to be confused by the address
which represents the network itself; in this example, 19 2. 16 8. 10 . 0 , or the subnet's broadcast
address; in this example 19 2. 16 8. 10 . 255.
C o n f ig u rin g St at ic R o u t es
To set a static route, open the IPv4 or IPv6 settings window for the connection you want to
configure. See Section 2.5.1, “ Connecting to a Network Using a GUI” for instructions on how to do
that.
R o utes
Ad d ress — Enter the IP address of a remote network, sub-net, or host.
N et mask — The netmask or prefix length of the IP address entered above.
G at eway — The IP address of the gateway leading to the remote network, sub-net, or host
entered above.
Met ric — A network cost, a preference value to give to this route. Lower values will be
preferred over higher values.
Auto mati c
52
⁠Chapt er 2 . Configure IP Net working
When Automatic is O N, routes from R A or D HC P are used, but you can also add additional
static routes. When O FF, only static routes you define are used.
Use thi s co nnecti o n o nl y fo r reso urces o n i ts netwo rk
Select this check box to prevent the connection from becoming the default route. Typical
examples are where a connection is a VPN tunnel or a leased line to a head office and you
do not want any Internet-bound traffic to pass over the connection. Selecting this option
means that only traffic specifically destined for routes learned automatically over the
connection or entered here manually will be routed over the connection.
2.6. Addit ional Resources
The following sources of information provide additional resources relevant to this chapter.
2.6.1. Inst alled Document at ion
i p(8) man page — D escribes the ip utility's command syntax.
nmcl i (1) man page — D escribes N et wo rkMan ag er's command‐line tool.
nmcl i -exampl es(5) man page — Gives examples of n mcli commands.
nm-setti ng s(5) man page — D escribes N et wo rkMan ag er properties and their settings.
2.6.2. Online Document at ion
Red Hat Enterprise Linux 7 Security Guide
D escribes IPsec based VPN and its configuration. D escribes the use of authenticated D NS
quires using D NSSEC.
RFC 1518 — C lassless In t er- D o main R o u t in g ( C ID R )
D escribes the CID R Address Assignment and Aggregation Strategy, including variablelength subnetting.
RFC 1918 — Ad d ress Allo cat io n f o r Privat e In t ern et s
D escribes the range of IP v4 addresses reserved for private use.
RFC 3330 — Sp ecial- U se IPv4 Ad d resses
D escribes the global and other specialized IP v4 address blocks that have been assigned
by the Internet Assigned Numbers Authority (IANA).
53
Red Hat Ent erprise Linux 7 Net working G uide
Chapter 3. Configure Host Names
3.1. Underst anding Host Names
There are three classes of ho stname: static, pretty, and transient.
The “ static” host name is the traditional ho stname, which can be chosen by the user, and is stored
in the /etc/ho stname file. The “ transient” ho stname is a dynamic host name maintained by the
kernel. It is initialized to the static host name by default, whose value defaults to “ localhost” . It can be
changed by D HC P or mD NS at runtime. The “ pretty” ho stname is a free-form UTF8 host name for
presentation to the user.
Note
A host name can be a free-form string up to 64 characters in length. However, Red Hat
recommends that both static and transient names match the fully-qualified domain name (FQD N)
used for the machine in D NS, such as ho st. exampl e. co m. It is also recommended that the
static and transient names consists only of 7 bit ASCII lower-case characters, no spaces or
dots, and limits itself to the format allowed for D NS domain name labels, even though this is
not a strict requirement. Older specifications do not permit the underscore, and so their use is
not recommended.
The h o st n amect l tool will enforce the following: Static and transient host names to consist of
a-z, A-Z, 0 -9 , “ -” , “ _” and “ . ” only, to not begin or end in a dot, and to not have two dots
immediately following each other. The size limit of 64 characters is enforced.
3.1.1. Recommended Naming Pract ices
The Internet Corporation for Assigned Names and Numbers (ICANN) sometimes adds previously
unregistered Top-Level D omains (such as . yo urco mpany) to the public register. Therefore, Red Hat
strongly recommends that you do not use a domain name that is not delegated to you, even on a
private network, as this can result in a domain name that resolves differently depending on network
configuration. As a result, network resources can become unavailable. Using domain names that are
not delegated to you also makes D NSSEC more difficult to deploy and maintain, as domain name
collisions require manual configuration to enable D NSSEC validation. See the ICANN FAQ on
domain name collision for more information on this issue.
3.2. Configuring Host Names Using T ext User Int erface, nmt ui
The text user interface tool n mt u i can be used to configure a host name in a terminal window. Issue
the following command to start the tool:
~]$ nmtui
The text user interface appears. Any invalid command prints a usage message.
54
⁠Chapt er 3. Configure Host Names
Fig u re 3.1. T h e N et wo rkMan ag er T ext U ser In t erf ace st art in g men u
To navigate, use the arrow keys or press T ab to step forwards and press Shi ft+T ab to step back
through the options. Press Enter to select an option. The Space bar toggles the status of a check
box.
See Section 1.5, “ Network Configuration Using a Text User Interface (nmtui)” for information on
installing n mt u i.
The N et wo rkMan ag er text user interface tool n mt u i can be used to query and set the static host
name in the /etc/ho stname file. Note that at time of writing, changing the host name in this way will
not be noticed by h o st n amect l.
To force h o st n amect l to notice the change in the static host name, restart ho stnamed as ro o t:
~]# systemctl restart systemd -ho stnamed
3.3. Configuring Host Names Using host namect l
The h o st n amect l tool is provided for administering the three separate classes of host names in use
on a given system.
3.3.1. View All t he Host Names
To view all the current host names, enter the following command:
~]$ ho stnamectl status
The status option is implied by default if no option is given.
3.3.2. Set All t he Host Names
To set all the host names on a system, enter the following command as ro o t:
~]# ho stnamectl set-ho stname name
55
Red Hat Ent erprise Linux 7 Net working G uide
This will alter the pretty, static, and transient host names alike. The static and transient host names
will be simplified forms of the pretty host name. Spaces will be replaced with “ -” and special
characters will be removed.
3.3.3. Set a Part icular Host Name
To set a particular host name, enter the following command as ro o t with the relevant option:
~]# ho stnamectl set-ho stname name [option. . . ]
Where option is one or more of: --pretty, --stati c, and --transi ent.
If the --stati c or --transi ent options are used together with the --pretty option, the static and
transient host names will be simplified forms of the pretty host name. Spaces will be replaced with “ -”
and special characters will be removed. If the --pretty option is not given, no simplification takes
place.
When setting a pretty host name, remember to use the appropriate quotation marks if the host name
contains spaces or a single quotation mark. For example:
~]# ho stnamectl set-ho stname "Stephen' s no tebo o k" --pretty
3.3.4 . Clear a Part icular Host Name
To clear a particular host name and allow it to revert to the default, enter the following command as
ro o t with the relevant option:
~]# ho stnamectl set-ho stname "" [option. . . ]
Where "" is a quoted empty string and where option is one or more of: --pretty, --stati c, and -transi ent.
3.3.5. Changing Host Names Remot ely
To execute a ho stnamectl command on a remote system, use the -H, --ho st option as follows:
~]# ho stnamectl set-ho stname -H [username]@ hostname
Where hostname is the remote host you want to configure. The username is optional. The
h o st n amect l tool will use SSH to connect to the remote system.
3.4 . Configuring Host Names Using nmcli
The N et wo rkMan ag er tool n mcli can be used to query and set the static host name in the
/etc/ho stname file. Note that at time of writing, changing the host name in this way will not be
noticed by h o st n amect l.
To query the static host name, issue the following command:
~]$ nmcl i g eneral ho stname
To set the static host name to my-server, issue the following command as ro o t:
56
⁠Chapt er 3. Configure Host Names
~]# nmcl i g eneral ho stname my-server
To force h o st n amect l to notice the change in the static host name, restart ho stnamed as ro o t:
~]# systemctl restart systemd -ho stnamed
3.5. Addit ional Resources
The following sources of information provide additional resources regarding h o st n amect l.
3.5.1. Inst alled Document at ion
ho stnamectl (1) man page — D escribes h o st n amect l including the commands and
command options.
ho stname(1) man page — Contains an explanation of the ho stname and d o mai nname
commands.
ho stname(5) man page — Contains an explanation of the host name file, its contents, and use.
ho stname(7) man page — Contains an explanation of host name resolution.
machi ne-i nfo (5) man page — D escribes the local machine information file and the
environment variables it contains.
machi ne-i d (5) man page — D escribes the local machine ID configuration file.
systemd -ho stnamed . servi ce(8) man page — D escribes the systemd -ho stnamed system
service used by h o st n amect l.
57
Red Hat Ent erprise Linux 7 Net working G uide
Chapter 4. Configure Network Bonding
Red Hat Enterprise Linux 7 allows administrators to bind multiple network interfaces together into a
single, bonded, channel. Channel bonding enables two or more network interfaces to act as one,
simultaneously increasing the bandwidth and providing redundancy.
Warning
The use of direct cable connections without network switches is not supported for bonding.
The failover mechanisms described here will not work as expected without the presence of
network switches. See the Red Hat Knowledgebase article Why is bonding in not supported with
direct connection using crossover cables? for more information.
Note
The active-backup, balance-tlb and balance-alb modes do not require any specific
configuration of the switch. Other bonding modes require configuring the switch to aggregate
the links. For example, a Cisco switch requires EtherChannel for Modes 0, 2, and 3, but for
Mode 4 LACP and EtherChannel are required. See the documentation supplied with your
switch and see https://www.kernel.org/doc/D ocumentation/networking/bonding.txt
4 .1. Underst anding t he Default Behavior of Mast er and Slave
Int erfaces
When controlling bonded slave interfaces using the Netwo rkManag er daemon, and especially when
fault finding, keep the following in mind:
1. Starting the master interface does not automatically start the slave interfaces.
2. Starting a slave interface always starts the master interface.
3. Stopping the master interface also stops the slave interfaces.
4. A master without slaves can start static IP connections.
5. A master without slaves waits for slaves when starting D HC P connections.
6. A master with a D HC P connection waiting for slaves completes when a slave with a carrier is
added.
7. A master with a D HC P connection waiting for slaves continues waiting when a slave without a
carrier is added.
4 .2. Configure Bonding Using t he T ext User Int erface, nmt ui
The text user interface tool n mt u i can be used to configure bonding in a terminal window. Issue the
following command to start the tool:
~]$ nmtui
58
⁠Chapt er 4 . Configure Net work Bonding
The text user interface appears. Any invalid command prints a usage message.
To navigate, use the arrow keys or press T ab to step forwards and press Shi ft+T ab to step back
through the options. Press Enter to select an option. The Space bar toggles the status of a check
box.
1. From the starting menu, select Ed i t a co nnecti o n. Select Ad d , the New C o nnecti o n
screen opens.
Fig u re 4 .1. T h e N et wo rkMan ag er T ext U ser In t erf ace Ad d a B o n d C o n n ect io n
men u
2. Select Bo nd and then C reate; the Ed i t co nnecti o n screen for the bond will open.
59
Red Hat Ent erprise Linux 7 Net working G uide
Fig u re 4 .2. T h e N et wo rkMan ag er T ext U ser In t erf ace C o n f ig u rin g a B o n d
C o n n ect io n men u
3. At this point slave interfaces will need to be added to the bond; to add these select Ad d , the
New C o nnecti o n screen opens. Once the type of Connection has been chosen select the
C reate button.
60
⁠Chapt er 4 . Configure Net work Bonding
Fig u re 4 .3. T h e N et wo rkMan ag er T ext U ser In t erf ace C o n f ig u rin g a N ew B o n d
Slave C o n n ect io n men u
4. The slave's Ed i t C o nnecti o n display appears; enter the desired slave's device name or
MAC address in the D evi ce section. If required, enter a clone MAC address to be used as the
bond's MAC address by selecting Sho w to the right of the Ethernet label. Select the O K
button to save the slave.
Note
If the device is specified without a MAC address the D evi ce section will be
automatically populated once the Ed i t C o nnecti o n window is reloaded, but only if
it successfully finds the device.
61
Red Hat Ent erprise Linux 7 Net working G uide
Fig u re 4 .4 . T h e N et wo rkMan ag er T ext U ser In t erf ace C o n f ig u rin g a B o n d Slave
C o n n ect io n men u
5. The name of the bond slave appears in the Sl aves section. Repeat the above steps to add
further slave connections.
6. Review and confirm the settings before selecting the O K button.
62
⁠Chapt er 4 . Configure Net work Bonding
Fig u re 4 .5. T h e N et wo rkMan ag er T ext U ser In t erf ace C o mp let ed B o n d
See Section 4.6.1.1, “ Configuring the Bond Tab” for definitions of the bond terms.
See Section 1.5, “ Network Configuration Using a Text User Interface (nmtui)” for information on
installing n mt u i.
4 .3. Using t he Net workManager Command Line T ool, nmcli
To create a bond, named mybond0, issue a command as follows:
~]$ nmcl i co n ad d type bo nd co n-name mybond0 i fname mybond0 mo d e
acti ve-backup
Connection 'mybond0' (9301ff97-abbc-4432-aad1-246d7faea7fb) successfully
added.
To add a slave interface, issue a command in the following form:
~]$ nmcl i co n ad d type bo nd -sl ave i fname ens7 master mybond0
To add additional slaves, repeat the previous command with a new interface, for example:
63
Red Hat Ent erprise Linux 7 Net working G uide
~]$ nmcl i co n ad d type bo nd -sl ave i fname ens3 master mybond0
Connection 'bond-slave-ens3-1' (50c59350-1531-45f4-ba04-33431c16e386)
successfully added.
Note that as no co n-name was given for the slaves, the name was derived from the interface name by
prepending the type. At time of writing, n mcli only supports Ethernet slaves.
In order to bring up a bond, the slaves must be brought up first as follows:
~]$ nmcl i co n up bo nd -sl ave-ens7
Connection successfully activated (D-Bus active path:
/org/freedesktop/NetworkManager/ActiveConnection/14)
~]$ nmcl i co n up bo nd -sl ave-ens3
Connection successfully activated (D-Bus active path:
/org/freedesktop/NetworkManager/ActiveConnection/15)
Now bring up the bond as follows:
~]$ nmcl i co n up bo nd -mybond0
Connection successfully activated (D-Bus active path:
/org/freedesktop/NetworkManager/ActiveConnection/16)
See Section 2.3, “ Using the NetworkManager Command Line Tool, nmcli” for an introduction to
n mcli
4 .4 . Using t he Command Line Int erface (CLI)
A bond is created using the bo nd i ng kernel module and a special network interface called a channel
bonding interface.
4 .4 .1. Check if Bonding Kernel Module is Inst alled
In Red Hat Enterprise Linux 7, the bonding module is not loaded by default. You can load the module
by issuing the following command as ro o t:
~]# mo d pro be --fi rst-ti me bo nd i ng
This activation will not persist across system restarts. See the Red Hat Enterprise Linux 7 System
Administrator's Guide for an explanation of persistent module loading. Note that given a correct
configuration file using the BO ND ING _O P T S directive, the bonding module will be loaded as
required and therefore does not need to be loaded separately.
To display information about the module, issue the following command:
~]$ mo d i nfo bo nd i ng
See the mo d pro be(8) man page for more command options.
4 .4 .2. Creat e a Channel Bonding Int erface
64
⁠Chapt er 4 . Configure Net work Bonding
To create a channel bonding interface, create a file in the /etc/sysco nfi g /netwo rk-scri pts/
directory called i fcfg -bo nd N, replacing N with the number for the interface, such as 0 .
The contents of the file can be based on a configuration file for whatever type of interface is getting
bonded, such as an Ethernet interface. The essential differences are that the D EVIC E directive is
bo nd N, replacing N with the number for the interface, and T Y P E= Bo nd . In addition, set
BO ND ING _MAST ER = yes.
Examp le 4 .1. Examp le if cf g - b o n d 0 In t erf ace C o n f ig u rat io n File
An example of a channel bonding interface.
DEVICE=bond0
NAME=bond0
TYPE=Bond
BONDING_MASTER=yes
IPADDR=192.168.1.1
PREFIX=24
ONBOOT=yes
BOOTPROTO=none
BONDING_OPTS="bonding parameters separated by spaces"
The NAME directive is useful for naming the connection profile in N et wo rkMan ag er. ONBOOT
says whether the profile should be started when booting (or more generally, when auto-connecting
a device).
Important
Parameters for the bonding kernel module must be specified as a space-separated list in the
BO ND ING _O P T S= "bonding parameters" directive in the i fcfg -bo nd N interface file. D o
not specify options for the bonding device in /etc/mo d pro be. d /bonding. co nf, or in the
deprecated /etc/mo d pro be. co nf file.
The max_bo nd s parameter is not interface specific and should not be set when using i fcfg bo nd N files with the BO ND ING _O P T S directive as this directive will cause the network scripts
to create the bond interfaces as required.
For further instructions and advice on configuring the bonding module and to view the list of
bonding parameters, see Section 4.5, “ Using Channel Bonding” .
4 .4 .3. Creat ing SLAVE Int erfaces
The channel bonding interface is the “ master” and the interfaces to be bonded are referred to as the
“ slaves” . After the channel bonding interface is created, the network interfaces to be bound together
must be configured by adding the MAST ER and SLAVE directives to the configuration files of the
slaves. The configuration files for each of the slave interfaces can be nearly identical.
Examp le 4 .2. Examp le Slave In t erf ace C o n f ig u rat io n File
For example, if two Ethernet interfaces are being channel bonded, eth0 and eth1, they can both
look like the following example:
65
Red Hat Ent erprise Linux 7 Net working G uide
DEVICE=ethN
NAME=bond0-slave
TYPE=Ethernet
BOOTPROTO=none
ONBOOT=yes
MASTER=bond0
SLAVE=yes
In this example, replace N with the numerical value for the interface. Note that if more than one
profile or configuration file exists with O NBO O T = yes for an interface, they may race with each
other and a plain T Y P E= Ethernet profile may be activated instead of a bond slave.
4 .4 .4 . Act ivat ing a Channel Bond
To activate a bond, bring up all the slaves. As ro o t, issue the following commands:
~]# i fup i fcfg -eth0
Connection successfully activated (D-Bus active path:
/org/freedesktop/NetworkManager/ActiveConnection/7)
~]# i fup i fcfg -eth1
Connection successfully activated (D-Bus active path:
/org/freedesktop/NetworkManager/ActiveConnection/8)
Note that if editing interface files for interfaces which are currently “ up” , set them down first as follows:
ifdown ethN
Then when complete, bring up all the slaves, which will bring up the bond (provided it was not set
“ down” ).
To make N et wo rkMan ag er aware of the changes, issue a command for every changed interface as
ro o t:
~]# nmcl i co n l o ad /etc/sysco nfi g /netwo rk-scri pts/i fcfg -device
Alternatively, to reload all interfaces:
~]# nmcl i co n rel o ad
The default behavior is for N et wo rkMan ag er not to be aware of the changes and to continue using
the old configuration data. The is set by the mo ni to r-co nnecti o n-fi l es option in the
Netwo rkManag er. co nf file. See the Netwo rkManag er. co nf(5) manual page for more
information.
To view the status of the bond interface, issue the following command:
~]# i p l i nk sho w
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode
DEFAULT
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
66
⁠Chapt er 4 . Configure Net work Bonding
2: eth0: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 1500 qdisc
pfifo_fast master bond0 state UP mode DEFAULT qlen 1000
link/ether 52:54:00:e9:ce:d2 brd ff:ff:ff:ff:ff:ff
3: eth1: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 1500 qdisc
pfifo_fast master bond0 state UP mode DEFAULT qlen 1000
link/ether 52:54:00:38:a6:4c brd ff:ff:ff:ff:ff:ff
4: bond0: <BROADCAST,MULTICAST,MASTER,UP,LOWER_UP> mtu 1500 qdisc noqueue
state UP mode DEFAULT
link/ether 52:54:00:38:a6:4c brd ff:ff:ff:ff:ff:ff
4 .4 .5. Creat ing Mult iple Bonds
In Red Hat Enterprise Linux 7, for each bond a channel bonding interface is created including the
BO ND ING _O P T S directive. This configuration method is used so that multiple bonding devices can
have different configurations. To create multiple channel bonding interfaces, proceed as follows:
Create multiple i fcfg -bo nd N files with the BO ND ING _O P T S directive; this directive will cause
the network scripts to create the bond interfaces as required.
Create, or edit existing, interface configuration files to be bonded and include the SLAVE directive.
Assign the interfaces to be bonded, the slave interfaces, to the channel bonding interfaces by
means of the MAST ER directive.
Examp le 4 .3. Examp le mu lt ip le if cf g - b o n d N in t erf ace co n f ig u rat io n f iles
The following is an example of a channel bonding interface configuration file:
DEVICE=bondN
NAME=bondN
TYPE=Bond
BONDING_MASTER=yes
IPADDR=192.168.1.1
PREFIX=24
ONBOOT=yes
BOOTPROTO=none
BONDING_OPTS="bonding parameters separated by spaces"
In this example, replace N with the number for the bond interface. For example, to create two bonds
create two configuration files, i fcfg -bo nd 0 and i fcfg -bo nd 1, with appropriate IP
addresses.
Create the interfaces to be bonded as per Example 4.2, “ Example Slave Interface Configuration File”
and assign them to the bond interfaces as required using the MAST ER = bo nd N directive. For
example, continuing on from the example above, if two interfaces per bond are required, then for two
bonds create four interface configuration files and assign the first two using MAST ER = bo nd 0 and
the next two using MAST ER = bo nd 1.
4 .5. Using Channel Bonding
67
Red Hat Ent erprise Linux 7 Net working G uide
To enhance performance, adjust available module options to ascertain what combination works
best. Pay particular attention to the mi i mo n or arp_i nterval and the arp_i p_targ et
parameters. See Section 4.5.1, “ Bonding Module D irectives” for a list of available options and how to
quickly determine the best ones for your bonded interface.
4 .5.1. Bonding Module Direct ives
It is a good idea to test which channel bonding module parameters work best for your bonded
interfaces before adding them to the BO ND ING _O P T S= "bonding parameters" directive in your
bonding interface configuration file (i fcfg -bo nd 0 for example). Parameters to bonded interfaces
can be configured without unloading (and reloading) the bonding module by manipulating files in
the sysfs file system.
sysfs is a virtual file system that represents kernel objects as directories, files and symbolic links.
sysfs can be used to query for information about kernel objects, and can also manipulate those
objects through the use of normal file system commands. The sysfs virtual file system is mounted
under the /sys/ directory. All bonding interfaces can be configured dynamically by interacting with
and manipulating files under the /sys/cl ass/net/ directory.
In order to determine the best parameters for your bonding interface, create a channel bonding
interface file such as i fcfg -bo nd 0 by following the instructions in Section 4.4.2, “ Create a Channel
Bonding Interface” . Insert the SLAVE= yes and MAST ER = bo nd 0 directives in the configuration files
for each interface bonded to bo nd 0 . Once this is completed, you can proceed to testing the
parameters.
First, bring up the bond you created by running i fup bo nd N as ro o t:
~]# i fup bo nd 0
If you have correctly created the i fcfg -bo nd 0 bonding interface file, you will be able to see bo nd 0
listed in the output of running i p l i nk sho w as ro o t:
~]# i p l i nk sho w
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode
DEFAULT
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
2: eth0: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 1500 qdisc
pfifo_fast master bond0 state UP mode DEFAULT qlen 1000
link/ether 52:54:00:e9:ce:d2 brd ff:ff:ff:ff:ff:ff
3: eth1: <BROADCAST,MULTICAST,SLAVE,UP,LOWER_UP> mtu 1500 qdisc
pfifo_fast master bond0 state UP mode DEFAULT qlen 1000
link/ether 52:54:00:38:a6:4c brd ff:ff:ff:ff:ff:ff
4: bond0: <BROADCAST,MULTICAST,MASTER,UP,LOWER_UP> mtu 1500 qdisc noqueue
state UP mode DEFAULT
link/ether 52:54:00:38:a6:4c brd ff:ff:ff:ff:ff:ff
To view all existing bonds, even if they are not up, run:
~]$ cat /sys/cl ass/net/bo nd i ng _masters
bond0
You can configure each bond individually by manipulating the files located in the
/sys/cl ass/net/bo nd N/bo nd i ng / directory. First, the bond you are configuring must be taken
down:
68
⁠Chapt er 4 . Configure Net work Bonding
~]# i fd o wn bo nd 0
As an example, to enable MII monitoring on bond0 with a 1 second interval, run as ro o t:
~]# echo 10 0 0 > /sys/cl ass/net/bo nd 0 /bo nd i ng /mi i mo n
To configure bond0 for balance-alb mode, run either:
~]# echo 6 > /sys/cl ass/net/bo nd 0 /bo nd i ng /mo d e
...or, using the name of the mode:
~]# echo bal ance-al b > /sys/cl ass/net/bo nd 0 /bo nd i ng /mo d e
After configuring options for the bond in question, you can bring it up and test it by running i fup
bo nd N. If you decide to change the options, take the interface down, modify its parameters using
sysfs, bring it back up, and re-test.
Once you have determined the best set of parameters for your bond, add those parameters as a
space-separated list to the BONDING_OPTS= directive of the /etc/sysco nfi g /netwo rkscri pts/i fcfg -bo nd N file for the bonding interface you are configuring. Whenever that bond is
brought up (for example, by the system during the boot sequence if the ONBOOT=yes directive is set),
the bonding options specified in the BONDING_OPTS will take effect for that bond.
The following list provides the names of many of the more common channel bonding parameters,
along with a description of what they do. For more information, see the brief descriptions for each
parm in mo d i nfo bo nd i ng output, or for more detailed information, see
https://www.kernel.org/doc/D ocumentation/networking/bonding.txt.
B o n d in g In t erf ace Paramet ers
ad _sel ect= value
Specifies the 802.3ad aggregation selection logic to use. Possible values are:
stabl e or 0 — D efault setting. The active aggregator is chosen by largest aggregate
bandwidth. Reselection of the active aggregator occurs only when all slaves of the
active aggregator are down or if the active aggregator has no slaves.
band wi d th or 1 — The active aggregator is chosen by largest aggregate bandwidth.
Reselection occurs if:
A slave is added to or removed from the bond;
Any slave's link state changes;
Any slave's 802.3ad association state changes;
The bond's administrative state changes to up.
co unt or 2 — The active aggregator is chosen by the largest number of slaves.
Reselection occurs as described for the band wi d th setting above.
The band wi d th and co unt selection policies permit failover of 802.3ad aggregations
when partial failure of the active aggregator occurs. This keeps the aggregator with the
highest availability, either in bandwidth or in number of slaves, active at all times.
69
Red Hat Ent erprise Linux 7 Net working G uide
arp_i nterval = time_in_milliseconds
Specifies, in milliseconds, how often AR P monitoring occurs.
Important
It is essential that both arp_i nterval and arp_i p_targ et parameters are
specified, or, alternatively, the mi i mo n parameter is specified. Failure to do so can
cause degradation of network performance in the event that a link fails.
If using this setting while in mo d e= 0 or mo d e= 2 (the two load-balancing modes), the
network switch must be configured to distribute packets evenly across the NICs. For more
information on how to accomplish this, see
https://www.kernel.org/doc/D ocumentation/networking/bonding.txt.
The value is set to 0 by default, which disables it.
arp_i p_targ et= ip_address[,ip_address_2,… ip_address_16]
Specifies the target IP address of AR P requests when the arp_i nterval parameter is
enabled. Up to 16 IP addresses can be specified in a comma separated list.
arp_val i d ate= value
Validate source/distribution of AR P probes; default is no ne. Other valid values are acti ve,
backup, and al l .
d o wnd el ay= time_in_milliseconds
Specifies (in milliseconds) how long to wait after link failure before disabling the link. The
value must be a multiple of the value specified in the mi i mo n parameter. The value is set to
0 by default, which disables it.
fai l _o ver_mac= value
Specifies whether active-backup mode should set all slaves to the same MAC address at
enslavement (the traditional behavior), or, when enabled, perform special handling of the
bond's MAC address in accordance with the selected policy. Possible values are:
no ne or 0 — D efault setting. This setting disables fai l _o ver_mac, and causes
bonding to set all slaves of an active-backup bond to the same MAC address at
enslavement time.
acti ve or 1 — The “ active” fai l _o ver_mac policy indicates that the MAC address of
the bond should always be the MAC address of the currently active slave. The MAC
address of the slaves is not changed; instead, the MAC address of the bond changes
during a failover.
This policy is useful for devices that cannot ever alter their MAC address, or for devices
that refuse incoming broadcasts with their own source MAC (which interferes with the
ARP monitor). The disadvantage of this policy is that every device on the network must
be updated via gratuitous ARP, as opposed to the normal method of switches snooping
incoming traffic to update their ARP tables. If the gratuitous ARP is lost, communication
may be disrupted.
70
⁠Chapt er 4 . Configure Net work Bonding
When this policy is used in conjunction with the MII monitor, devices which assert link
up prior to being able to actually transmit and receive are particularly susceptible to
loss of the gratuitous ARP, and an appropriate updelay setting may be required.
fo l l o w or 2 — The “ follow” fai l _o ver_mac policy causes the MAC address of the
bond to be selected normally (normally the MAC address of the first slave added to the
bond). However, the second and subsequent slaves are not set to this MAC address
while they are in a backup role; a slave is programmed with the bond's MAC address at
failover time (and the formerly active slave receives the newly active slave's MAC
address).
This policy is useful for multiport devices that either become confused or incur a
performance penalty when multiple ports are programmed with the same MAC address.
lacp _rat e= value
Specifies the rate at which link partners should transmit LACPD U packets in 802.3ad mode.
Possible values are:
sl o w or 0 — D efault setting. This specifies that partners should transmit LACPD Us
every 30 seconds.
fast or 1 — Specifies that partners should transmit LACPD Us every 1 second.
mi i mo n= time_in_milliseconds
Specifies (in milliseconds) how often MII link monitoring occurs. This is useful if high
availability is required because MII is used to verify that the NIC is active. To verify that the
driver for a particular NIC supports the MII tool, type the following command as root:
~]# ethto o l interface_name | g rep "Li nk d etected : "
In this command, replace interface_name> with the name of the device interface, such as
eth0 , not the bond interface. If MII is supported, the command returns:
Link detected: yes
If using a bonded interface for high availability, the module for each NIC must support MII.
Setting the value to 0 (the default), turns this feature off. When configuring this setting, a
good starting point for this parameter is 10 0 .
Important
It is essential that both arp_i nterval and arp_i p_targ et parameters are
specified, or, alternatively, the mi i mo n parameter is specified. Failure to do so can
cause degradation of network performance in the event that a link fails.
mo d e= value
Allows you to specify the bonding policy. The value can be one of:
bal ance-rr or 0 — Sets a round-robin policy for fault tolerance and load balancing.
Transmissions are received and sent out sequentially on each bonded slave interface
beginning with the first one available.
71
Red Hat Ent erprise Linux 7 Net working G uide
acti ve-backup or 1 — Sets an active-backup policy for fault tolerance. Transmissions
are received and sent out via the first available bonded slave interface. Another bonded
slave interface is only used if the active bonded slave interface fails.
bal ance-xo r or 2 — Transmissions are based on the selected hash policy. The
default is to derive a hash by XOR of the source and destination MAC addresses
multiplied by the modulo of the number of slave interfaces. In this mode traffic destined
for specific peers will always be sent over the same interface. As the destination is
determined by the MAC addresses this method works best for traffic to peers on the same
link or local network. If traffic has to pass through a single router then this mode of traffic
balancing will be suboptimal.
bro ad cast or 3 — Sets a broadcast policy for fault tolerance. All transmissions are sent
on all slave interfaces.
80 2. 3ad or 4 — Sets an IEEE 802.3ad dynamic link aggregation policy. Creates
aggregation groups that share the same speed and duplex settings. Transmits and
receives on all slaves in the active aggregator. Requires a switch that is 802.3ad
compliant.
bal ance-tl b or 5 — Sets a Transmit Load Balancing (TLB) policy for fault tolerance
and load balancing. The outgoing traffic is distributed according to the current load on
each slave interface. Incoming traffic is received by the current slave. If the receiving
slave fails, another slave takes over the MAC address of the failed slave. This mode is
only suitable for local addresses known to the kernel bonding module and therefore
cannot be used behind a bridge with virtual machines.
bal ance-al b or 6 — Sets an Adaptive Load Balancing (ALB) policy for fault tolerance
and load balancing. Includes transmit and receive load balancing for IP v4 traffic.
Receive load balancing is achieved through AR P negotiation. This mode is only
suitable for local addresses known to the kernel bonding module and therefore cannot
be used behind a bridge with virtual machines.
pri mary= interface_name
Specifies the interface name, such as eth0 , of the primary device. The pri mary device is
the first of the bonding interfaces to be used and is not abandoned unless it fails. This
setting is particularly useful when one NIC in the bonding interface is faster and, therefore,
able to handle a bigger load.
This setting is only valid when the bonding interface is in acti ve-backup mode. See
https://www.kernel.org/doc/D ocumentation/networking/bonding.txt for more information.
pri mary_resel ect= value
Specifies the reselection policy for the primary slave. This affects how the primary slave is
chosen to become the active slave when failure of the active slave or recovery of the primary
slave occurs. This parameter is designed to prevent flip-flopping between the primary slave
and other slaves. Possible values are:
al ways or 0 (default) — The primary slave becomes the active slave whenever it comes
back up.
better or 1 — The primary slave becomes the active slave when it comes back up, if the
speed and duplex of the primary slave is better than the speed and duplex of the current
active slave.
fai l ure or 2 — The primary slave becomes the active slave only if the current active
slave fails and the primary slave is up.
72
⁠Chapt er 4 . Configure Net work Bonding
The pri mary_resel ect setting is ignored in two cases:
If no slaves are active, the first slave to recover is made the active slave.
When initially enslaved, the primary slave is always made the active slave.
Changing the pri mary_resel ect policy via sysfs will cause an immediate selection of
the best active slave according to the new policy. This may or may not result in a change of
the active slave, depending upon the circumstances
resend _i g mp= range
Specifies the number of IGMP membership reports to be issued after a failover event. One
membership report is issued immediately after the failover, subsequent packets are sent in
each 200ms interval.
The valid range is 0 to 255; the default value is 1. A value of 0 prevents the IGMP
membership report from being issued in response to the failover event.
This option is useful for bonding modes bal ance-rr (mode 0), acti ve-backup (mode 1),
bal ance-tl b (mode 5) and bal ance-al b (mode 6), in which a failover can switch the
IGMP traffic from one slave to another. Therefore a fresh IGMP report must be issued to
cause the switch to forward the incoming IGMP traffic over the newly selected slave.
upd el ay= time_in_milliseconds
Specifies (in milliseconds) how long to wait before enabling a link. The value must be a
multiple of the value specified in the mi i mo n parameter. The value is set to 0 by default,
which disables it.
use_carri er= number
Specifies whether or not mi i mo n should use MII/ETHTOOL ioctls or
neti f_carri er_o k() to determine the link state. The neti f_carri er_o k() function
relies on the device driver to maintains its state with neti f_carri er_on/off; most device
drivers support this function.
The MII/ETHTOOL ioctls tools utilize a deprecated calling sequence within the kernel.
However, this is still configurable in case your device driver does not support
neti f_carri er_on/off.
Valid values are:
1 — D efault setting. Enables the use of neti f_carri er_o k().
0 — Enables the use of MII/ETHTOOL ioctls.
Note
If the bonding interface insists that the link is up when it should not be, it is possible
that your network device driver does not support neti f_carri er_on/off.
xmi t_hash_po l i cy= value
Selects the transmit hash policy used for slave selection in bal ance-xo r and 80 2. 3ad
modes. Possible values are:
73
Red Hat Ent erprise Linux 7 Net working G uide
0 or l ayer2 — D efault setting. This parameter uses the XOR of hardware MAC
addresses to generate the hash. The formula used is:
(source_MAC_address XOR destination_MAC) MODULO slave_count
This algorithm will place all traffic to a particular network peer on the same slave, and is
802.3ad compliant.
1 or l ayer3+ 4 — Uses upper layer protocol information (when available) to generate
the hash. This allows for traffic to a particular network peer to span multiple slaves,
although a single connection will not span multiple slaves.
The formula for unfragmented TCP and UD P packets used is:
((source_port XOR dest_port) XOR
((source_IP XOR dest_IP) AND 0 xffff)
MODULO slave_count
For fragmented TCP or UD P packets and all other IP protocol traffic, the source and
destination port information is omitted. For non-IP traffic, the formula is the same as the
l ayer2 transmit hash policy.
This policy intends to mimic the behavior of certain switches; particularly, Cisco
switches with PFC2 as well as some Foundry and IBM products.
The algorithm used by this policy is not 802.3ad compliant.
2 or l ayer2+ 3 — Uses a combination of layer2 and layer3 protocol information to
generate the hash.
Uses XOR of hardware MAC addresses and IP addresses to generate the hash. The
formula is:
(((source_IP XOR dest_IP) AND 0 xffff) XOR
( source_MAC XOR destination_MAC ))
MODULO slave_count
This algorithm will place all traffic to a particular network peer on the same slave. For
non-IP traffic, the formula is the same as for the layer2 transmit hash policy.
This policy is intended to provide a more balanced distribution of traffic than layer2
alone, especially in environments where a layer3 gateway device is required to reach
most destinations.
This algorithm is 802.3ad compliant.
4 .6. Creat ing a Bond Connect ion Using a GUI
You can use the GNOME co n t ro l- cen t er utility to direct N et wo rkMan ag er to create a Bond from
two or more Wired or InfiniBand connections. It is not necessary to create the connections to be
bonded first. They can be configured as part of the process to configure the bond. You must have the
MAC addresses of the interfaces available in order to complete the configuration process.
4 .6.1. Est ablishing a Bond Connect ion
74
⁠Chapt er 4 . Configure Net work Bonding
Pro ced u re 4 .1. Ad d in g a N ew B o n d C o n n ect io n
Follow the below steps to create a new bond connection.
1. Press the Super key to enter the Activities Overview, type co ntro l netwo rk and then press
Enter. The N et wo rk settings tool appears. This step is fully covered in Section 2.5, “ Using
NetworkManager with the GNOME Graphical User Interface” .
2. Click the plus symbol to open the selection list. Select Bo nd . The Ed i ti ng Bo nd
co nnecti o n 1 window appears.
Fig u re 4 .6 . T h e N et wo rkMan ag er G rap h ical U ser In t erf ace Ad d a B o n d men u
75
Red Hat Ent erprise Linux 7 Net working G uide
3. On the Bo nd tab, click Ad d and select the type of interface you want to use with the bond
connection. Click the C reate button. Note that the dialog to select the slave type only comes
up when you create the first slave; after that, it will automatically use that same type for all
further slaves.
4. The Ed i ti ng bo nd 0 sl ave 1 window appears. Use the D evi ce MAC ad d ress dropdown menu to select the MAC address of the interface to be bonded. The first slave's MAC
address will be used as the MAC address for the bond interface. If required, enter a clone MAC
address to be used as the bond's MAC address. Click the Save button.
Fig u re 4 .7. T h e N et wo rkMan ag er G rap h ical U ser In t erf ace Ad d a B o n d
C o n n ect io n men u
5. The name of the bonded slave appears in the Bo nd ed co nnecti o ns window. Click the
Ad d button to add further slave connections.
6. Review and confirm the settings and then click the Save button.
7. Edit the bond-specific settings by referring to Section 4.6.1.1, “ Configuring the Bond Tab”
below.
Pro ced u re 4 .2. Ed it in g an Exist in g B o n d C o n n ect io n
Follow these steps to edit an existing bond connection.
76
⁠Chapt er 4 . Configure Net work Bonding
1. Press the Super key to enter the Activities Overview, type co ntro l netwo rk and then press
Enter. The N et wo rk settings tool appears.
2. Select the connection you want to edit and click the O pti o ns button.
3. Select the G eneral tab.
4. Configure the connection name, auto-connect behavior, and availability settings.
Five settings in the Ed i ti ng dialog are common to all connection types, see the G eneral
tab:
C o nnecti o n name — Enter a descriptive name for your network connection. This name
will be used to list this connection in the menu of the Netwo rk window.
Auto mati cal l y co nnect to thi s netwo rk when i t i s avai l abl e — Select
this box if you want N et wo rkMan ag er to auto-connect to this connection when it is
available. See Section 2.5.3, “ Connecting to a Network Automatically” for more
information.
Al l users may co nnect to thi s netwo rk — Select this box to create a connection
available to all users on the system. Changing this setting may require root privileges. See
Section 2.5.4, “ System-wide and Private Connection Profiles” for details.
Auto mati cal l y co nnect to VP N when usi ng thi s co nnecti o n — Select this
box if you want N et wo rkMan ag er to auto-connect to a VPN connection when it is
available. Select the VPN from the drop-down menu.
Fi rewal l Zo ne — Select the firewall zone from the drop-down menu. See the Red Hat
Enterprise Linux 7 Security Guide for more information on firewall zones.
5. Edit the bond-specific settings by referring to Section 4.6.1.1, “ Configuring the Bond Tab”
below.
Saving Yo ur Ne w (o r Mo difie d) Co nne ct io n and Making Furt he r Co nfigurat io ns
Once you have finished editing your bond connection, click the Save button to save your
customized configuration. If the profile was in use while being edited, power cycle the connection to
make N et wo rkMan ag er apply the changes. If the profile is OFF, set it to ON or select it in the
network connection icon's menu. See Section 2.5.1, “ Connecting to a Network Using a GUI” for
information on using your new or altered connection.
You can further configure an existing connection by selecting it in the Netwo rk window and clicking
O pti o ns to return to the Ed i ti ng dialog.
Then, to configure:
IP v4 settings for the connection, click the IP v4 Setti ng s tab and proceed to Section 2.5.10.4,
“ Configuring IPv4 Settings” ; or,
IP v6 settings for the connection, click the IP v6 Setti ng s tab and proceed to Section 2.5.10.5,
“ Configuring IPv6 Settings” .
Once saved the Bond will appear in the Network settings tool with each slave showing in the display.
77
Red Hat Ent erprise Linux 7 Net working G uide
Fig u re 4 .8. T h e N et wo rkMan ag er G rap h ical U ser In t erf ace wit h B o n d
4 .6 .1 .1 . Co nfiguring t he Bo nd T ab
If you have already added a new bond connection (refer to Procedure 4.1, “ Adding a New Bond
Connection” for instructions), you can edit the Bo nd tab to set the load sharing mode and the type of
link monitoring to use to detect failures of a slave connection.
Mo d e
The mode that is used to share traffic over the slave connections which make up the bond.
The default is R o und -ro bi n. Other load sharing modes, such as 80 2. 3ad , can be
selected by means of the drop-down list.
Li nk Mo ni to ri ng
The method of monitoring the slaves ability to carry network traffic.
The following modes of load sharing are selectable from the Mo d e drop-down list:
R o und -ro bi n
Sets a round-robin policy for fault tolerance and load balancing. Transmissions are
received and sent out sequentially on each bonded slave interface beginning with the first
one available. This mode might not work behind a bridge with virtual machines without
additional switch configuration.
Acti ve backup
Sets an active-backup policy for fault tolerance. Transmissions are received and sent out
via the first available bonded slave interface. Another bonded slave interface is only used if
the active bonded slave interface fails. Note that this is the only mode available for bonds of
InfiniBand devices.
XO R
78
⁠Chapt er 4 . Configure Net work Bonding
Sets an XOR (exclusive-or) policy. Transmissions are based on the selected hash policy.
The default is to derive a hash by XOR of the source and destination MAC addresses
multiplied by the modulo of the number of slave interfaces. In this mode traffic destined for
specific peers will always be sent over the same interface. As the destination is determined
by the MAC addresses this method works best for traffic to peers on the same link or local
network. If traffic has to pass through a single router then this mode of traffic balancing will
be suboptimal.
Bro ad cast
Sets a broadcast policy for fault tolerance. All transmissions are sent on all slave
interfaces. This mode might not work behind a bridge with virtual machines without
additional switch configuration.
80 2. 3ad
Sets an IEEE 80 2. 3ad dynamic link aggregation policy. Creates aggregation groups that
share the same speed and duplex settings. Transmits and receives on all slaves in the
active aggregator. Requires a network switch that is 80 2. 3ad compliant.
Ad apti ve transmi t l o ad bal anci ng
Sets an adaptive Transmit Load Balancing (TLB) policy for fault tolerance and load
balancing. The outgoing traffic is distributed according to the current load on each slave
interface. Incoming traffic is received by the current slave. If the receiving slave fails,
another slave takes over the MAC address of the failed slave. This mode is only suitable for
local addresses known to the kernel bonding module and therefore cannot be used behind
a bridge with virtual machines.
Ad apti ve l o ad bal anci ng
Sets an Adaptive Load Balancing (ALB) policy for fault tolerance and load balancing.
Includes transmit and receive load balancing for IP v4 traffic. Receive load balancing is
achieved through AR P negotiation. This mode is only suitable for local addresses known to
the kernel bonding module and therefore cannot be used behind a bridge with virtual
machines.
The following types of link monitoring can be selected from the Li nk Mo ni to ri ng drop-down list. It
is a good idea to test which channel bonding module parameters work best for your bonded
interfaces.
MII (Med i a Ind epend ent Interface)
The state of the carrier wave of the interface is monitored. This can be done by querying the
driver, by querying MII registers directly, or by using et h t o o l to query the device. Three
options are available:
Mo ni to ri ng Freq uency
The time interval, in milliseconds, between querying the driver or MII registers.
Li nk up d el ay
The time in milliseconds to wait before attempting to use a link that has been
reported as up. This delay can be used if some gratuitous AR P requests are lost
in the period immediately following the link being reported as “ up” . This can
happen during switch initialization for example.
Li nk d o wn d el ay
79
Red Hat Ent erprise Linux 7 Net working G uide
The time in milliseconds to wait before changing to another link when a previously
active link has been reported as “ down” . This delay can be used if an attached
switch takes a relatively long time to change to backup mode.
AR P
The address resolution protocol (AR P ) is used to probe one or more peers to determine how
well the link-layer connections are working. It is dependent on the device driver providing
the transmit start time and the last receive time.
Two options are available:
Mo ni to ri ng Freq uency
The time interval, in milliseconds, between sending AR P requests.
AR P targ ets
A comma separated list of IP addresses to send AR P requests to.
4 .7. Addit ional Resources
The following sources of information provide additional resources regarding network bonding.
4 .7.1. Inst alled Document at ion
nmcl i (1) man page — D escribes N et wo rkMan ag er's command‐line tool.
nmcl i -exampl es(5) man page — Gives examples of n mcli commands.
nm-setti ng s(5) man page — D escription of settings and parameters of N et wo rkMan ag er
connections.
4 .7.2. Online Document at ion
Red Hat Enterprise Linux 7 System Administrator's Guide
Explains the use of kernel module capabilities.
h t t p s://access.red h at .co m/sit e/n o d e/284 21/C o n f ig u rin g _VLAN _d evices_o ver_a_b o n
d ed _in t erf ace
A Red Hat Knowledgebase article about Configuring VLAN devices over a bonded interface.
80
⁠Chapt er 5. Configure Net work T eaming
Chapter 5. Configure Network Teaming
5.1. Underst anding Net work T eaming
The combining or aggregating together of network links in order to provide a logical link with higher
throughput, or to provide redundancy, is known by many names such as “ channel bonding” ,
“ Ethernet bonding” , “ port trunking” , “ channel teaming” , “ NIC teaming” , “ link aggregation” , and so
on. This concept as originally implemented in the Linux kernel is widely referred to as “ bonding” . The
term Network Teaming has been chosen to refer to this new implementation of the concept. The
existing bonding driver is unaffected, Network Teaming is offered as an alternative and does not
replace bonding in Red Hat Enterprise Linux 7.
Network Teaming, or Team, is designed to implement the concept in a different way by providing a
small kernel driver to implement the fast handling of packet flows, and various user-space
applications to do everything else in user space. The driver has an Application Programming Interface
(API), referred to as “ Team Netlink API” , which implements Netlink communications. User-space
applications can use this API to communicate with the driver. A library, referred to as “ lib” , has been
provided to do user space wrapping of Team Netlink communications and RT Netlink messages. An
application daemon, teamd , which uses Libteam lib is also provided. One instance of teamd can
control one instance of the Team driver. The daemon implements the load-balancing and activebackup logic, such as round-robin, by using additional code referred to as “ runners” . By separating
the code in this way, the Network Teaming implementation presents an easily extensible and scalable
solution for load-balancing and redundancy requirements. For example, custom runners can be
relatively easily written to implement new logic via teamd , and even teamd is optional, users can
write their own application to use lib t eam.
A tool to control a running instance of teamd using D -bus is provided by t eamd ct l. It provides a D Bus wrapper around the teamd D -Bus API. By default, teamd listens and communicates using Unix
D omain Sockets but still monitors D -Bus. This is to ensure that teamd can be used in environments
where D -Bus is not present or not yet loaded. For example, when booting over teamd links D -Bus
would not yet be loaded. The t eamd ct l tool can be used during run time to read the configuration,
the state of link-watchers, check and change the state of ports, add and remove ports, and to change
ports between active and backup states.
Team Netlink API communicates with user-space applications using Netlink messages. The userspace library lib t eam does not directly interact with the API, but uses lib n l or t eamn l to interact
with the driver API.
To sum up, the instances of Team driver, running in the kernel, do not get configured or controlled
directly. All configuration is done with the aid of user space applications, such as the t eamd
application. The application then directs the kernel driver part accordingly.
5.2. Underst anding t he Default Behavior of Mast er and Slave Int erfaces
When controlling teamed port interfaces using the Netwo rkManag er daemon, and especially when
fault finding, keep the following in mind:
1. Starting the master interface does not automatically start the port interfaces.
2. Starting a port interface always starts the master interface.
3. Stopping the master interface also stops the port interfaces.
4. A master without ports can start static IP connections.
5. A master without ports waits for ports when starting D HC P connections.
81
Red Hat Ent erprise Linux 7 Net working G uide
6. A master with a D HC P connection waiting for ports completes when a port with a carrier is
added.
7. A master with a D HC P connection waiting for ports continues waiting when a port without a
carrier is added.
Warning
The use of direct cable connections without network switches is not supported for teaming.
The failover mechanisms described here will not work as expected without the presence of
network switches. See the Red Hat Knowledgebase article Why is bonding not supported with
direct connection using crossover cables? for more information.
5.3. Comparison of Net work T eaming t o Bonding
T ab le 5.1. A C o mp ariso n o f Feat u res in B o n d in g an d T eam
Feat u re
B o n d in g
T eam
broadcast Tx policy
round-robin Tx policy
active-backup Tx policy
LACP (802.3ad) support
Hash-based Tx policy
User can set hash function
Tx load-balancing support
(TLB)
LACP hash port select
load-balancing for LACP
support
Ethtool link monitoring
ARP link monitoring
NS/NA (IPv6) link monitoring
ports up/down delays
port priorities and stickiness
(“ primary” option
enhancement)
separate per-port link
monitoring setup
multiple link monitoring setup
lockless Tx/Rx path
VLAN support
user-space runtime control
Logic in user-space
Extensibility
Modular design
Performance overhead
D -Bus interface
multiple device stacking
zero config using LLD P
Yes
Yes
Yes
Yes (passive only)
Yes
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
Yes
Yes
Yes
Yes
No
Yes
No
Yes
Yes
Yes
Yes
Yes
No
Yes
Limited
No (rwlock)
Yes
Limited
No
Hard
No
Low
No
Yes
No
Yes
Yes (RCU)
Yes
Full
Yes
Easy
Yes
Very Low
Yes
Yes
(in planning)
82
⁠Chapt er 5. Configure Net work T eaming
Feat u re
B o n d in g
T eam
NetworkManager support
Yes
Yes
5.4 . Underst anding t he Net work T eaming Daemon and t he "Runners"
The Team daemon, teamd , uses lib t eam to control one instance of the team driver. This instance of
the team driver adds instances of a hardware device driver to form a “ team” of network links. The
team driver presents a network interface, team0 for example, to the other parts of the kernel. The
interfaces created by instances of the team driver are given names such as team0, team1, and so
forth in the documentation. This is for ease of understanding and other names can be used. The
logic common to all methods of teaming is implemented by teamd ; those functions that are unique to
the different load sharing and backup methods, such as round-robin, are implemented by separate
units of code referred to as “ runners” . Because words such as “ module” and “ mode” already have
specific meanings in relation to the kernel, the word “ runner” was chosen to refer to these units of
code. The user specifies the runner in the JSON format configuration file and the code is then
compiled into an instance of teamd when the instance is created. A runner is not a plug-in because
the code for a runner is compiled into an instance of teamd as it is being created. Code could be
created as a plug-in for teamd should the need arise.
The following runners are available at time of writing.
broadcast (data is transmitted over all ports)
round-robin (data is transmitted over all ports in turn)
active-backup (one port or link is used while others are kept as a backup)
loadbalance (with active Tx load balancing and BPF-based Tx port selectors)
lacp (implements the 802.3ad Link Aggregation Control Protocol)
In addition, the following link-watchers are available:
et h t o o l (Libteam lib uses et h t o o l to watch for link state changes). This is the default if no other
link-watcher is specified in the configuration file.
arp _p in g (The arp _p in g utility is used to monitor the presence of a far-end hardware address
using ARP packets.)
n sn a_p in g (Neighbor Advertisements and Neighbor Solicitation from the IP v6 Neighbor
D iscovery protocol are used to monitor the presence of a neighbor's interface)
There are no restrictions in the code to prevent a particular link-watcher from being used with a
particular runner, however when using the lacp runner, et h t o o l is the only recommended linkwatcher.
5.5. Inst all t he Net work T eaming Daemon
The networking teaming daemon, teamd , is not installed by default. To install teamd , issue the
following command as ro o t:
~]# yum i nstal l teamd
5.6. Convert ing a Bond t o a T eam
83
Red Hat Ent erprise Linux 7 Net working G uide
It is possible to convert existing bonding configuration files to team configuration files using the
b o n d 2t eam tool. It can convert bond configuration files in i fcfg format to team configuration files
in either i fcfg or JSON format. Note that firewall rules, alias interfaces, and anything that might be
tied to the original interface name can break after the renaming because the tool will only change the
i fcfg file, nothing else.
To see some examples of the command format, issue the following command:
~]$ bo nd 2team --exampl es
New files will be created in a directory whose name starts with /tmp/bo nd 2team. XXXXXX/, where
XXXXXX is a random string. After creating the new configuration files, move the old bonding files to a
backup folder and then move the new files to the /etc/sysco nfi g /netwo rk-scri pts/ directory.
Examp le 5.1. C o n vert a B o n d t o a T eam
To convert a current bo nd 0 configuration to team i fcfg , issue a command as ro o t:
~]# /usr/bi n/bo nd 2team --master bo nd 0
Note that this will retain the name bo nd 0 . To use a new name to save the configuration, use the -rename as follows:
~]# /usr/bi n/bo nd 2team --master bo nd 0 --rename team0
add the --jso n option to output JSON format files instead of i fcfg files. See the
teamd . co nf(5) man page for examples of JSON format.
Examp le 5.2. C o n vert a B o n d t o a T eam an d Sp ecif y t h e File Pat h
To convert a current bo nd 0 configuration to team i fcfg , and to manually specify the path to the
i fcfg file, issue a command as ro o t:
~]# /usr/bi n/bo nd 2team --master bo nd 0 --co nfi g d i r /path/to/ifcfg-file
add the --jso n option to output JSON format files instead of i fcfg files.
Examp le 5.3. C reat e a T eam C o n f ig u rat io n U sin g B o n d 2t eam
It is also possible to create a team configuration by supplying the b o n d 2t eam tool with a list of
bonding parameters. For example:
~]# /usr/bi n/bo nd 2team --bo nd i ng _o pts "mo d e= 1 mi i mo n= 50 0 "
Ports can also be supplied on the command line as follows:
~]# /usr/bi n/bo nd 2team --bo nd i ng _o pts "mo d e= 1 mi i mo n= 50 0
pri mary= eth1 \
pri mary_resel ect-0 " --po rt eth1 --po rt eth2 --po rt eth3 --po rt eth4
See the bo nd 2team(1) man page for further details. For an explanation of bonding parameters, see
84
⁠Chapt er 5. Configure Net work T eaming
Section 4.5, “ Using Channel Bonding”
5.7. Select ing Int erfaces t o Use as Port s for a Net work T eam
To view the available interfaces, issue the following command:
~]$ i p l i nk sho w
1: lo: <LOOPBACK,UP,LOWER_UP > mtu 65536 qdisc noqueue state UNKNOWN
mode DEFAULT
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
2: em1: <BROADCAST,MULTICAST,UP,LOWER_UP > mtu 1500 qdisc pfifo_fast
state UP mode DEFAULT qlen 1000
link/ether 52:54:00:6a:02:8a brd ff:ff:ff:ff:ff:ff
3: em2: <BROADCAST,MULTICAST,UP,LOWER_UP > mtu 1500 qdisc pfifo_fast
state UP mode DEFAULT qlen 1000
link/ether 52:54:00:9b:6d:2a brd ff:ff:ff:ff:ff:ff
From the available interfaces, determine which are suitable for adding to your network team and then
proceed to Section 5.8, “ Selecting Network Team Configuration Methods”
Note
The Team developers prefer the term “ port” rather than “ slave” , however N et wo rkMan ag er
uses the term “ team-slave” to refer to interfaces that make up a team.
5.8. Select ing Net work T eam Configurat ion Met hods
T o co n f ig u re a n et wo rk t eam u sin g N et wo rkMan ag er's text user interface tool, n mt u i,
proceed to Section 5.9, “ Configure a Network Team Using the Text User Interface, nmtui”
T o creat e a n et wo rk t eam u sin g t h e co mman d - lin e t o o l, n mcli, proceed to Section 5.10.1,
“ Configure Network Teaming Using nmcli” .
T o creat e a n et wo rk t eam u sin g t h e T eam d aemo n , teamd , proceed to Section 5.10.2,
“ Creating a Network Team Using teamd” .
T o creat e a n et wo rk t eam u sin g co n f ig u rat io n f iles, proceed to Section 5.10.3, “ Creating a
Network Team Using ifcfg Files” .
T o co n f ig u re a n et wo rk t eam u sin g a g rap h ical u ser in t erf ace, see Section 5.13, “ Creating a
Network Team Using a GUI”
5.9. Configure a Net work T eam Using t he T ext User Int erface, nmt ui
The text user interface tool n mt u i can be used to configure teaming in a terminal window. Issue the
following command to start the tool:
~]$ nmtui
The text user interface appears. Any invalid command prints a usage message.
85
Red Hat Ent erprise Linux 7 Net working G uide
To navigate, use the arrow keys or press T ab to step forwards and press Shi ft+T ab to step back
through the options. Press Enter to select an option. The Space bar toggles the status of a check
box.
1. From the starting menu, select Ed i t a co nnecti o n. Select Ad d , the New C o nnecti o n
screen opens.
Fig u re 5.1. T h e N et wo rkMan ag er T ext U ser In t erf ace Ad d a T eam C o n n ect io n
men u
2. Select T eam, the Ed i t co nnecti o n screen opens.
86
⁠Chapt er 5. Configure Net work T eaming
Fig u re 5.2. T h e N et wo rkMan ag er T ext U ser In t erf ace C o n f ig u rin g a T eam
C o n n ect io n men u
3. To add port interfaces to the team select Ad d , the New C o nnecti o n screen opens. Once the
type of Connection has been chosen select the C reate button to cause the team's Ed i t
C o nnecti o n display to appear.
87
Red Hat Ent erprise Linux 7 Net working G uide
Fig u re 5.3. T h e N et wo rkMan ag er T ext U ser In t erf ace C o n f ig u rin g a n ew T eam
Po rt In t erf ace C o n n ect io n men u
4. Enter the desired slave's device name or MAC address in the D evi ce section. If required,
enter a clone MAC address to be used as the team's MAC address by selecting Sho w to the
right of the Ethernet label. Select the O K button.
Note
If the device is specified without a MAC address the D evi ce section will be
automatically populated once the Ed i t C o nnecti o n window is reloaded, but only if
it successfully finds the device.
88
⁠Chapt er 5. Configure Net work T eaming
Fig u re 5.4 . T h e N et wo rkMan ag er T ext U ser In t erf ace C o n f ig u rin g a T eam' s Po rt
In t erf ace C o n n ect io n men u
5. The name of the teamed slave appears in the Sl aves section. Repeat the above steps to add
further slave connections.
6. If custom port settings are to be applied select the Ed i t button under the JSO N
co nfi g urati o n section. This will launch a vim console where changes may be applied.
Once finished write the changes from vim and then confirm that the displayed JSON string
under JSO N co nfi g urati o n matches what is intended.
7. Review and confirm the settings before selecting the O K button.
89
Red Hat Ent erprise Linux 7 Net working G uide
Fig u re 5.5. T h e N et wo rkMan ag er T ext U ser In t erf ace C o n f ig u rin g a T eam
C o n n ect io n men u
See Section 5.12, “ Configure teamd Runners” for examples of JSON strings. Note that only the
relevant sections from the example strings should be used for a team or port configuration using
n mt u i. D o not specify the “ D evice” as part of the JSON string. For example, only the JSON string
after “ device” but before “ port” should be used in the Team JSON configuration field. All JSON
strings relevant to a port must only be added in the port configuration field.
See Section 1.5, “ Network Configuration Using a Text User Interface (nmtui)” for information on
installing n mt u i.
5.10. Configure a Net work T eam Using t he Command Line
5.10.1. Configure Net work T eaming Using nmcli
To view the devices available on the system, issue the following command:
~]$ nmcl i co nnecti o n sho w
NAME UUID
eth1 0e8185a1-f0fd-4802-99fb-bedbb31c689b
eth0 dfe1f57b-419d-4d1c-aaf5-245deab82487
90
TYPE
802-3-ethernet
802-3-ethernet
DEVICE
---
⁠Chapt er 5. Configure Net work T eaming
To create a new team interface, with name team-ServerA, issue a command as follows:
~]$ nmcl i co nnecti o n ad d type team i fname team-ServerA
Connection 'team-ServerA' (b954c62f-5fdd-4339-97b0-40efac734c50)
successfully added.
N et wo rkMan ag er will set its internal parameter co nnecti o n. auto co nnect to yes and as no IP
address was given i pv4 . metho d will be set to auto . N et wo rkMan ag er will also write a
configuration file to /etc/sysco nfi g /netwo rk-scri pts/i fcfg -team-ServerA where the
corresponding ONBOOT will be set to yes and BOOTPROTO will be set to d hcp.
Note that manual changes to the ifcfg file will not be noticed by N et wo rkMan g er until the interface is
next brought up. See Section 1.9, “ Network Configuration Using sysconfig Files” for more information
on using configuration files.
To view the other values assigned, issue a command as follows:
~]$ nmcl i co n sho w team-ServerA
connection.id:
connection.uuid:
40efac734c50
connection.interface-name:
connection.type:
connection.autoconnect:
…
ipv4.method:
[output truncated]
team-ServerA
b954c62f-5fdd-4339-97b0ServerA
team
yes
auto
As no JSON configuration file was specified the default values apply. See the teamd . co nf(5) man
page for more information on the team JSON parameters and their default values. Notice that the
name was derived from the interface name by prepending the type. Alternatively, specify a name with
the co n-name option as follows:
~]$ nmcl i co nnecti o n ad d type team co n-name Team0 i fname ServerB
Connection 'Team0' (5f7160a1-09f6-4204-8ff0-6d96a91218a7) successfully
added.
To view the team interfaces just configured, issue a command as follows:
~]$ nmcl i co n sho w
NAME
UUID
TYPE
DEVICE
team-ServerA
b954c62f-5fdd-4339-97b0-40efac734c50 team
ServerA
eth1
0e8185a1-f0fd-4802-99fb-bedbb31c689b 802-3-ethernet
-eth0
dfe1f57b-419d-4d1c-aaf5-245deab82487 802-3-ethernet
-Team0
5f7160a1-09f6-4204-8ff0-6d96a91218a7 team
ServerB
To change the name assigned to a team, issue a command in the following format:
nmcl i co n mo d old-team-name co nnecti o n. i d new-team-name
91
Red Hat Ent erprise Linux 7 Net working G uide
To load a team configuration file for a team that already exists, issue a command in the following
format:
nmcl i co nnecti o n mo d i fy team-name team. co nfi g JSON-config
You can specify the team configuration either as a JSON string or provide a file name containing the
configuration. The file name can include the path. In both cases, what is stored in the team.config
property is the JSON string. In the case of a JSON string, use single quotes around the string and
paste the entire string to the command line.
To review the team. co nfi g property, enter a command in the following format:
nmcl i co n sho w team-name | g rep team. co nfi g
To add an interface eth0 to T eam0 , with the name Team0-port1, issue a command as follows:
~]$ nmcl i co n ad d type team-sl ave co n-name Team0-port1 i fname eth0
master Team0
Connection 'Team0-port1' (ccd87704-c866-459e-8fe7-01b06cf1cffc)
successfully added.
Similarly, to add another interface, eth1, with the name Team0-port2, issue a command as follows:
~]$ nmcl i co n ad d type team-sl ave co n-name Team0-port2 i fname eth1
master Team0
Connection 'Team0-port2' (a89ccff8-8202-411e-8ca6-2953b7db52dd)
successfully added.
At time of writing, n mcli only supports Ethernet ports.
In order to bring up a team, the ports must be brought up first as follows:
~]$ nmcl i co nnecti o n up Team0-port1
Connection successfully activated (D-Bus active path:
/org/freedesktop/NetworkManager/ActiveConnection/2)
~]$ nmcl i co nnecti o n up Team0-port2
Connection successfully activated (D-Bus active path:
/org/freedesktop/NetworkManager/ActiveConnection/3)
You can verify that the team interface was brought up by the activation of the ports, as follows:
~]$ i p l i nk
3: Team0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state
UP mode DEFAULT
link/ether 52:54:00:76:6f:f0 brd ff:ff:ff:ff:ff:f
Alternatively, issue a command to bring up the team as follows:
~]$ nmcl i co nnecti o n up Team0
Connection successfully activated (D-Bus active path:
/org/freedesktop/NetworkManager/ActiveConnection/4)
92
⁠Chapt er 5. Configure Net work T eaming
See Section 2.3, “ Using the NetworkManager Command Line Tool, nmcli” for an introduction to
n mcli
5.10.2. Creat ing a Net work T eam Using t eamd
Note
Configurations created using teamd are not persistent, and as such it may be necessary to
create a team using the steps defined in Section 5.10.1, “ Configure Network Teaming Using
nmcli” or Section 5.10.3, “ Creating a Network Team Using ifcfg Files” .
To create a network team, a JSON format configuration file is required for the virtual interface that will
serve as the interface to the team of ports or links. A quick way is to copy the example configuration
files and then edit them using an editor running with ro o t privileges. To list the available example
configurations, enter the following command:
~]$ l s /usr/share/d o c/teamd -*/exampl e_co nfi g s/
activebackup_arp_ping_1.conf activebackup_multi_lw_1.conf
loadbalance_2.conf
activebackup_arp_ping_2.conf activebackup_nsna_ping_1.conf
loadbalance_3.conf
activebackup_ethtool_1.conf
broadcast.conf
activebackup_ethtool_2.conf
lacp_1.conf
roundrobin_2.conf
activebackup_ethtool_3.conf
loadbalance_1.conf
roundrobin.conf
random.conf
To view one of the included files, such as acti vebackup_ethto o l _1. co nf, enter the following
command:
~]$ cat /usr/share/d o c/teamd */exampl e_co nfi g s/acti vebackup_ethto o l _1. co nf
{
"device": "team0",
"runner": {"name": "activebackup"},
"link_watch": {"name": "ethtool"},
"ports": {
"eth1": {
"prio": -10,
"sticky": true
},
"eth2": {
"prio": 100
}
}
}
Create a working configurations directory to store teamd configuration files. For example, as normal
user, enter a command with the following format:
~]$ mkd i r ~ /teamd_working_configs
93
Red Hat Ent erprise Linux 7 Net working G uide
Copy the file you have chosen to your working directory and edit it as necessary. As an example, you
could use a command with the following format:
~]$ cp /usr/share/d o c/teamd */exampl e_co nfi g s/acti vebackup_ethto o l _1. co nf \
~ /teamd_working_configs/acti vebackup_ethto o l _1. co nf
To edit the file to suit your environment, for example to change the interfaces to be used as ports for
the network team, open the file for editing as follows:
~]$ vi ~ /teamd_working_configs/acti vebackup_ethto o l _1. co nf
Make any necessary changes and save the file. See the vi (1) man page for help on using the vi
editor or use your preferred editor.
Note that it is essential that the interfaces to be used as ports within the team must not be active, that
is to say, they must be “ down” , when adding them into a team device. To check their status, issue the
following command:
~]$ i p l i nk sho w
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode
DEFAULT
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
2: em1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state
UP mode DEFAULT qlen 1000
link/ether 52:54:00:d5:f7:d4 brd ff:ff:ff:ff:ff:ff
3: em2: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state
UP mode DEFAULT qlen 1000
link/ether 52:54:00:d8:04:70 brd ff:ff:ff:ff:ff:ff
In this example we see that both the interfaces we plan to use are “ UP” .
To take down an interface, issue a command as ro o t in the following format:
~]# i p l i nk set d o wn em1
Repeat for each interface as necessary.
To create a team interface based on the configuration file, as ro o t user, change to the working
configurations directory (teamd_working_configs in this example):
~]# cd /ho me/userteamd_working_configs
Then issue a command in the following format:
~]# teamd -g -f acti vebackup_ethto o l _1. co nf -d
Using team device "team0".
Using PID file "/var/run/teamd/team0.pid"
Using config file
"/home/user/teamd_working_configs/activebackup_ethtool_1.conf"
The -g option is for debug messages, -f option is to specify the configuration file to load, and the d option is to make the process run as a daemon after startup. See the teamd (8) man page for other
options.
94
⁠Chapt er 5. Configure Net work T eaming
To check the status of the team, issue the following command as ro o t:
~]# teamd ctl team0 state
setup:
runner: activebackup
ports:
em1
link watches:
link summary: up
instance[link_watch_0]:
name: ethtool
link: up
em2
link watches:
link summary: up
instance[link_watch_0]:
name: ethtool
link: up
runner:
active port: em1
To apply an address to the network team interface, team0, issue a command as ro o t in the following
format:
~]# i p ad d r ad d 19 2. 16 8. 23. 2/24 d ev team0
To check the IP address of a team interface, issue a command as follows:
~]$ i p ad d r sho w team0
4: team0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue state
UP
link/ether 16:38:57:60:20:6f brd ff:ff:ff:ff:ff:ff
inet 192.168.23.2/24 scope global team0
valid_lft forever preferred_lft forever
inet6 2620:52:0:221d:1438:57ff:fe60:206f/64 scope global dynamic
valid_lft 2591880sec preferred_lft 604680sec
inet6 fe80::1438:57ff:fe60:206f/64 scope link
valid_lft forever preferred_lft forever
To activate the team interface, or to bring it “ up” , issue a command as ro o t in the following format:
~]# i p l i nk set d ev team0 up
To temporarily deactivate the team interface, or to take it “ down” , issue a command as ro o t in the
following format:
~]# i p l i nk set d ev team0 d o wn
To terminate, or kill, an instance of the team daemon, as ro o t user, issue a command in the
following format:
~]# teamd -t team0 -k
95
Red Hat Ent erprise Linux 7 Net working G uide
The -k option is to specify that the instance of the daemon associated with the device team0 is to be
killed. See the teamd (8) man page for other options.
For help on command-line options for teamd , issue the following command:
~]$ teamd -h
In addition, see the teamd (8) man page.
5.10.3. Creat ing a Net work T eam Using ifcfg Files
To create a networking team using i fcfg files, create a file in the /etc/sysco nfi g /netwo rkscri pts/ directory as follows:
DEVICE=team0
DEVICETYPE=Team
ONBOOT=yes
BOOTPROTO=none
IPADDR=192.168.11.1
PREFIX=24
TEAM_CONFIG='{"runner": {"name": "activebackup"}, "link_watch": {"name":
"ethtool"}}'
This creates the interface to the team, in other words, this is the master.
To create a port to be a member of team0, create one or more files in the
/etc/sysco nfi g /netwo rk-scri pts/ directory as follows:
DEVICE=eth1
HWADDR=D4:85:64:01:46:9E
DEVICETYPE=TeamPort
ONBOOT=yes
TEAM_MASTER=team0
TEAM_PORT_CONFIG='{"prio": 100}'
Add additional port interfaces similar to the above as required, changing the D EVICE and HWAD D R
field to match the ports (the network devices) being added. If port priority is not specified by pri o it
defaults to 0 ; it accepts negative and positive values in the range -32,76 7 to + 32,76 7.
Specifying the hardware or MAC address using the HWAD D R directive will influence the device
naming procedure. This is explained in Chapter 8, Consistent Network Device Naming.
To bring up the network team, issue the following command as ro o t:
~]# i fup team0
To view the network team, issue the following command:
~]$ i p l i nk sho w
5.10.4 . Add a Port t o a Net work T eam Using iput ils
To add a port em1 to a network team team0, using the ip utility, issue the following commands as
ro o t:
96
⁠Chapt er 5. Configure Net work T eaming
~]# i p l i nk set d ev em1 d o wn
~]# i p l i nk set d ev em1 master team0
Add additional ports as required. Team driver will bring ports up automatically.
5.10.5. List ing t he port s of a T eam Using t eamnl
To view or list the ports in a network team, using the t eamn l utility, issue the following command as
ro o t:
~]# teamnl team0 po rts
em2: up 100 fullduplex
em1: up 100 fullduplex
5.10.6. Configuring Opt ions of a T eam Using t eamnl
To view or list all currently available options, using the t eamn l utility, issue the following command
as ro o t:
~]# teamnl team0 o pti o ns
To configure a team to use active backup mode, issue the following command as ro o t:
~]# teamnl team0 seto pti o n mo d e acti vebackup
5.10.7. Add an Address t o a Net work T eam Using iput ils
To add an address to a team team0, using the ip utility, issue the following command as ro o t:
~]# i p ad d r ad d 19 2. 16 8. 252. 2/24 d ev team0
5.10.8. Bring up an Int erface t o a Net work T eam Using iput ils
To activate or “ bring up” an interface to a network team, team0, using the ip utility, issue the
following command as ro o t:
~]# i p l i nk set team0 up
5.10.9. Viewing t he Act ive Port Opt ions of a T eam Using t eamnl
To view or list the acti vepo rt option in a network team, using the t eamn l utility, issue the following
command as ro o t:
~]# teamnl team0 g eto pti o n acti vepo rt
0
5.10.10. Set t ing t he Act ive Port Opt ions of a T eam Using t eamnl
97
Red Hat Ent erprise Linux 7 Net working G uide
To set the acti vepo rt option in a network team, using the t eamn l utility, issue the following
command as ro o t:
~]# teamnl team0 seto pti o n acti vepo rt 5
To check the change in team port options, issue the following command as ro o t:
~]# teamnl team0 g eto pti o n acti vepo rt
5
5.11. Cont rolling t eamd wit h t eamdct l
In order to query a running instance of teamd for statistics or configuration information, or to make
changes, the control tool t eamd ct l is used.
To view the current team state of a team team0, enter the following command as ro o t:
~]# teamd ctl team0 state vi ew
For a more verbose output:
~]# teamd ctl team0 state vi ew -v
For a complete state dump in JSON format (useful for machine processing) of team0, use the
following command:
~]# teamd ctl team0 state d ump
For a configuration dump in JSON format of team0, use the following command:
~]# teamd ctl team0 co nfi g d ump
To view the configuration of a port em1, that is part of a team team0, enter the following command:
~]# teamd ctl team0 po rt co nfi g d ump em1
5.11.1. Add a Port t o a Net work T eam
To add a port em1 to a network team team0, issue the following command as ro o t:
~]# teamd ctl team0 po rt ad d em1
5.11.2. Remove a Port From a Net work T eam
To remove an interface em1 from a network team team0, issue the following command as ro o t:
~]# teamd ctl team0 po rt remo ve em1
5.11.3. Apply a Configurat ion t o a Port in a Net work T eam
98
⁠Chapt er 5. Configure Net work T eaming
To apply a JSON format configuration to a port em1 in a network team team0, issue a command as
ro o t in the following format:
~]# teamd ctl team0 po rt co nfi g upd ate em1 JSON-config-string
Where JSON-config-string is the configuration as a string of text in JSON format. This will update the
configuration of the port using the JSON format string supplied. An example of a valid JSON string
for configuring a port would be the following:
{
"prio": -10,
"sticky": true
}
Use single quotes around the JSON configuration string and omit the line breaks.
Note that the old configuration will be overwritten and that any options omitted will be reset to the
default values. See the teamd ctl (8) man page for more team daemon control tool command
examples.
5.11.4 . View t he Configurat ion of a Port in a Net work T eam
To copy the configuration of a port em1 in a network team team0, issue the following command as
ro o t:
~]# teamd ctl team0 po rt co nfi g d ump em1
This will dump the JSON format configuration of the port to standard output.
5.12. Configure t eamd Runners
Runners are units of code which are compiled into the Team daemon when an instance of the
daemon is created. For an introduction to the teamd runners, see Section 5.4, “ Understanding the
Network Teaming D aemon and the " Runners" ” .
5.12.1. Configure t he broadcast Runner
To configure the broadcast runner, using an editor as ro o t, add the following to the team JSON
format configuration file:
{
"device": "team0",
"runner": {"name": "broadcast"},
"ports": {"em1": {}, "em2": {}}
}
Please see the teamd . co nf(5) man page for more information.
5.12.2. Configure t he random Runner
The random runner behaves similarly to the roundrobin runner.
99
Red Hat Ent erprise Linux 7 Net working G uide
To configure the random runner, using an editor as ro o t, add the following to the team JSON format
configuration file:
{
"device": "team0",
"runner": {"name": "random"},
"ports": {"em1": {}, "em2": {}}
}
Please see the teamd . co nf(5) man page for more information.
5.12.3. Configure t he roundrobin Runner
To configure the roundrobin runner, using an editor as ro o t, add the following to the team JSON
format configuration file:
{
"device": "team0",
"runner": {"name": "roundrobin"},
"ports": {"em1": {}, "em2": {}}
}
A very basic configuration for roundrobin.
Please see the teamd . co nf(5) man page for more information.
5.12.4 . Configure t he act ivebackup Runner
The active backup runner can use all of the link-watchers to determine the status of links in a team.
Any one of the following examples can be added to the team JSON format configuration file:
{
"device": "team0",
"runner": {
"name": "activebackup"
},
"link_watch": {
"name": "ethtool"
},
"ports": {
"em1": {
"prio": -10,
"sticky": true
},
"em2": {
"prio": 100
}
}
}
This example configuration uses the active-backup runner with et h t o o l as the link watcher. Port
em2 has higher priority. The sticky flag ensures that if em1 becomes active, it stays active as long as
the link remains up.
100
⁠Chapt er 5. Configure Net work T eaming
{
"device": "team0",
"runner": {
"name": "activebackup"
},
"link_watch": {
"name": "ethtool"
},
"ports": {
"em1": {
"prio": -10,
"sticky": true,
"queue_id": 4
},
"em2": {
"prio": 100
}
}
}
This example configuration adds a queue ID of 4 . It uses active-backup runner with et h t o o l as the
link watcher. Port em2 has higher priority. But the sticky flag ensures that if em1 becomes active, it
will stay active as long as the link remains up.
To configure the activebackup runner using et h t o o l as the link watcher and applying a delay,
using an editor as ro o t, add the following to the team JSON format configuration file:
{
"device": "team0",
"runner": {
"name": "activebackup"
},
"link_watch": {
"name": "ethtool",
"delay_up": 2500,
"delay_down": 1000
},
"ports": {
"em1": {
"prio": -10,
"sticky": true
},
"em2": {
"prio": 100
}
}
}
This example configuration uses the active-backup runner with et h t o o l as the link watcher. Port
em2 has higher priority. But the sticky flag ensures that if em1 becomes active, it stays active while
the link remains up. Link changes are not propagated to the runner immediately, but delays are
applied.
Please see the teamd . co nf(5) man page for more information.
101
Red Hat Ent erprise Linux 7 Net working G uide
5.12.5. Configure t he loadbalance Runner
This runner can be used for two types of load balancing, active and passive. In active mode,
constant re-balancing of traffic is done by using statistics of recent traffic to share out traffic as
evenly as possible. In static mode, streams of traffic are distributed randomly across the available
links. This has a speed advantage due to lower processing overhead. In high volume traffic
applications this is often preferred as traffic usually consists of multiple stream which will be
distributed randomly between the available links, in his way load sharing is accomplished without
intervention by teamd .
To configure the loadbalance runner for passive transmit (Tx) load balancing, using an editor as
ro o t, add the following to the team JSON format configuration file:
{
"device": "team0",
"runner": {
"name": "loadbalance",
"tx_hash": ["eth", "ipv4", "ipv6"]
},
"ports": {"em1": {}, "em2": {}}
}
Configuration for hash-based passive transmit (Tx) load balancing.
To configure the loadbalance runner for active transmit (Tx) load balancing, using an editor as
ro o t, add the following to the team JSON format configuration file:
{
"device": "team0",
"runner": {
"name": "loadbalance",
"tx_hash": ["eth", "ipv4", "ipv6"],
"tx_balancer": {
"name": "basic"
}
},
"ports": {"em1": {}, "em2": {}}
}
Configuration for active transmit (Tx) load balancing using basic load balancer.
Please see the teamd . co nf(5) man page for more information.
5.12.6. Configure t he LACP (802.3ad) Runner
To configure the LACP runner using et h t o o l as a link watcher, using an editor as ro o t, add the
following to the team JSON format configuration file:
{
"device": "team0",
"runner": {
"name": "lacp",
"active": true,
"fast_rate": true,
"tx_hash": ["eth", "ipv4", "ipv6"]
102
⁠Chapt er 5. Configure Net work T eaming
},
"link_watch": {"name": "ethtool"},
"ports": {"em1": {}, "em2": {}}
}
Configuration for connection to a link aggregation control protocol (LACP) capable counterpart. The
LACP runner should use et h t o o l to monitor the status of a link. It does not make sense to use any
other link monitoring method besides the et h t o o l because, for example in the case of arp _p in g , the
link would never come up. The reason is that the link has to be established first and only after that
can packets, ARP included, go through. Using et h t o o l prevents this because it monitors each link
layer individually.
Active load balancing is possible with this runner in the same way as it is done for the loadbalance
runner. To enable active transmit (Tx) load balancing, add the following section:
"tx_balancer": {
"name": "basic"
}
Please see the teamd . co nf(5) man page for more information.
5.12.7. Configure Monit oring of t he Link St at e
The following methods of link state monitoring are available. To implement one of the methods, add
the JSON format string to the team JSON format configuration file using an editor running with ro o t
privileges.
5 .1 2 .7 .1 . Co nfigure Et ht o o l fo r link-st at e Mo nit o ring
To add or edit an existing delay, in milliseconds, between the link coming up and the runner being
notified about it, add or edit a section as follows:
"link_watch": {
"name": "ethtool",
"delay_up": 2500
}
To add or edit an existing delay, in milliseconds, between the link going down and the runner being
notified about it, add or edit a section as follows:
"link_watch": {
"name": "ethtool",
"delay_down": 1000
}
5 .1 2 .7 .2 . Co nfigure ARP Ping fo r Link-st at e Mo nit o ring
The team daemon teamd sends an ARP REQUEST to an address at the remote end of the link in
order to determine if the link is up. The method used is the same as the arp in g utility but it does not
use that utility.
Prepare a file containing the new configuration in JSON format similar to the following example:
{
103
Red Hat Ent erprise Linux 7 Net working G uide
"device": "team0",
"runner": {"name": "activebackup"},
"link_watch":{
"name": "arp_ping",
"interval": 100,
"missed_max": 30,
"source_host": "192.168.23.2",
"target_host": "192.168.23.1"
},
"ports": {
"em1": {
"prio": -10,
"sticky": true
},
"em2": {
"prio": 100
}
}
}
This configuration uses arp _p in g as the link watcher. The mi ssed _max option is a limit value of the
maximum allowed number of missed replies (ARP replies for example). It should be chosen in
conjunction with the i nterval option in order to determine the total time before a link is reported as
down.
To load a new configuration for a team port em2, from a file containing a JSON configuration, issue
the following command as ro o t:
~]# port config update em2 JSON-config-file
Note that the old configuration will be overwritten and that any options omitted will be reset to the
default values. See the teamd ctl (8) man page for more team daemon control tool command
examples.
5 .1 2 .7 .3. Co nfigure IPv6 NA/NS fo r Link-st at e Mo nit o ring
{
"device": "team0",
"runner": {"name": "activebackup"},
"link_watch": {
"name": "nsna_ping",
"interval": 200,
"missed_max": 15,
"target_host": "fe80::210:18ff:feaa:bbcc"
},
"ports": {
"em1": {
"prio": -10,
"sticky": true
},
"em2": {
"prio": 100
}
}
}
104
⁠Chapt er 5. Configure Net work T eaming
To configure the interval between sending NS/NA packets, add or edit a section as follows:
"link_watch": {
"name": "nsna_ping",
"interval": 200
}
Value is positive number in milliseconds. It should be chosen in conjunction with the mi ssed _max
option in order to determine the total time before a link is reported as down.
To configure the maximum number of missed NS/NA reply packets to allow before reporting the link
as down, add or edit a section as follows:
"link_watch": {
"name": "nsna_ping",
"missed_max": 15
}
Maximum number of missed NS/NA reply packets. If this number is exceeded, the link is reported as
down. The mi ssed _max option is a limit value of the maximum allowed number of missed replies
(ARP replies for example). It should be chosen in conjunction with the i nterval option in order to
determine the total time before a link is reported as down.
To configure the host name that is resolved to the IP v6 address target address for the NS/NA
packets, add or edit a section as follows:
"link_watch": {
"name": "nsna_ping",
"target_host": "MyStorage"
}
The “ target_host” option contains the host name to be converted to an IP v6 address which will be
used as the target address for the NS/NA packets. An IP v6 address can be used in place of a host
name.
Please see the teamd . co nf(5) man page for more information.
5.12.8. Configure Port Select ion Override
The physical port which transmits a frame is normally selected by the kernel part of the team driver,
and is not relevant to the user or system administrator. The output port is selected using the policies
of the selected team mode (teamd runner). On occasion however, it is helpful to direct certain classes
of outgoing traffic to certain physical interfaces to implement slightly more complex policies. By
default the team driver is multiqueue aware and 16 queues are created when the driver initializes. If
more or less queues are desired, the Netlink attribute tx_q ueues can be used to change this value
during the team driver instance creation.
The queue ID for a port can be set by the port configuration option q ueue_i d as follows:
{
"queue_id": 3
}
105
Red Hat Ent erprise Linux 7 Net working G uide
These queue ID 's can be used in conjunction with the t c utility to configure a multiqueue queue
discipline and filters to bias certain traffic to be transmitted on certain port devices. For example, if
using the above configuration and wanting to force all traffic bound to 19 2. 16 8. 1. 10 0 to use eth1
in the team as its output device, issue commands as ro o t in the following format:
~]# tc q d i sc ad d d ev team0 hand l e 1 ro o t mul ti q
~]# tc fi l ter ad d d ev team0 pro to co l i p parent 1: pri o 1 u32 match i p
d st \
19 2. 16 8. 1. 10 0 acti o n skbed i t q ueue_mappi ng 3
This mechanism of overriding runner selection logic in order to bind traffic to a specific port can be
used with all runners.
5.12.9. Configure BPF-based T x Port Select ors
The loadbalance and LACP runners uses hashes of packets to sort network traffic flow. The hash
computation mechanism is based on the Berkeley Packet Filter (BPF) code. The BPF code is used to
generate a hash rather than make a policy decision for outgoing packets. The hash length is 8 bits
giving 256 variants. This means many different socket buffers (SKB) can have the same hash and
therefore pass traffic over the same link. The use of a short hash is a quick way to sort traffic into
different streams for the purposes of load balancing across multiple links. In static mode, the hash is
only used to decide out of which port the traffic should be sent. In active mode, the runner will
continually reassign hashes to different ports in an attempt to reach a perfect balance.
The following fragment types or strings can be used for packet Tx hash computation:
eth — Uses source and destination MAC addresses.
vl an — Uses VLAN ID .
i pv4 — Uses source and destination IP v4 addresses.
i pv6 — Uses source and destination IP v6 addresses.
i p — Uses source and destination IP v4 and IP v6 addresses.
l 3 — Uses source and destination IP v4 and IP v6 addresses.
tcp — Uses source and destination T C P ports.
ud p — Uses source and destination UD P ports.
sctp — Uses source and destination SC T P ports.
l 4 — Uses source and destination T C P and UD P and SC T P ports.
These strings can be used by adding a line in the following format to the load balance runner:
"tx_hash": ["eth", "ipv4", "ipv6"]
See Section 5.12.5, “ Configure the loadbalance Runner” for an example.
5.13. Creat ing a Net work T eam Using a GUI
5.13.1. Est ablishing a T eam Connect ion
106
⁠Chapt er 5. Configure Net work T eaming
You can use the GNOME co n t ro l- cen t er utility to direct N et wo rkMan ag er to create a team from
two or more Wired or InfiniBand connections. It is not necessary to create the connections to be
teamed first. They can be configured as part of the process to configure the team. You must have the
MAC addresses of the interfaces available in order to complete the configuration process.
Pro ced u re 5.1. Ad d in g a N ew T eam C o n n ect io n
Follow the below steps to add a new team connection.
1. Press the Super key to enter the Activities Overview, type co ntro l netwo rk and then press
Enter. The N et wo rk settings tool appears. This step is fully covered in Section 2.5, “ Using
NetworkManager with the GNOME Graphical User Interface” .
2. Click the plus symbol to open the selection list. Select T eam. The Ed i ti ng T eam
C o nnecti o n 1 window appears.
107
Red Hat Ent erprise Linux 7 Net working G uide
Fig u re 5.6 . T h e N et wo rkMan ag er G rap h ical U ser In t erf ace Ad d a men u
3. On the T eam tab, click Ad d and select the type of interface you want to use with the team
connection. Click the C reate button. Note that the dialog to select the port type only comes
up when you create the first port; after that, it will automatically use that same type for all
further ports.
4. The Ed i ti ng team0 po rt 1 window appears. Fill in the MAC address of the first interface
to be added to the team.
108
⁠Chapt er 5. Configure Net work T eaming
Fig u re 5.7. T h e N et wo rkMan ag er G rap h ical U ser In t erf ace Ad d a Slave
C o n n ect io n
5. If custom port settings are to be applied, click on the T eam P o rt tab and enter a JSON
configuration string or import it from a file.
6. Click the Save button.
7. The name of the teamed port appears in the T eamed co nnecti o ns window. Click the Ad d
button to add further port connections.
8. Review and confirm the settings and then click the Save button.
9. Edit the team-specific settings by referring to Section 5.13.1.1, “ Configuring the Team Tab”
below.
Pro ced u re 5.2. Ed it in g an Exist in g T eam C o n n ect io n
Follow the below steps to edit an existing team connection.
1. Press the Super key to enter the Activities Overview, type co ntro l netwo rk and then press
Enter. The N et wo rk settings tool appears.
2. Select the connection you want to edit and click the O pti o ns button.
3. Select the G eneral tab.
4. Configure the connection name, auto-connect behavior, and availability settings.
109
Red Hat Ent erprise Linux 7 Net working G uide
Five settings in the Ed i ti ng dialog are common to all connection types, see the G eneral
tab:
C o nnecti o n name — Enter a descriptive name for your network connection. This name
will be used to list this connection in the menu of the Netwo rk window.
Auto mati cal l y co nnect to thi s netwo rk when i t i s avai l abl e — Select
this box if you want N et wo rkMan ag er to auto-connect to this connection when it is
available. See Section 2.5.3, “ Connecting to a Network Automatically” for more
information.
Al l users may co nnect to thi s netwo rk — Select this box to create a connection
available to all users on the system. Changing this setting may require root privileges. See
Section 2.5.4, “ System-wide and Private Connection Profiles” for details.
Auto mati cal l y co nnect to VP N when usi ng thi s co nnecti o n — Select this
box if you want N et wo rkMan ag er to auto-connect to a VPN connection when it is
available. Select the VPN from the drop-down menu.
Fi rewal l Zo ne — Select the firewall zone from the drop-down menu. See the Red Hat
Enterprise Linux 7 Security Guide for more information on firewall zones.
5. Edit the team-specific settings by referring to Section 5.13.1.1, “ Configuring the Team Tab”
below.
Saving Yo ur Ne w (o r Mo difie d) Co nne ct io n and Making Furt he r Co nfigurat io ns
Once you have finished editing your team connection, click the Save button to save your customized
configuration. If the profile was in use while being edited, power cycle the connection to make
N et wo rkMan ag er apply the changes. If the profile is OFF, set it to ON or select it in the network
connection icon's menu. See Section 2.5.1, “ Connecting to a Network Using a GUI” for information
on using your new or altered connection.
You can further configure an existing connection by selecting it in the Netwo rk window and clicking
O pti o ns to return to the Ed i ti ng dialog.
Then, to configure:
IP v4 settings for the connection, click the IP v4 Setti ng s tab and proceed to Section 2.5.10.4,
“ Configuring IPv4 Settings” ; or,
IPv6 settings for the connection, click the IP v6 Setti ng s tab and proceed to Section 2.5.10.5,
“ Configuring IPv6 Settings” .
Once saved the Team will appear in the Network settings tool.
110
⁠Chapt er 5. Configure Net work T eaming
Fig u re 5.8. T h e N et wo rkMan ag er G rap h ical U ser In t erf ace wit h T eam
5 .1 3.1 .1 . Co nfiguring t he T e am T ab
If you have already added a new team connection you can enter a custom JSON configuration string
in the text box or import a configuration file. Click Save to apply the JSON configuration to the team
interface.
For examples of JSON strings, see Section 5.12, “ Configure teamd Runners”
See Procedure 5.1, “ Adding a New Team Connection” for instructions on how to add a new team.
5.14 . Addit ional Resources
The following sources of information provide additional resources regarding network teaming.
5.14 .1. Inst alled Document at ion
teamd (8) man page — D escribes the teamd service.
teamd ctl (8) man page — D escribes the teamd control tool.
teamd . co nf(5) man page — D escribes the teamd configuration file.
teamnl (8) man page — D escribes the teamd Netlink library.
bo nd 2team(1) man page — D escribes a tool to convert bonding options to team.
5.14 .2. Online Document at ion
h t t p ://www.w3sch o o ls.co m/jso n /jso n _syn t ax.asp
An explanation of JSON syntax.
111
Red Hat Ent erprise Linux 7 Net working G uide
Chapter 6. Configure Network Bridging
A network bridge is a link-layer device which forwards traffic between networks based on MAC
addresses. It makes forwarding decisions based on a table of MAC addresses which it builds by
listening to network traffic and thereby learning what hosts are connected to each network. A software
bridge can be used within a Linux host in order to emulate a hardware bridge, for example in
virtualization applications for sharing a NIC with one or more virtual NICs.
Note that a bridge cannot be established over Wi-Fi networks operating in Ad-Hoc or Infrastructure
modes. This is due to the IEEE 802.11 standard that specifies the use of 3-address frames in Wi-Fi for
the efficient use of airtime.
6.1. Configure Bridging Using t he T ext User Int erface, nmt ui
The text user interface tool n mt u i can be used to configure bridging in a terminal window. Issue the
following command to start the tool:
~]$ nmtui
The text user interface appears. Any invalid command prints a usage message.
To navigate, use the arrow keys or press T ab to step forwards and press Shi ft+T ab to step back
through the options. Press Enter to select an option. The Space bar toggles the status of a check
box.
1. From the starting menu, select Ed i t a co nnecti o n. Select Ad d , the New C o nnecti o n
screen opens.
112
⁠Chapt er 6 . Configure Net work Bridging
Fig u re 6 .1. T h e N et wo rkMan ag er T ext U ser In t erf ace Ad d a B rid g e C o n n ect io n
men u
2. Select Bri d g e, the Ed i t co nnecti o n screen opens.
3. To add slave interfaces to the bridge select Ad d , the New C o nnecti o n screen opens. Once
the type of Connection has been chosen select the C reate button to cause the bridge's Ed i t
C o nnecti o n display to appear.
Fig u re 6 .2. T h e N et wo rkMan ag er T ext U ser In t erf ace Ad d in g a n ew B rid g e Slave
C o n n ect io n men u
4. Enter the desired slave's device name or MAC address in the D evi ce section. If required,
enter a clone MAC address to be used as the bridge's MAC address by selecting Sho w to the
right of the Ethernet label. Select the O K button.
113
Red Hat Ent erprise Linux 7 Net working G uide
Note
If the device is specified without a MAC address the D evi ce section will be
automatically populated once the Ed i t C o nnecti o n window is reloaded, but only if
it successfully finds the device.
Fig u re 6 .3. T h e N et wo rkMan ag er T ext U ser In t erf ace C o n f ig u rin g a B rid g e Slave
C o n n ect io n men u
5. The name of the bridge slave appears in the Sl aves section. Repeat the above steps to add
further slave connections.
6. Review and confirm the settings before selecting the O K button.
114
⁠Chapt er 6 . Configure Net work Bridging
Fig u re 6 .4 . T h e N et wo rkMan ag er T ext U ser In t erf ace C o n f ig u rin g a B rid g e men u
See Section 6.4.1.1, “ Configuring the Bridge Tab” for definitions of the bridge terms.
See Section 1.5, “ Network Configuration Using a Text User Interface (nmtui)” for information on
installing n mt u i.
6.2. Using t he Net workManager Command Line T ool, nmcli
To create a bridge, named bridge-br0, issue a command as follows as ro o t:
~]# nmcl i co n ad d type bri d g e i fname br0
Connection 'bridge-br0' (6ad5bba6-98a0-4f20-839d-c997ba7668ad)
successfully added.
If no interface name is specified, the name will default to bridge, bridge-1, bridge-2, and so on.
To view the connections, issue the following command:
115
Red Hat Ent erprise Linux 7 Net working G uide
~]$ nmcl i co n sho w
NAME
UUID
TYPE
DEVICE
bridge-br0 79cf6a3e-0310-4a78-b759-bda1cc3eef8d bridge
br0
eth0
4d5c449a-a6c5-451c-8206-3c9a4ec88bca 802-3-ethernet eth0
Spanning tree protocol (STP) is enabled by default. The values used are from the IEEE 802.1D -1998
standard. To disable ST P for this bridge, issue a command as follows as ro o t:
~]# nmcl i co n mo d i fy bri d g e-br0 bri d g e. stp no
To re-enable 80 2. 1D ST P for this bridge, issue a command as follows as ro o t:
~]# nmcl i co n mo d i fy bri d g e-br0 bri d g e. stp yes
The default bridge priority for 80 2. 1D ST P is 3276 8. The lower number is preferred in root bridge
selection. For example, a bridge with priority of 286 72 would be selected as the root bridge in
preference to a bridge with priority value of 3276 8 (the default). To create a bridge with a non-default
value, issue a command as follows:
~]$ nmcl i co n ad d type bri d g e i fname br5 stp yes pri o ri ty 286 72
Connection 'bridge-br5' (86b83ad3-b466-4795-aeb6-4a66eb1856c7)
successfully added.
The allowed values are in the range 0 to 6 5535.
To change the bridge priority of an existing bridge to a non-default value, issue a command in the
following format:
~]$ nmcl i co nnecti o n mo d i fy bri d g e-br5 bri d g e. pri o ri ty 36 86 4
The allowed values are in the range 0 to 6 5535.
To view the bridge settings, issue the following command:
~]$ nmcl i -f bri d g e co n sho w bri d g e-br0
Further options for 80 2. 1D ST P are listed in the bridge section of the nmcl i (1) man page.
To add, or enslave an interface, for example eth1, to the bridge bridge-br0, issue a command as
follows:
~]$ nmcl i co n ad d type bri d g e-sl ave i fname eth1 master bri d g e-br0
Connection 'bridge-slave-eth1' (70ffae80-7428-4d9c-8cbd-2e35de72476e)
successfully added.
At time of writing, n mcli only supports Ethernet slaves.
To change a value using interactive mode, issue the following command:
~]$ nmcl i co nnecti o n ed i t bri d g e-br0
You will be placed at the n mcli prompt.
116
⁠Chapt er 6 . Configure Net work Bridging
nmcli> set bri d g e. pri o ri ty 4 0 9 6
nmcli> save
Connection 'bridge-br0' (79cf6a3e-0310-4a78-b759-bda1cc3eef8d)
successfully saved.
nmcli> q ui t
See Section 2.3, “ Using the NetworkManager Command Line Tool, nmcli” for an introduction to
n mcli.
6.3. Using t he Command Line Int erface (CLI)
6.3.1. Check if Bridging Kernel Module is Inst alled
In Red Hat Enterprise Linux 7, the bridging module is loaded by default. If necessary, you can make
sure that the module is loaded by issuing the following command as ro o t:
~]# mo d pro be --fi rst-ti me bri d g e
modprobe: ERROR: could not insert 'bridge': Module already in kernel
To display information about the module, issue the following command:
~]$ mo d i nfo bri d g e
See the mo d pro be(8) man page for more command options.
6.3.2. Creat e a Net work Bridge
To create a network bridge, create a file in the /etc/sysco nfi g /netwo rk-scri pts/ directory
called i fcfg -brN, replacing N with the number for the interface, such as 0 .
The contents of the file is similar to whatever type of interface is getting bridged to, such as an
Ethernet interface. The differences in this example are as follows:
The D EVIC E directive is given an interface name as its argument in the format brN, where N is
replaced with the number of the interface.
The T Y P E directive is given an argument Bri d g e. This directive determines the device type and
the argument is case sensitive.
The bridge interface configuration file is given an IP address whereas the physical interface
configuration file must only have a MAC address (see below).
An extra directive, D ELAY = 0 , is added to prevent the bridge from waiting while it monitors traffic,
learns where hosts are located, and builds a table of MAC addresses on which to base its filtering
decisions. The default delay of 15 seconds is not needed if no routing loops are possible.
Examp le 6 .1. Examp le if cf g - b r0 In t erf ace C o n f ig u rat io n File
The following is an example of a bridge interface configuration file using a static IP address:
DEVICE=br0
TYPE=Bridge
IPADDR=192.168.1.1
117
Red Hat Ent erprise Linux 7 Net working G uide
PREFIX=24
BOOTPROTO=none
ONBOOT=yes
DELAY=0
To complete the bridge another interface is created, or an existing interface is modified, and pointed
to the bridge interface.
Examp le 6 .2. Examp le if cf g - et h X In t erf ace C o n f ig u rat io n File
The following is an example of an Ethernet interface configuration file pointing to a bridge
interface. Configure your physical interface in /etc/sysco nfi g /netwo rk-scri pts/i fcfg ethX, where X is a unique number corresponding to a specific interface, as follows:
DEVICE=ethX
TYPE=Ethernet
HWADDR=AA:BB:CC:DD:EE:FF
BOOTPROTO=none
ONBOOT=yes
BRIDGE=br0
Optionally specify a name using the NAME directive. If no name is specified, the N et wo rkMan ag er
plug-in, i fcfg -rh, will create a name for the connection profile in the form “ Type Interface” . In this
example, this means the bridge will be named Bri d g e br0 . Alternately, if NAME= bri d g e-br0 is
added to the i fcfg -br0 file the connection profile will be named bri d g e-br0 .
Note
For the D EVIC E directive, almost any interface name could be used as it does not determine
the device type. T Y P E= Ethernet is not strictly required. If the T Y P E directive is not set, the
device is treated as an Ethernet device (unless its name explicitly matches a different interface
configuration file).
The directives are case sensitive.
Specifying the hardware or MAC address using the HWAD D R directive will influence the device
naming procedure as explained in Chapter 8, Consistent Network Device Naming.
Warning
If you are configuring bridging on a remote host, and you are connected to that host over the
physical NIC you are configuring, please consider the implications of losing connectivity
before proceeding. You will lose connectivity when restarting the service and may not be able
to regain connectivity if any errors have been made. Console, or out-of-band access is
advised.
To bring up the new or recently configured interfaces, issue a command as ro o t in the following
format:
118
⁠Chapt er 6 . Configure Net work Bridging
i fup device
This command will detect if N et wo rkMan ag er is running and call nmcl i co n l o ad UUID and
then call nmcl i co n up UUID.
Alternatively, to reload all interfaces, issue the following command as ro o t:
~]# systemctl restart netwo rk
This command will stop the network service, start the network service, and then call i fup for all ifcfg
files with O NBO O T = yes.
Note
The default behavior is for N et wo rkMan ag er not to be aware of changes to ifcfg files and to
continue using the old configuration data until the interface is next brought up. The is set by
the mo ni to r-co nnecti o n-fi l es option in the Netwo rkManag er. co nf file. See the
Netwo rkManag er. co nf(5) manual page for more information.
6.3.3. Net work Bridge wit h Bond
An example of a network bridge formed from two or more bonded Ethernet interfaces will now be given
as this is another common application in a virtualization environment. If you are not very familiar with
the configuration files for bonded interfaces then please refer to Section 4.4.2, “ Create a Channel
Bonding Interface”
Create or edit two or more Ethernet interface configuration files, which are to be bonded, as follows:
DEVICE=ethX
TYPE=Ethernet
SLAVE=yes
MASTER=bond0
BOOTPROTO=none
HWADDR=AA:BB:CC:DD:EE:FF
Note
Using ethX as the interface name is common practice but almost any name could be used.
Create or edit one interface configuration file, /etc/sysco nfi g /netwo rk-scri pts/i fcfg bo nd 0 , as follows:
DEVICE=bond0
ONBOOT=yes
BONDING_OPTS='mode=1 miimon=100'
BRIDGE=brbond0
For further instructions and advice on configuring the bonding module and to view the list of
bonding parameters, see Section 4.5, “ Using Channel Bonding” .
119
Red Hat Ent erprise Linux 7 Net working G uide
Create or edit one interface configuration file, /etc/sysco nfi g /netwo rk-scri pts/i fcfg brbo nd 0 , as follows:
DEVICE=brbond0
ONBOOT=yes
TYPE=Bridge
IPADDR=192.168.1.1
PREFIX=24
We now have two or more interface configuration files with the MAST ER = bo nd 0 directive. These point
to the configuration file named /etc/sysco nfi g /netwo rk-scri pts/i fcfg -bo nd 0 , which
contains the D EVIC E= bo nd 0 directive. This i fcfg -bo nd 0 in turn points to the
/etc/sysco nfi g /netwo rk-scri pts/i fcfg -brbo nd 0 configuration file, which contains the IP
address, and acts as an interface to the virtual networks inside the host.
To bring up the new or recently configured interfaces, issue a command as ro o t in the following
format:
i fup device
This command will detect if N et wo rkMan ag er is running and call nmcl i co n l o ad UUID and
then call nmcl i co n up UUID.
Alternatively, to reload all interfaces, issue the following command as ro o t:
~]# systemctl restart netwo rk
This command will stop the network service, start the network service, and then call i fup for all ifcfg
files with O NBO O T = yes.
Note
The default behavior is for N et wo rkMan ag er not to be aware of changes to ifcfg files and to
continue using the old configuration data until the interface is next brought up. The is set by
the mo ni to r-co nnecti o n-fi l es option in the Netwo rkManag er. co nf file. See the
Netwo rkManag er. co nf(5) manual page for more information.
6.4 . Configure Net work Bridging Using a GUI
When starting a bridge interface, N et wo rkMan ag er waits for at least one port to enter the
“ forwarding” state before beginning any network-dependent IP configuration such as D HC P or
IP v6 autoconfiguration. Static IP addressing is allowed to proceed before any slaves or ports are
connected or begin forwarding packets.
6.4 .1. Est ablishing a Bridge Connect ion
Pro ced u re 6 .1. Ad d in g a N ew B rid g e C o n n ect io n
Follow the below instructions to create a new bridge connection.
120
⁠Chapt er 6 . Configure Net work Bridging
1. To use the graphical N et wo rk settings tool, press the Super key to enter the Activities
Overview, type co ntro l netwo rk and then press Enter. The N et wo rk settings tool
appears. This steps is fully covered in Section 2.5, “ Using NetworkManager with the GNOME
Graphical User Interface” .
2. Select the plus symbol below the menu. The Ad d Netwo rk C o nnecti o n window appears.
3. Select the Bri d g e menu entry. The Ed i ti ng Bri d g e co nnecti o n 1 window appears.
Fig u re 6 .5. Ed it in g B rid g e C o n n ect io n 1
121
Red Hat Ent erprise Linux 7 Net working G uide
4. Add slave devices by referring to Procedure 6.3, “ Adding a Slave Interface to a Bridge”
below.
Pro ced u re 6 .2. Ed it in g an Exist in g B rid g e C o n n ect io n
You can configure an existing bridge connection by opening the Netwo rk window and selecting the
name of the connection from the list. Then click the Ed i t button.
1. Press the Super key to enter the Activities Overview, type co ntro l netwo rk and then press
Enter. The N et wo rk settings tool appears.
2. Select the Bri d g e connection you want to edit from the left hand menu.
3. Click the O pti o ns button.
Co nfiguring t he Co nne ct io n Nam e , Aut o -Co nne ct Be havio r, and Availabilit y
Se t t ings
Five settings in the Ed i ti ng dialog are common to all connection types, see the G eneral tab:
C o nnecti o n name — Enter a descriptive name for your network connection. This name will be
used to list this connection in the menu of the Netwo rk window.
Auto mati cal l y co nnect to thi s netwo rk when i t i s avai l abl e — Select this box if
you want N et wo rkMan ag er to auto-connect to this connection when it is available. See
Section 2.5.3, “ Connecting to a Network Automatically” for more information.
Al l users may co nnect to thi s netwo rk — Select this box to create a connection
available to all users on the system. Changing this setting may require root privileges. See
Section 2.5.4, “ System-wide and Private Connection Profiles” for details.
Auto mati cal l y co nnect to VP N when usi ng thi s co nnecti o n — Select this box if
you want N et wo rkMan ag er to auto-connect to a VPN connection when it is available. Select the
VPN from the dropdown menu.
Fi rewal l Zo ne — Select the Firewall Z one from the dropdown menu. See the Red Hat
Enterprise Linux 7 Security Guide for more information on Firewall Z ones.
6 .4 .1 .1 . Co nfiguring t he Bridge T ab
Interface name
The name of the interface to the bridge.
Bri d g ed co nnecti o ns
One or more slave interfaces.
Ag i ng ti me
The time, in seconds, a MAC address is kept in the MAC address forwarding database.
Enabl e ST P (Spanni ng T ree P ro to co l )
If required, select the check box to enable ST P .
P ri o ri ty
The bridge priority; the bridge with the lowest priority will be elected as the root bridge.
122
⁠Chapt er 6 . Configure Net work Bridging
Fo rward d el ay
The time, in seconds, spent in both the Listening and Learning states before entering the
Forwarding state. The default is 15 seconds.
Hel l o ti me
The time interval, in seconds, between sending configuration information in bridge protocol
data units (BPD U).
Max ag e
The maximum time, in seconds, to store the configuration information from BPD Us. This
value should be twice the Hello Time plus 1 but less than twice the Forwarding delay minus
1.
Pro ced u re 6 .3. Ad d in g a Slave In t erf ace t o a B rid g e
1. To add a port to a bridge, select the Bri d g e tab in the Ed i ti ng Bri d g e co nnecti o n 1
window. If necessary, open this window by following the procedure in Procedure 6.2, “ Editing
an Existing Bridge Connection” .
2. Click Ad d . The C ho o se a C o nnecti o n T ype menu appears.
3. Select the type of connection to be created from the list. Click C reate. A window appropriate
to the connection type selected appears.
123
Red Hat Ent erprise Linux 7 Net working G uide
Fig u re 6 .6 . T h e N et wo rkMan ag er G rap h ical U ser In t erf ace Ad d a B rid g e
C o n n ect io n
4. Select the Bri d g e P o rt tab. Configure P ri o ri ty and P ath co st as required. Note the
STP priority for a bridge port is limited by the Linux kernel. Although the standard allows a
range of 0 to 255, Linux only allows 0 to 6 3. The default is 32 in this case.
Fig u re 6 .7. T h e N et wo rkMan ag er G rap h ical U ser In t erf ace B rid g e Po rt t ab
5. If required, select the Hai rpi n mo d e check box to enable forwarding of frames for external
processing. Also known as virtual Ethernet port aggregator (VEPA) mode.
Then, to configure:
An Ethernet slave, click the Ethernet tab and proceed to Section 2.5.5.1, “ Configuring the
Connection Name, Auto-Connect Behavior, and Availability Settings” , or;
A Bond slave, click the Bo nd tab and proceed to Section 4.6.1.1, “ Configuring the Bond Tab” , or;
A Team slave, click the T eam tab and proceed to Section 5.13.1.1, “ Configuring the Team Tab” ,
or;
An VLAN slave, click the VLAN tab and proceed to Section 7.5.1.1, “ Configuring the VLAN Tab” , or;
Savin g Yo u r N ew ( o r Mo d if ied ) C o n n ect io n an d Makin g Fu rt h er C o n f ig u rat io n s
124
⁠Chapt er 6 . Configure Net work Bridging
Once you have finished editing your new bridge connection, click the Save button to save your
customized configuration. If the profile was in use while being edited, power cycle the connection to
make N et wo rkMan ag er apply the changes. If the profile is OFF, set it to ON or select it in the
network connection icon's menu. See Section 2.5.1, “ Connecting to a Network Using a GUI” for
information on using your new or altered connection.
You can further configure an existing connection by selecting it in the Netwo rk window and clicking
O pti o ns to return to the Ed i ti ng dialog.
Then, to configure:
IP v4 settings for the connection, click the IP v4 Setti ng s tab and proceed to Section 2.5.10.4,
“ Configuring IPv4 Settings” , or;
IP v6 settings for the connection, click the IP v6 Setti ng s tab and proceed to Section 2.5.10.5,
“ Configuring IPv6 Settings” .
Once saved the Bridge will appear in the Network settings tool with each slave showing in the
display.
Fig u re 6 .8. T h e N et wo rkMan ag er G rap h ical U ser In t erf ace wit h B rid g e
6.5. Addit ional Resources
The following sources of information provide additional resources regarding network bridging.
6.5.1. Inst alled Document at ion
nmcl i (1) man page — D escribes N et wo rkMan ag er's command‐line tool.
nmcl i -exampl es(5) man page — Gives examples of n mcli commands.
125
Red Hat Ent erprise Linux 7 Net working G uide
nm-setti ng s(5) man page — D escription of settings and parameters of N et wo rkMan ag er
connections.
126
⁠Chapt er 7 . Configure 8 0 2 .1 Q VLAN t agging
Chapter 7. Configure 802.1Q VLAN tagging
To create a VLAN, an interface is created on another interface referred to as the parent interface. The
VLAN interface will tag packets with the VLAN ID as they pass through the interface, and returning
packets will be untagged. VLAN interface can be configured similarly to any other interface. The
parent interface need not be an Ethernet interface. An 802.1Q VLAN tagging interface can be created
on bridge, bond, and team interfaces, however there are some things to note:
In the case of VLANs over bonds, it is important that the bond has slaves and that they are “ up”
before bringing up the VLAN interface. At the time of writing, adding a VLAN interface to a bond
without slaves does not work.
A VLAN slave cannot be configured on a bond with the fai l _o ver_mac= fo l l o w option,
because the VLAN virtual device cannot change its MAC address to match the parent's new MAC
address. In such a case, traffic would still be sent with the now incorrect source MAC address.
Sending VLAN tagged packets through a network switch requires configuration of the switch.
Refer to the documentation for the switch. For example, for Cisco switches the port must be
assigned to one VLAN or configured to be a trunk port to accept tagged packets for multiple
VLANs. Untagged packets can also be processed by a trunk port and processed as belonging to
the native VLAN, however this is a security risk and may have been disabled, or by default not
enabled, depending on the make of the switch.
Some older network interface cards, loopback interfaces, Wimax cards, and some InfiniBand
devices, are said to be VLAN challenged, meaning they cannot support VLANs. This is usually
because the devices cannot cope with VLAN headers and the larger MTU size associated with
tagged packets.
7.1. Select ing VLAN Int erface Configurat ion Met hods
T o co n f ig u re a VLAN in t erf ace u sin g N et wo rkMan ag er's text user interface tool, n mt u i,
proceed to Section 7.2, “ Configure 802.1Q VLAN tagging Using the Text User Interface, nmtui”
T o co n f ig u re a VLAN in t erf ace u sin g N et wo rkMan ag er's command-line tool, n mcli,
proceed to Section 7.3, “ Configure 802.1Q VLAN Tagging Using the Command Line Tool, nmcli”
T o co n f ig u re a n et wo rk in t erf ace man u ally, see Section 7.4, “ Configure 802.1Q VLAN
Tagging Using the Command Line” .
T o co n f ig u re a n et wo rk u sin g g rap h ical u ser in t erf ace t o o ls, proceed to Section 7.5,
“ Configure 802.1Q VLAN Tagging Using a GUI”
7.2. Configure 802.1Q VLAN t agging Using t he T ext User Int erface,
nmt ui
The text user interface tool n mt u i can be used to configure 802.1Q VLANs in a terminal window.
Issue the following command to start the tool:
~]$ nmtui
The text user interface appears. Any invalid command prints a usage message.
127
Red Hat Ent erprise Linux 7 Net working G uide
To navigate, use the arrow keys or press T ab to step forwards and press Shi ft+T ab to step back
through the options. Press Enter to select an option. The Space bar toggles the status of a check
box.
From the starting menu, select Ed i t a co nnecti o n. Select Ad d , the New C o nnecti o n screen
opens.
Fig u re 7.1. T h e N et wo rkMan ag er T ext U ser In t erf ace Ad d a VLAN C o n n ect io n men u
Select VLAN, the Ed i t co nnecti o n screen opens. Follow the on-screen prompts to complete the
configuration.
128
⁠Chapt er 7 . Configure 8 0 2 .1 Q VLAN t agging
Fig u re 7.2. T h e N et wo rkMan ag er T ext U ser In t erf ace C o n f ig u rin g a VLAN C o n n ect io n
men u
See Section 7.5.1.1, “ Configuring the VLAN Tab” for definitions of the VLAN terms.
See Section 1.5, “ Network Configuration Using a Text User Interface (nmtui)” for information on
installing n mt u i.
7.3. Configure 802.1Q VLAN T agging Using t he Command Line T ool,
nmcli
To view the available interfaces on the system, issue a command as follows:
~]$ nmcl i co n sho w
NAME
UUID
DEVICE
System eth1 9c92fad9-6ecb-3e6c-eb4d-8a47c6f50c04
System eth0 5fb06bd0-0bb0-7ffb-45f1-d6edd65f3e03
TYPE
802-3-ethernet
802-3-ethernet
eth1
eth0
Note that the NAME field in the output always denotes the connection ID . It is not the interface name
even though it might look the same. The ID can be used in nmcl i co nnecti o n commands to
identify a connection. Use the D EVICE name with other applications such as fi rewal l d .
To create an 802.1Q VLAN interface on Ethernet interface eth0, with VLAN interface VLAN10 and ID 10 ,
issue a command as follows:
~]$ nmcl i co n ad d type vl an i fname VLAN10 d ev eth0 i d 10
Connection 'vlan-VLAN10' (37750b4a-8ef5-40e6-be9b-4fb21a4b6d17)
successfully added.
129
Red Hat Ent erprise Linux 7 Net working G uide
Note that as no co n-name was given for the VLAN interface, the name was derived from the interface
name by prepending the type. Alternatively, specify a name with the co n-name option as follows:
~]$ nmcl i co n ad d type vl an co n-name VLAN12 d ev eth0 i d 12
Connection 'VLAN12' (b796c16a-9f5f-441c-835c-f594d40e6533) successfully
added.
Assigning Addresses t o VLAN Int erfaces
You can use the same n mcli commands to assign static and dynamic interface addresses as with
any other interface.
For example, a command to create a VLAN interface with a static IP v4 address and gateway is as
follows:
~]$ nmcl i co n ad d type vl an co n-name VLAN20 i fname eth0 i p4
10 . 10 . 10 . 10 /24 \
g w4 10 . 10 . 10 . 254
To create a VLAN interface with dynamically assigned addressing, issue a command as follows:
~]$ nmcl i co n ad d type vl an co n-name VLAN30 i fname eth0
See Section 2.3.2, “ Connecting to a Network Using nmcli” for examples of using n mcli commands to
configure interfaces.
To review the VLAN interfaces created, issue a command as follows:
~]$ nmcl i co n sho w
NAME
UUID
TYPE
DEVICE
VLAN12
4129a37d-4feb-4be5-ac17-14a193821755 vlan
eth0.12
System eth1 9c92fad9-6ecb-3e6c-eb4d-8a47c6f50c04 802-3-ethernet eth1
System eth0 5fb06bd0-0bb0-7ffb-45f1-d6edd65f3e03 802-3-ethernet eth0
vlan-VLAN10 1be91581-11c2-461a-b40d-893d42fed4f4 vlan
VLAN10
To view detailed information about the newly configured connection, issue a command as follows:
~]$ nmcl i -p co n sho w VLAN12
========================================================================
=======
Connection profile details (VLAN12)
========================================================================
=======
connection.id:
VLAN12
connection.uuid:
4129a37d-4feb-4be5-ac1714a193821755
connection.interface-name:
-connection.type:
vlan
connection.autoconnect:
yes
…
------------------------------------------------------------------------------
130
⁠Chapt er 7 . Configure 8 0 2 .1 Q VLAN t agging
802-3-ethernet.port:
-802-3-ethernet.speed:
0
802-3-ethernet.duplex:
-802-3-ethernet.auto-negotiate:
yes
802-3-ethernet.mac-address:
-802-3-ethernet.cloned-mac-address:
-802-3-ethernet.mac-address-blacklist:
802-3-ethernet.mtu:
auto
…
vlan.interface-name:
-vlan.parent:
eth0
vlan.id:
12
vlan.flags:
0 (NONE)
vlan.ingress-priority-map:
vlan.egress-priority-map:
-----------------------------------------------------------------------------========================================================================
=======
Activate connection details (4129a37d-4feb-4be5-ac17-14a193821755)
========================================================================
=======
GENERAL.NAME:
VLAN12
GENERAL.UUID:
4129a37d-4feb-4be5-ac1714a193821755
GENERAL.DEVICES:
eth0.12
GENERAL.STATE:
activating
[output truncated]
Further options for the VLAN command are listed in the VLAN section of the nmcl i (1) man page. In
the man pages the device on which the VLAN is created is referred to as the parent device. In the
example above the device was specified by its interface name, eth0 , it can also be specified by the
connection UUID or MAC address.
To create an 802.1Q VLAN connection profile with ingress priority mapping on Ethernet interface eth1,
with name VLAN1 and ID 13, issue a command as follows:
~]$ nmcl i co n ad d type vl an co n-name VLAN1 d ev eth2 i d 13 i ng ress
"2: 3,3: 5"
To view all the parameters associated with the VLAN created above, issue a command as follows:
~]$ nmcl i co nnecti o n sho w vlan-VLAN10
To change the MTU, issue a command as follows:
~]$ nmcl i co nnecti o n mo d i fy vlan-VLAN10 80 2. mtu 14 9 6
The MTU setting determines the maximum size of the network layer packet. The maximum size of the
payload the link-layer frame can carry in turn limits the network layer MTU. For standard Ethernet
frames this means an MTU of 1500 bytes. It should not be necessary to change the MTU when setting
up a VLAN as the link-layer header is increased in size by 4 bytes to accommodate the 802.1Q tag.
131
Red Hat Ent erprise Linux 7 Net working G uide
At time of writing, co nnecti o n. i nterface-name and vl an. i nterface-name have to be the
same (if they are set). They must therefore be changed simultaneously using n mcli's interactive
mode. To change a VLAN connections name, issue commands as follows:
~]$ nmcl i co n ed i t vlan-VLAN10
nmcli> set vl an. i nterface-name superVLAN
nmcli> set co nnecti o n. i nterface-name superVLAN
nmcli> save
nmcli> q ui t
The n mcli utility can be used to set and clear i o ctl flags which change the way the 802.1Q code
functions. The following VLAN flags are supported by N et wo rkMan ag er:
0x01 - reordering of output packet headers
0x02 - use GVRP protocol
0x04 - loose binding of the interface and its master
The state of the VLAN is synchronized to the state of the parent or master interface (the interface or
device on which the VLAN is created). If the parent interface is set to the “ down” administrative state
then all associated VLANs are set down and all routes are flushed from the routing table. Flag 0 x0 4
enables a loose binding mode, in which only the operational state is passed from the parent to the
associated VLANs, but the VLAN device state is not changed.
To set a VLAN flag, issue a command as follows:
~]$ nmcl i co nnecti o n mo d i fy vl an-VLAN10 vl an. fl ag s 1
See Section 2.3, “ Using the NetworkManager Command Line Tool, nmcli” for an introduction to
n mcli.
7.4 . Configure 802.1Q VLAN T agging Using t he Command Line
In Red Hat Enterprise Linux 7, the 80 21q module is loaded by default. If necessary, you can make
sure that the module is loaded by issuing the following command as ro o t:
~]# mo d pro be --fi rst-ti me 80 21q
modprobe: ERROR: could not insert '8021q': Module already in kernel
To display information about the module, issue the following command:
~]$ mo d i nfo 80 21q
See the mo d pro be(8) man page for more command options.
7.4 .1. Set t ing Up 802.1Q VLAN T agging Using ifcfg Files
1. Configure the parent interface in /etc/sysco nfi g /netwo rk-scri pts/i fcfg -ethX,
where X is a unique number corresponding to a specific interface, as follows:
132
⁠Chapt er 7 . Configure 8 0 2 .1 Q VLAN t agging
DEVICE=ethX
TYPE=Ethernet
BOOTPROTO=none
ONBOOT=yes
2. Configure the VLAN interface configuration in the /etc/sysco nfi g /netwo rk-scri pts/
directory. The configuration file name should be the parent interface plus a . character plus
the VLAN ID number. For example, if the VLAN ID is 192, and the parent interface is eth0, then
the configuration file name should be i fcfg -eth0 . 19 2:
DEVICE=ethX.192
BOOTPROTO=none
ONBOOT=yes
IPADDR=192.168.1.1
PREFIX=24
NETWORK=192.168.1.0
VLAN=yes
If there is a need to configure a second VLAN, with for example, VLAN ID 193, on the same
interface, eth0, add a new file with the name eth0 . 19 3 with the VLAN configuration details.
3. Restart the networking service in order for the changes to take effect. As ro o t issue the
following command:
~]# systemctl restart netwo rk
7.4 .2. Configure 802.1Q VLAN T agging Using ip Commands
To create an 802.1Q VLAN interface on Ethernet interface eth0, with name VLAN8 and ID 8, issue a
command as ro o t as follows:
~]# i p l i nk ad d l i nk eth0 name eth0.8 type vl an i d 8
To view the VLAN, issue the following command:
~]$ i p -d l i nk sho w eth0.8
4: eth0.8@ eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc noqueue
state UP mode DEFAULT
link/ether 52:54:00:ce:5f:6c brd ff:ff:ff:ff:ff:ff promiscuity 0
vlan protocol 802.1Q id 8 <REORDER_HDR>
Note that the ip utility interprets the VLAN ID as a hexadecimal value if it is preceded by 0 x and as an
octal value if it has a leading 0 . This means that in order to assign a VLAN ID with a decimal value of
22, you must not add any leading zeros.
To remove the VLAN, issue a command as ro o t as follows:
~]# i p l i nk d el ete eth0.8
133
Red Hat Ent erprise Linux 7 Net working G uide
Note
VLAN interfaces created using ip commands at the command prompt will be lost if the system
is shutdown or restarted. To configure VLAN interfaces to be persistent after a system restart,
use i fcfg files. See Section 7.4.1, “ Setting Up 802.1Q VLAN Tagging Using ifcfg Files”
7.5. Configure 802.1Q VLAN T agging Using a GUI
7.5.1. Est ablishing a VLAN Connect ion
You can use the GNOME co n t ro l- cen t er utility to direct N et wo rkMan ag er to create a VLAN using
an existing interface as the parent interface. At time of writing, you can only make VLANs on Ethernet
devices. Note that VLAN devices are only created automatically if the parent interface is set to connect
automatically.
Pro ced u re 7.1. Ad d in g a N ew VLAN C o n n ect io n
You can configure a VLAN connection by opening the Netwo rk window, clicking the plus symbol,
and selecting VLAN from the list.
1. Press the Super key to enter the Activities Overview, type co ntro l netwo rk and then press
Enter. The Netwo rk settings tool appears.
2. Click the plus symbol to open the selection list. Select VLAN. The Ed i ti ng VLAN
C o nnecti o n 1 window appears.
3. On the VLAN tab, select the parent interface from the drop-down list you want to use for the
VLAN connection.
4. Enter the VLAN ID
5. Enter a VLAN interface name. This is the name of the VLAN interface that will be created. For
example, eth0 . 1 or vl an2. (Normally this is either the parent interface name plus “ . ” and
the VLAN ID , or “ vl an” plus the VLAN ID .)
6. Review and confirm the settings and then click the Save button.
7. To edit the VLAN-specific settings see Section 7.5.1.1, “ Configuring the VLAN Tab” .
134
⁠Chapt er 7 . Configure 8 0 2 .1 Q VLAN t agging
Fig u re 7.3. Ad d in g a N ew VLAN C o n n ect io n
Pro ced u re 7.2. Ed it in g an Exist in g VLAN C o n n ect io n
Follow these steps to edit an existing VLAN connection.
1. Press the Super key to enter the Activities Overview, type co ntro l netwo rk and then press
Enter. The Netwo rk settings tool appears.
2. Select the connection you want to edit and click the O pti o ns button.
3. Select the G eneral tab.
4. Configure the connection name, auto-connect behavior, and availability settings.
These settings in the Ed i ti ng dialog are common to all connection types:
C o nnecti o n name — Enter a descriptive name for your network connection. This name
will be used to list this connection in the VLAN section of the Netwo rk window.
Auto mati cal l y co nnect to thi s netwo rk when i t i s avai l abl e — Select
this box if you want N et wo rkMan ag er to auto-connect to this connection when it is
available. Refer to Section 2.5.3, “ Connecting to a Network Automatically” for more
information.
Avai l abl e to al l users — Select this box to create a connection available to all
users on the system. Changing this setting may require root privileges. Refer to
Section 2.5.4, “ System-wide and Private Connection Profiles” for details.
5. To edit the VLAN-specific settings see Section 7.5.1.1, “ Configuring the VLAN Tab” .
Saving Yo ur Ne w (o r Mo difie d) Co nne ct io n and Making Furt he r Co nfigurat io ns
135
Red Hat Ent erprise Linux 7 Net working G uide
Once you have finished editing your VLAN connection, click the Save button to save your
customized configuration. If the profile was in use while being edited, power cycle the connection to
make N et wo rkMan ag er apply the changes. If the profile is OFF, set it to ON or select it in the
network connection icon's menu. See Section 2.5.1, “ Connecting to a Network Using a GUI” for
information on using your new or altered connection.
You can further configure an existing connection by selecting it in the Netwo rk window and clicking
O pti o ns to return to the Ed i ti ng dialog.
Then, to configure:
IPv4 settings for the connection, click the IP v4 Setti ng s tab and proceed to Section 2.5.10.4,
“ Configuring IPv4 Settings” .
7 .5 .1 .1 . Co nfiguring t he VLAN T ab
If you have already added a new VLAN connection (refer to Procedure 7.1, “ Adding a New VLAN
Connection” for instructions), you can edit the VLAN tab to set the parent interface and the VLAN ID .
P arent Interface
A previously configured interface can be selected in the drop-down list.
VLAN ID
The identification number to be used to tag the VLAN network traffic.
VLAN i nterface name
The name of the VLAN interface that will be created. For example, eth0 . 1 or vl an2.
C l o ned MAC ad d ress
Optionally sets an alternate MAC address to use for identifying the VLAN interface. This can
be used to change the source MAC address for packets sent on this VLAN.
MT U
Optionally sets a Maximum Transmission Unit (MTU) size to be used for packets to be sent
over the VLAN connection.
7.6. Addit ional Resources
The following sources of information provide additional resources regarding Network Teaming.
7.6.1. Inst alled Document at ion
i p-l i nk(8) man page — D escribes the ip utility's network device configuration commands.
nmcl i (1) man page — D escribes N et wo rkMan ag er's command‐line tool.
nmcl i -exampl es(5) man page — Gives examples of n mcli commands.
nm-setti ng s(5) man page — D escription of settings and parameters of N et wo rkMan ag er
connections.
136
⁠Chapt er 8 . Consist ent Net work Device Naming
Chapter 8. Consistent Network Device Naming
Red Hat Enterprise Linux 7 provides methods for consistent and predictable network device naming
for network interfaces. These features change the name of network interfaces on a system in order to
make locating and differentiating the interfaces easier.
Traditionally, network interfaces in Linux are enumerated as eth[0 123… ], but these names do not
necessarily correspond to actual labels on the chassis. Modern server platforms with multiple
network adapters can encounter non-deterministic and counter-intuitive naming of these interfaces.
This affects both network adapters embedded on the motherboard (Lan-on-Motherboard, or LOM) and
add-in (single and multiport) adapters.
In Red Hat Enterprise Linux 7, u d ev supports a number of different naming schemes. The default is to
assign fixed names based on firmware, topology, and location information. This has the advantage
that the names are fully automatic, fully predictable, that they stay fixed even if hardware is added or
removed (no re-enumeration takes place), and that broken hardware can be replaced seamlessly.
The disadvantage is that they are sometimes harder to read than the eth0 or wlan0 names
traditionally used. For example: enp5s0.
8.1. Naming Schemes Hierarchy
By default, systemd will name interfaces using the following policy to apply the supported naming
schemes:
Sch eme 1: Names incorporating Firmware or BIOS provided index numbers for on-board
devices (example: eno 1), are applied if that information from the firmware or BIOS is applicable
and available, else falling back to scheme 2.
Sch eme 2: Names incorporating Firmware or BIOS provided PCI Express hotplug slot index
numbers (example: ens1) are applied if that information from the firmware or BIOS is applicable
and available, else falling back to scheme 3.
Sch eme 3: Names incorporating physical location of the connector of the hardware (example:
enp2s0 ), are applied if applicable, else falling directly back to scheme 5 in all other cases.
Sch eme 4 : Names incorporating interface's MAC address (example: enx78e7d 1ea4 6 d a), is not
used by default, but is available if the user chooses.
Sch eme 5: The traditional unpredictable kernel naming scheme, is used if all other methods fail
(example: eth0 ).
This policy, the procedure outlined above, is the default. If the system has b io sd evn ame enabled, it
will be used. Note that enabling b io sd evn ame requires passing bi o sd evname= 1 as a commandline parameter except in the case of a D ell system, where b io sd evn ame will be used by default as
long as it is installed. If the user has added u d ev rules which change the name of the kernel devices,
those rules will take precedence.
8.2. Underst anding t he Device Renaming Procedure
The device name procedure in detail is as follows:
1. A rule in /usr/l i b/ud ev/rul es. d /6 0 -net. rul es instructs the u d ev helper utility,
/lib /u d ev/ren ame_d evice, to look into all /etc/sysco nfi g /netwo rkscri pts/i fcfg -suffix files. If it finds an i fcfg file with a HWAD D R entry matching the
MAC address of an interface it renames the interface to the name given in the i fcfg file by
137
Red Hat Ent erprise Linux 7 Net working G uide
the D EVIC E directive.
2. A rule in /usr/l i b/ud ev/rul es. d /71-bi o sd evname. rul es instructs b io sd evn ame
to rename the interface according to its naming policy, provided that it was not renamed in a
previous step, b io sd evn ame is installed, and bi o sd evname= 0 was not given as a kernel
command on the boot command line.
3. A rule in /l i b/ud ev/rul es. d /75-net-d escri pti o n. rul es instructs u d ev to fill in the
internal u d ev device property values ID _NET_NAME_ONBOARD , ID _NET_NAME_SLOT,
ID _NET_NAME_PATH, ID _NET_NAME_MAC by examining the network interface device. Note,
that some device properties might be undefined.
4. A rule in /usr/l i b/ud ev/rul es. d /80 -net-name-sl o t. rul es instructs u d ev to
rename the interface, provided that it was not renamed in step 1 or 2, and the kernel
parameter net. i fnames= 0 was not given, according to the following priority:
ID _NET_NAME_ONBOARD , ID _NET_NAME_SLOT, ID _NET_NAME_PATH. It falls through to
the next in the list, if one is unset. If none of these are set, then the interface will not be
renamed.
Steps 3 and 4 are implementing the naming schemes 1, 2, 3, and optionally 4, described in
Section 8.1, “ Naming Schemes Hierarchy” . Step 2 is explained in more detail in Section 8.6,
“ Consistent Network D evice Naming Using biosdevname” .
8.3. Underst anding t he Predict able Net work Int erface Device Names
The names have two character prefixes based on the type of interface:
1. en for Ethernet,
2. wl for wireless LAN (WLAN),
3. ww for wireless wide area network (WWAN).
The names have the following types:
T ab le 8.1. D evice N ame T yp es
Fo rmat
D escrip t io n
o<index>
s<slot>[f<function>][d<dev_id>]
x<MAC>
p<bus>s<slot>[f<function>][d<dev_id>]
p<bus>s<slot>[f<function>][u<port>][..][c<config>][i<interface>]
on-board device index number
hotplug slot index number
MAC address
PCI geographical location
USB port number chain
All multi-function PCI devices will carry the [f<function>] number in the device name, including the
function 0 device.
For USB devices the full chain of port numbers of hubs is composed. If the name gets longer than
the maximum number of 15 characters, the name is not exported.
The USB configuration descriptors == 1 and USB interface descriptors == 0 values are
suppressed (configuration == 1 and interface == 0 are the default values if only one USB
configuration or interface exists).
8.4 . Naming Scheme for Net work Devices Available for Linux on
Syst em z
138
⁠Chapt er 8 . Consist ent Net work Device Naming
Syst em z
Use the bus-ID to create predictable device names for network interfaces in Linux on System z
instances. The bus-ID identifies a device in the s390 channel subsystem. A bus ID identifies the
device within the scope of a Linux instance. For a CCW device, the bus ID is the device's device
number with a leading 0 . n, where n is the subchannel set ID . For example, 0 . 1. 0 ab1.
Network interfaces of device type Ethernet are named as follows:
enccw0.0.1234
CTC network devices of device type SLIP are named as follows:
slccw0.0.1234
Use the znetco nf -c command or the l scss -a command to display available network devices
and their bus-ID s.
T ab le 8.2. D evice N ame T yp es f o r Lin u x o n Syst em z
Fo rmat
D escrip t io n
enccwbus-ID
slccwbus-ID
device type Ethernet
CTC network devices of device type
SLIP
8.5. Naming Scheme for VLAN Int erfaces
Traditionally, VLAN interface names in the format: interface-name.VLAN-ID are used. The VLAN-ID
ranges from 0 to 4 0 9 6 , which is a maximum of four characters and the total interface name has a
limit of 15 characters. The maximum interface name length is defined by the kernel headers and is a
global limit, affecting all applications.
In Red Hat Enterprise Linux 7, four naming conventions for VLAN interface names are supported:
VLAN p lu s VLAN ID
The word vl an plus the VLAN ID . For example: vlan0005
VLAN p lu s VLAN ID wit h o u t p ad d in g
The word vl an plus the VLAN ID with out padding by means of an additional two zeros.
For example: vlan5
D evice n ame p lu s VLAN ID
The name of the parent interface plus the VLAN ID . For example: eth0.0005
D evice n ame p lu s VLAN ID wit h o u t p ad d in g
The name of the parent interface plus the VLAN ID with out padding by means of an
additional two zeros. For example: eth0.05
8.6. Consist ent Net work Device Naming Using biosdevname
This feature, implemented via the b io sd evn ame u d ev helper utility, will change the name of all
139
Red Hat Ent erprise Linux 7 Net working G uide
embedded network interfaces, PCI card network interfaces, and virtual function network interfaces
from the existing eth[0 123… ] to the new naming convention as shown in Table 8.3, “ The
biosdevname Naming Convention” . Note that unless the system is a D ell system, or b io sd evn ame is
explicitly enabled as described in Section 8.6.2, “ Enabling and D isabling the Feature” , the systemd
naming scheme will take precedence.
T ab le 8.3. T h e b io sd evn ame N amin g C o n ven t io n
D evice
O ld N ame
N ew N ame
Embedded network interface
(LOM)
PCI card network interface
eth[0 123…
]
eth[0 123…
]
eth[0 123…
]
em[1234 … ] ⁠ [a]
Virtual function
p<slot>p<ethernet port> ⁠ [b ]
p<slot>p<ethernet port>_<virtual int
erface> ⁠ [c ]
[a] New enumeratio n s tarts at 1 .
[b ] Fo r examp le: p3p4
[c ] Fo r examp le: p3p4 _1
8.6.1. Syst em Requirement s
The b io sd evn ame program uses information from the system's BIOS, specifically the type 9 (System
Slot) and type 41 (Onboard D evices Extended Information) fields contained within the SMBIOS. If the
system's BIOS does not have SMBIOS version 2.6 or higher and this data, the new naming
convention will not be used. Most older hardware does not support this feature because of a lack of
BIOSes with the correct SMBIOS version and field information. For BIOS or SMBIOS version
information, contact your hardware vendor.
For this feature to take effect, the biosdevname package must also be installed. To install it, issue the
following command as ro o t:
~]# yum i nstal l bi o sd evname
8.6.2. Enabling and Disabling t he Feat ure
To disable this feature, pass the following option on the boot command line, both during and after
installation:
bi o sd evname= 0
To enable this feature, pass the following option on the boot command line, both during and after
installation:
bi o sd evname= 1
Unless the system meets the minimum requirements, this option will be ignored and the system will
use the systemd naming scheme as described in the beginning of the chapter.
If the bi o sd evname install option is specified, it must remain as a boot option for the lifetime of the
system.
8.7. Not es for Administ rat ors
14 0
⁠Chapt er 8 . Consist ent Net work Device Naming
Many system customization files can include network interface names, and thus will require updates
if moving a system from the old convention to the new convention. If you use the new naming
convention, you will also need to update network interface names in areas such as custom ip t ab les
rules, scripts altering i rq bal ance, and other similar configuration files. Also, enabling this change
for installation will require modification to existing kickst art files that use device names via the
ksd evi ce parameter; these kickst art files will need to be updated to use the network device's MAC
address or the network device's new name.
Note
The maximum interface name length is defined by the kernel headers and is a global limit,
affecting all applications.
8.8. Cont rolling t he Select ion of Net work Device Names
D evice naming can be controlled in the following manner:
B y id en t if yin g t h e n et wo rk in t erf ace d evice
Setting the MAC address in an i fcfg file using the HWAD D R directive enables it to be
identified by u d ev. The name will be taken from the string given by the D EVIC E directive,
which by convention is the same as the i fcfg suffix. For example, i fcfg -eth0.
B y t u rn in g o n o r o f f b io sd evn ame
The name provided by b io sd evn ame will be used (if b io sd evn ame can determine one).
B y t u rn in g o n o r o f f t h e systemd -ud ev n amin g sch eme
The name provided by systemd -ud ev will be used (if systemd -ud ev can determine one).
8.9. Disabling Consist ent Net work Device Naming
To disable consistent network device naming, choose from one of the following:
D isable the assignment of fixed names, so that the unpredictable kernel names are used again,
by masking u d ev's rule file for the default policy. This “ masking” can be done by making a
symbolic link to /d ev/nul l . As ro o t, issue a command as follows:
~]# l n -s /d ev/nul l /etc/ud ev/rul es. d /80 -net-name-sl o t. rul es
Create your own manual naming scheme, for example by naming your interfaces “ internet0” ,
“ dmz0” or “ lan0” . To do that create your own u d ev rules file and set the NAME property for the
devices. Make sure to order it before the default policy file, for example by naming it
/etc/ud ev/rul es. d /70 -my-net-names. rul es.
Alter the default policy file to pick a different naming scheme, for example to name all interfaces
after their MAC address by default. As ro o t copy the default policy file as follows:
~]# cp /usr/l i b/ud ev/rul es. d /80 -net-name-sl o t. rul es
/etc/ud ev/rul es. d /80 -net-name-sl o t. rul es
Edit the file in the /etc/ud ev/rul es. d / directory and change the lines as necessary.
14 1
Red Hat Ent erprise Linux 7 Net working G uide
Add the following line to the /etc/d efaul t/g rub file:
net.ifnames=0
or pass it to the kernel at boot time using the GRUB2 command-line interface. For more
information about GRUB2 see Red Hat Enterprise Linux 7 System Administrator's Guide.
8.10. T roubleshoot ing Net work Device Naming
Predictable interface names will be assigned for each interface, if applicable, as per the procedure
described in Section 8.2, “ Understanding the D evice Renaming Procedure” . To view the list of
possible names u d ev will use, issue a command in the following form as ro o t:
~]# ud evad m i nfo /sys/cl ass/net/ifname | g rep ID _NET _NAME
where ifname is one of the interfaces listed by the following command:
~]$ l s /sys/cl ass/net/
One of the possible names will be applied by u d ev according to the rules as described in
Section 8.2, “ Understanding the D evice Renaming Procedure” , and summarized here:
/usr/l i b/ud ev/rul es. d /6 0 -net. rul es - from initscripts,
/usr/l i b/ud ev/rul es. d /71-bi o sd evname. rul es - from b io sd evn ame,
/usr/l i b/ud ev/rul es. d /80 -net-name-sl o t. rul es - from systemd
From the above list of rule files it can be seen that if interface naming is done via initscripts or
b io sd evn ame it always takes precedence over u d ev native naming. However if initscripts renaming
is not taking place and b io sd evn ame is disabled, then to alter the interface names copy the 80 net-name-sl o t. rul es from /usr/ to /etc/ and edit the file appropriately. In other words,
comment out or arrange schemes to be used in a certain order.
Examp le 8.1. So me in t erf aces h ave n ames f ro m t h e kern el n amesp ace ( et h [ 0,1,2...] )
wh ile o t h ers are su ccessf u lly ren amed b y u d ev
Mixed up schemes most likely means that either for some hardware there is no usable information
provided by the kernel to u d ev, thus it could not figure out any names, or the information provided
to u d ev is not suitable, for example non-unique device ID s. The latter is more common and the
solution is to use a custom naming scheme in ifcfg files or alter which u d ev scheme is in use by
editing 80-net-name-slot.rules.
Examp le 8.2. In /var/lo g /messag es o r t h e syst emd jo u rn al, ren amin g is seen t o b e d o n e
t wice f o r each in t erf ace
Systems with the naming scheme encoded in ifcfg files but which do not have a regenerated
i ni trd image are likely to encounter this issue. The interface name is initially assigned (via
b io sd evn ame or u d ev or dracut parameters on the kernel command line) during early-boot while
still in i ni trd . Then after switching to real ro o tfs, renaming is done a second time and a new
interface name is determined by the /usr/l i b/ud ev/rename_d evi ce binary spawned by u d ev
because of processing 60-net.rules. You can safely ignore such messages.
14 2
⁠Chapt er 8 . Consist ent Net work Device Naming
Examp le 8.3. U sin g n amin g sch eme in if cf g f iles wit h et h X n ames d o es n o t wo rk
Use of interface names from kernel namespace is discouraged. To get predictable and stable
interface names please use some other prefix than " eth" .
8.11. Addit ional Resources
The following sources of information provide additional resources regarding Network Teaming.
8.11.1. Inst alled Document at ion
ud ev(7) man page — D escribes the Linux dynamic device management daemon, ud evd .
systemd (1) man page — D escribes systemd system and service manager.
bi o sd evname(1) man page — D escribes the utility for obtaining the BIOS-given name of a
device.
8.11.2. Online Document at ion
The IBM Knowledge Center Publication SC34-2710-00 Device Drivers, Features, and Commands on
Red Hat Enterprise Linux 7 includes information on “ Predictable network device names” for IBM
System z devices and attachments.
14 3
Red Hat Ent erprise Linux 7 Net working G uide
⁠Part II. InfiniBand and RDMA Networking
This part discusses how to set up RD MA, InfiniBand, and IP over InfiniBand network connections.
14 4
⁠Chapt er 9 . Configure InfiniBand and RDMA Net works
Chapter 9. Configure InfiniBand and RDMA Networks
9.1. Underst anding InfiniBand and RDMA t echnologies
InfiniBand refers to two distinctly different things. The first is a physical link-layer protocol for
InfiniBand networks. The second is a higher level programming API called the InfiniBand Verbs API.
The InfiniBand Verbs API is an implementation of a remote direct memory access (RD MA) technology.
RD MA communications differ from normal IP communications because they bypass kernel
intervention in the communication process, and in the process greatly reduce the CPU overhead
normally needed to process network communications. In a typical IP data transfer, application X on
machine A will send some data to application Y on machine B. As part of the transfer, the kernel on
machine B must first receive the data, decode the packet headers, determine that the data belongs to
application Y, wake up application Y, wait for application Y to perform a read syscall into the kernel,
then it must manually copy the data from the kernel's own internal memory space into the buffer
provided by application Y. This process means that most network traffic must be copied across the
system's main memory bus at least twice (once when the host adapter uses D MA to put the data into
the kernel-provided memory buffer, and again when the kernel moves the data to the application's
memory buffer) and it also means the computer must execute a number of context switches to switch
between kernel context and application Y context. Both of these things impose extremely high CPU
loads on the system when network traffic is flowing at very high rates.
The RD MA protocol allows the host adapter in the machine to know when a packet comes in from the
network, just which application should receive that packet, and just where in the application's
memory space it should go. Instead of sending the packet to the kernel to be processed and then
copied into the user application's memory, it places the contents of the packet directly in the
application's buffer without any further intervention necessary. This tremendously reduces the
overhead of high speed network communications. However, it cannot be accomplished using the
standard Berkeley Sockets API that most IP networking applications are built upon, so it must
provide its own API, the InfiniBand Verbs API, and applications must be ported to this API before they
can use RD MA technology directly.
Red Hat Enterprise Linux 7 supports both the InfiniBand hardware and the InfiniBand Verbs API. In
addition, there are two additional supported technologies that allow the InfiniBand Verbs API to be
utilized on non-InfiniBand hardware. They are iWARP (Internet Wide Area RD MA Protocol) and
RoCE/IBoE (RD MA over Converged Ethernet, which was later renamed to InfiniBand over Ethernet).
Both of these technologies have a normal IP network link layer as their underlying technology, and
so the majority of their configuration is actually covered in the Chapter 2, Configure IP Networking
chapter of this document. For the most part, once their IP networking features are properly
configured, their RD MA features are all automatic and will show up as long as the proper drivers for
the hardware are installed. The kernel drivers are always included with each kernel Red Hat provides,
however the user space drivers must be installed manually if the user did not select the InfiniBand
package group at machine install time.
T h ese are t h e n ecessary u ser- sp ace p ackag es:
iWAR P
C hel si o hard ware — lib cxg b 3 or lib cxg b 4 depending on version of hardware
R o C E/IB o E
Mel l ano x hard ware — lib mlx4 or lib mlx5 depending on version of hardware.
Additionally, the user may need to edit /etc/rd ma/ml x4 . co nf or
/etc/rd ma/ml x5. co nf to set the port types properly for RoCE/IBoE usage. The user may
need to edit /etc/mo d pro be. d /ml x4 . co nf or /etc/mo d pro be. d /ml x5. co nf to
14 5
Red Hat Ent erprise Linux 7 Net working G uide
instruct the driver which packet priority is configured for no-drop service on the Ethernet
switches the cards are plugged into.
With these driver packages installed (in addition to the normal RD MA packages typically installed for
any InfiniBand installation), a user should be able to utilize most of the normal RD MA applications to
test and see RD MA protocol communication taking place on their adapters. However, not all of the
programs included in Red Hat Enterprise Linux 7 will properly support iWARP or RoCE/IBoE
devices. This is because the connection establishment protocol on iWARP in particular is different
than it is on real InfiniBand link-layer connections. If the program in question uses the lib rd macm
connection management library, it will handle the differences between iWARP and InfiniBand silently
and the program should work. If the application tries to do its own connection management, then it
must specifically support iWARP or else it will not work.
9.2. InfiniBand and RDMA relat ed soft ware packages
Because RD MA applications are so different from Berkeley Sockets based applications, and from
normal IP networking, most applications that are used on an IP network cannot be used directly on
an RD MA network. Red Hat Enterprise Linux 7 comes with a number of different software packages for
RD MA network administration, testing and debugging, high level software development APIs, and
performance analysis.
In order to utilize these networks, some or all of these packages need to be installed (this list is not
exhaustive, but does cover the most important packages related to RD MA).
R eq u ired p ackag es:
rdma — responsible for kernel initialization of the RD MA stack.
libibverbs — provides the InfiniBand Verbs API.
opensm — subnet manager (only required on one machine, and only if there is no subnet
manager active on the fabric).
user space d ri ver fo r i nstal l ed hard ware — one or more of: infinipath-psm, libcxgb3,
libcxgb4, libehca, libipathverbs, libmthca, libmlx4, libmlx5, libnes, and libocrdma. Note that libehca is only
available for IBM Power Systems servers.
R eco mmen d ed p ackag es:
librdmacm, librdmacm-utils, and ibacm — Connection management library that is
aware of the differences between InfiniBand, iWARP, and RoCE and is able to properly open
connections across all of these hardware types, some simple test programs for verifying the
operation of the network, and a caching daemon that integrates with the library to make remote
host resolution faster in large clusters.
libibverbs-utils — Simple Verbs based programs for querying the installed hardware and
verifying communications over the fabric.
infiniband-diags and ibutils — Provide a number of useful debugging tools for
InfiniBand fabric management. These provide only very limited functionality on iWARP or RoCE
as most of the tools work at the InfiniBand link layer, not the Verbs API layer.
perftest and qperf — Performance testing applications for various types of RD MA
communications.
O p t io n al p ackag es:
14 6
⁠Chapt er 9 . Configure InfiniBand and RDMA Net works
These packages are available in the Optional channel. Before installing packages from the Optional
channel, see Scope of Coverage D etails. Information on subscribing to the Optional channel can be
found in the Red Hat Knowledgebase solution How to access Optional and Supplementary channels.
dapl, dapl-devel, and dapl-utils — Provide a different API for RD MA than the Verbs
API. There is both a runtime component and a development component to these packages.
openmpi, mvapich2, and mvapich2-psm — MPI stacks that have the ability to use RD MA
communications. User-space applications writing to these stacks are not necessarily aware that
RD MA communications are taking place.
9.3. Configuring t he Base RDMA Subsyst em
9.3.1. T he RDMA package Inst allat ion
The rdma package is not part of the default install package set. If the InfiniBand package group was
not selected during install, the rdma package (as well as a number of others as listed in the previous
section) can be installed after the initial installation is complete. If it was not installed at machine
installation time and instead was installed manually later, then it is necessary to rebuild the
i ni tramfs images using d racu t in order for it to function fully as intended. Issue the following
commands as ro o t:
~]# yum i nstal l rd ma
d racut -f
Startup of the rd ma service is automatic. When RD MA capable hardware, whether InfiniBand or
iWARP or RoCE/IBoE is detected, u d ev instructs systemd to start the rd ma service. Users need not
enable the rd ma service, but they can if they want to force it on all the time. To do that, issue the
following command:
~]# systemctl enabl e rd ma
9.3.2. Configurat ion of t he rdma.conf file
The rd ma service reads /etc/rd ma/rd ma. co nf to find out which kernel-level and user-level RD MA
protocols the administrator wants to be loaded by default. Users should edit this file to turn various
drivers on or off.
The various drivers that can be enabled and disabled are:
IP o IB — This is an IP network emulation layer that allows IP applications to run over
InfiniBand networks.
SR P — This is the SCSI Request Protocol. It allows a machine to mount a remote drive or drive
array that is exported via the SR P protocol on the machine as though it were a local hard disk.
SR P T — This is the target mode, or server mode, of the SR P protocol. This loads the kernel
support necessary for exporting a drive or drive array for other machines to mount as though it
were local on their machine. Further configuration of the target mode support is required before
any devices will actually be exported. See the documentation in the targetd and targetcli packages
for further information.
ISER — This is a low-level driver for the general iSCSI layer of the Linux kernel that provides
transport over InfiniBand networks for iSCSI devices.
14 7
Red Hat Ent erprise Linux 7 Net working G uide
R D S — This is the Reliable D atagram Service in the Linux kernel. It is not enabled in Red Hat
Enterprise Linux 7 kernels and so cannot be loaded.
9.3.3. Usage of 70-persist ent -ipoib.rules
The rdma package provides the file /etc/ud ev. d /rul es. d /70 -persi stent-i po i b. rul es.
This u d ev rules file is used to rename IPoIB devices from their default names (such as i b0 and i b1)
to more descriptive names. Users must edit this file to change how their devices are named. First, find
out the GUID address for the device to be renamed:
~]$ i p l i nk sho w i b0
8: ib0: >BROADCAST,MULTICAST,UP,LOWER_UP< mtu 65520 qdisc pfifo_fast
state UP mode DEFAULT qlen 256
link/infiniband
80:00:02:00:fe:80:00:00:00:00:00:00:f4 : 52: 14 : 0 3: 0 0 : 7b: cb: a1 brd
00:ff:ff:ff:ff:12:40:1b:ff:ff:00:00:00:00:00:00:ff:ff:ff:ff
Immediately after l i nk/i nfi ni band is the 20 byte hardware address for the IPoIB interface. The
final 8 bytes of the address, marked in bold above, is all that is required to make a new name. Users
can make up whatever naming scheme suits them. For example, use a device_fabric naming
convention such as ml x4 _i b0 if a ml x4 device is connected to an i b0 subnet fabric. The only
thing that is not recommended is to use the standard names, like i b0 or i b1, as these can conflict
with the kernel assigned automatic names. Next, add an entry in the rules file. Copy the existing
example in the rules file, replace the 8 bytes in the AT T R {ad d ress} entry with the highlighted 8
bytes from the device to be renamed, and enter the new name to be used in the NAME field.
9.3.4 . Relaxing memlock rest rict ions for users
RD MA communications require that physical memory in the computer be pinned (meaning that the
kernel is not allowed to swap that memory out to a paging file in the event that the overall computer
starts running short on available memory). Pinning memory is normally a very privileged operation.
In order to allow users other than ro o t to run large RD MA applications, it will likely be necessary to
increase the amount of memory that non-ro o t users are allowed to pin in the system. This is done by
adding a file in the /etc/securi ty/l i mi ts. d / directory with contents such as the following:
~]$ mo re /etc/securi ty/l i mi ts. d /rd ma. co nf
# configuration for rdma tuning
*
soft
memlock
unlimited
*
hard
memlock
unlimited
# rdma tuning end
9.3.5. Configuring Mellanox cards for Et hernet operat ion
Certain hardware from Mellanox is capable of running in either InfiniBand or Ethernet mode. These
cards generally default to InfiniBand. Users can set the cards to Ethernet mode. There is currently
support for setting the mode only on ConnectX family hardware (which uses the mlx4 driver). To set
the mode, users should follow the instruction in the file /etc/rd ma/ml x4 . co nf to find the right PCI
device ID for their given hardware. They should then create a line in the file using that device ID and
the requested port type. They should then rebuild their i ni tramfs to make sure the updated port
settings are copied into the i ni tramfs.
Once the port type has been set, if one or both ports are set to Ethernet, then users might see this
message in their logs: ml x4 _co re 0 0 0 0 : 0 5: 0 0 . 0 : R eq uested po rt type fo r po rt 1 i s
no t suppo rted o n thi s HC A This is normal and will not effect operation. The script responsible
14 8
⁠Chapt er 9 . Configure InfiniBand and RDMA Net works
for setting the port type has no way of knowing when the driver has finished switching port 2 to the
requested type internally, and from the time that the script issues a request for port 2 to switch until
that switch is complete, the attempts to set port 1 to a different type get rejected. The script retries until
the command succeeds or until a timeout has passed indicating that the port switch never completed.
9.4 . Configuring t he Subnet Manager
9.4 .1. Det ermining Necessit y
Most InfiniBand switches come with an embedded subnet manager. However, if a more up to date
subnet manager is required than the one in the switch firmware, or if more complete control than the
switch manager allows is required, Red Hat Enterprise Linux 7 includes the o p en sm subnet
manager. All InfiniBand networks mu st have a subnet manager running for the network to
function. This is true even when doing a simple network of two machines with no switch and the
cards are plugged in back to back, a subnet manager is required for the link on the cards to come
up. It is possible to have more than one, in which case one will act as master, and any other subnet
managers will act as slaves that will take over should the master subnet manager fail.
9.4 .2. Configuring t he opensm mast er configurat ion file
The o p en sm program keeps its master configuration file in /etc/rd ma/o pensm. co nf. Users may
edit this file at any time and edits will be kept on upgrade. There is extensive documentation of the
options in the file itself. However, for the two most common edits needed, setting the GUID to bind to
and the PRIORITY to run with, it is highly recommended that the o pensm. co nf file is not edited but
instead edit /etc/sysco nfi g /o pensm. If there are no edits to the base
/etc/rd ma/o pensm. co nf file, it will get upgraded whenever the opensm package is upgraded. As
new options are added to this file regularly, this makes it easier to keep the current configuration up
to date. If the o pensm. co nf file has been changed, then on upgrade, it might be necessary to merge
new options into the edited file.
9.4 .3. Configuring t he opensm st art up opt ions
The options in the /etc/sysco nfi g /o pensm file control how the subnet manager is actually
started, as well as how many copies of the subnet manager are started. For example, a dual port
InfiniBand card, with each port plugged into physically separate networks, will need a copy of the
subnet manager running on each port. The o p en sm subnet manager will only manage one subnet
per instance of the application and must be started once for each subnet that needs to be managed.
In addition, if there is more than one o p en sm server, then set the priorities on each server to control
which are to be slaves and which are to be master.
The file /etc/sysco nfi g /o pensm is used to provide a simple means to set the priority of the
subnet manager and to control which GUID the subnet manager binds to. There is an extensive
explanation of the options in the /etc/sysco nfi g /o pensm file itself. Users need only read and
follow the directions in the file itself to enable failover and multifabric operation of o p en sm.
9.4 .4 . Creat ing a P_Key definit ion
By default, o pensm. co nf looks for the file /etc/rd ma/parti ti o ns. co nf to get a list of partitions
to create on the fabric. All fabrics must contain the 0 x7fff subnet, and all switches and all hosts
must belong to that fabric. Any other partition can be created in addition to that, and all hosts and all
switches do not have to be members of these additional partitions. This allows an administrator to
create subnets akin to Ethernet’s VLANs on InfiniBand fabrics. If a partition is defined with a given
speed, such as 40 Gbps, and there is a host on the network unable to do 40 Gbps, then that host will
14 9
Red Hat Ent erprise Linux 7 Net working G uide
be unable to join the partition even if it has permission to do so as it will be unable to match the
speed requirements, therefore it is recommended that the speed of a partition be set to the slowest
speed of any host with permission to join the partition. If a faster partition for some subset of hosts is
required, create a different partition with the higher speed.
The following partition file would result in a default 0 x7fff partition at a reduced speed of 10 Gbps,
and a partition of 0 x0 0 0 2 with a speed of 40 Gbps:
~]$ mo re /etc/rd ma/parti ti o ns. co nf
# For reference:
# IPv4 IANA reserved multicast addresses:
#
http://www.iana.org/assignments/multicast-addresses/multicastaddresses.txt
# IPv6 IANA reserved multicast addresses:
#
http://www.iana.org/assignments/ipv6-multicast-addresses/ipv6multicast-addresses.xml
#
# mtu =
#
1 = 256
#
2 = 512
#
3 = 1024
#
4 = 2048
#
5 = 4096
#
# rate =
#
2 =
2.5 GBit/s
#
3 = 10
GBit/s
#
4 = 30
GBit/s
#
5 =
5
GBit/s
#
6 = 20
GBit/s
#
7 = 40
GBit/s
#
8 = 60
GBit/s
#
9 = 80
GBit/s
#
10 = 120
GBit/s
Default=0x7fff, rate=3 mtu=4 scope=2, defmember=full:
ALL, ALL_SWITCHES=full;
Default=0x7fff, ipoib, rate=3 mtu=4 scope=2:
mgid=ff12:401b::ffff:ffff
# IPv4 Broadcast address
mgid=ff12:401b::1
# IPv4 All Hosts group
mgid=ff12:401b::2
# IPv4 All Routers group
mgid=ff12:401b::16
# IPv4 IGMP group
mgid=ff12:401b::fb
# IPv4 mDNS group
mgid=ff12:401b::fc
# IPv4 Multicast Link Local Name
Resolution group
mgid=ff12:401b::101
# IPv4 NTP group
mgid=ff12:401b::202
# IPv4 Sun RPC
mgid=ff12:601b::1
# IPv6 All Hosts group
mgid=ff12:601b::2
# IPv6 All Routers group
mgid=ff12:601b::16
# IPv6 MLDv2-capable Routers
group
mgid=ff12:601b::fb
# IPv6 mDNS group
mgid=ff12:601b::101
# IPv6 NTP group
mgid=ff12:601b::202
# IPv6 Sun RPC group
mgid=ff12:601b::1:3
# IPv6 Multicast Link Local Name
Resolution group
150
⁠Chapt er 9 . Configure InfiniBand and RDMA Net works
ALL=full, ALL_SWITCHES=full;
ib0_2=0x0002, rate=7 mtu=4 scope=2, defmember=full:
ALL, ALL_SWITCHES=full;
ib0_2=0x0002, ipoib, rate=7 mtu=4 scope=2:
mgid=ff12:401b::ffff:ffff
# IPv4 Broadcast address
mgid=ff12:401b::1
# IPv4 All Hosts group
mgid=ff12:401b::2
# IPv4 All Routers group
mgid=ff12:401b::16
# IPv4 IGMP group
mgid=ff12:401b::fb
# IPv4 mDNS group
mgid=ff12:401b::fc
# IPv4 Multicast Link Local Name
Resolution group
mgid=ff12:401b::101
# IPv4 NTP group
mgid=ff12:401b::202
# IPv4 Sun RPC
mgid=ff12:601b::1
# IPv6 All Hosts group
mgid=ff12:601b::2
# IPv6 All Routers group
mgid=ff12:601b::16
# IPv6 MLDv2-capable Routers
group
mgid=ff12:601b::fb
# IPv6 mDNS group
mgid=ff12:601b::101
# IPv6 NTP group
mgid=ff12:601b::202
# IPv6 Sun RPC group
mgid=ff12:601b::1:3
# IPv6 Multicast Link Local Name
Resolution group
ALL=full, ALL_SWITCHES=full;
9.4 .5. Enabling opensm
Users need to enable the o p en sm service as it is not enabled by default when installed. Issue the
following command as ro o t:
~]# systemctl enabl e o pensm
9.5. T est ing Early InfiniBand RDMA operat ion
Note
This section applies only to InfiniBand devices. Since iWARP and RoCE/IBoE devices are IP
based devices, users should proceed to the section on testing RD MA operations once IPoIB
has been configured and the devices have IP addresses.
Once the rd ma service is enabled, and the o p en sm service (if needed) is enabled, and the proper
user-space library for the specific hardware has been installed, user space rd ma operation should
be possible. Simple test programs from the libibverbs-utils package are helpful in determining that
RD MA operations are working properly. The ib v_d evices program will show which devices are
present in the system and the i bv_d evi nfo command will give detailed information about each
device. For example:
~]$ i bv_d evi ces
device
-----mlx4_0
node GUID
---------------0002c903003178f0
151
Red Hat Ent erprise Linux 7 Net working G uide
mlx4_1
f4521403007bcba0
~]$ i bv_d evi nfo -d ml x4 _1
hca_id: mlx4_1
transport:
InfiniBand (0)
fw_ver:
2.30.8000
node_guid:
f452:1403:007b:cba0
sys_image_guid:
f452:1403:007b:cba3
vendor_id:
0x02c9
vendor_part_id:
4099
hw_ver:
0x0
board_id:
MT_1090120019
phys_port_cnt:
2
port:
1
state:
PORT_ACTIVE (4)
max_mtu:
4096 (5)
active_mtu:
2048 (4)
sm_lid:
2
port_lid:
2
port_lmc:
0x01
link_layer:
InfiniBand
port:
2
state:
max_mtu:
active_mtu:
sm_lid:
port_lid:
port_lmc:
link_layer:
~ ]$ i bstat ml x4 _1
CA 'mlx4_1'
CA type: MT4099
Number of ports: 2
Firmware version: 2.30.8000
Hardware version: 0
Node GUID: 0xf4521403007bcba0
System image GUID: 0xf4521403007bcba3
Port 1:
State: Active
Physical state: LinkUp
Rate: 56
Base lid: 2
LMC: 1
SM lid: 2
Capability mask: 0x0251486a
Port GUID: 0xf4521403007bcba1
Link layer: InfiniBand
Port 2:
State: Active
Physical state: LinkUp
Rate: 40
Base lid: 0
LMC: 0
152
PORT_ACTIVE (4)
4096 (5)
4096 (5)
0
0
0x00
Ethernet
⁠Chapt er 9 . Configure InfiniBand and RDMA Net works
SM lid: 0
Capability mask: 0x04010000
Port GUID: 0xf65214fffe7bcba2
Link layer: Ethernet
The i bv_d evi nfo and i bstat commands output slightly different information (such as port MTU
exists in i bv_d evi nfo but not in i bstat output, and the Port GUID exists in i bstat output but not
in i bv_d evi nfo output), and a few things are named differently (for example, the Base local identifier
(LID ) in i bstat output is the same as the po rt_l i d output of i bv_d evi nfo )
Simple ping programs, such as ib p in g from the infiniband-diags package, can be used to test RD MA
connectivity. The ib p in g program uses a client/server model. You must first start an ib p in g server
on one machine, then run ib p in g as a client on another machine and tell it to connect to the ib p in g
server. Since we are wanting to test the base RD MA capability, we need to use an RD MA specific
address resolution method instead of IP addresses for specifying the server.
On the server machine, the user can use the i bv_d evi nfo and i bstat commands to print out the
po rt_l i d (or Base lid) and the Port GUID of the port they want to test (assuming port 1 of the above
interface, the po rt_l i d /Base LID is 2 and Port GUID is 0 xf4 5214 0 30 0 7bcba1)). Then start
ib p in g with the necessary options to bind specifically to the card and port to be tested, and also
specifying ib p in g should run in server mode. You can see the available options to ib p in g by
passing -? or --hel p, but in this instance we will need either the -S or --Server option and for
binding to the specific card and port we will need either -C or --C a and -P or --P o rt. Note: port in
this instance does not denote a network port number, but denotes the physical port number on the
card when using a multi-port card. To test connectivity to the RD MA fabric using, for example, the
second port of a multi-port card, requires telling ib p in g to bind to port 2 on the card. When using a
single port card, or testing the first port on a card, this option is not needed. For example:
~]$ i bpi ng -S -C ml x4 _1 -P 1
Then change to the client machine and run ib p in g . Make note of either the port GUID of the port the
server ib p in g program is bound to, or the local identifier (LID ) of the port the server ib p in g program
is bound to. Also, take note which card and port in the client machine is physically connected to the
same network as the card and port that was bound to on the server. For example, if the second port
of the first card on the server was bound to, and that port is connected to a secondary RD MA fabric,
then on the client specify whichever card and port are necessary to also be connected to that
secondary fabric. Once these things are known, run the ib p in g program as a client and connect to
the server using either the port LID or GUID that was collected on the server as the address to
connect to. For example:
~]$ i bpi ng -c 10 0 0 0 -f -C ml x4 _0 -P 1 -L 2
--- rdma-host.example.com.(none) (Lid 2) ibping statistics --10000 packets transmitted, 10000 received, 0% packet loss, time 816 ms
rtt min/avg/max = 0.032/0.081/0.446 ms
or
~]$ i bpi ng -c 10 0 0 0 -f -C ml x4 _0 -P 1 -G 0 xf4 5214 0 30 0 7bcba1 \
--- rdma-host.example.com.(none) (Lid 2) ibping statistics --10000 packets transmitted, 10000 received, 0% packet loss, time 769 ms
rtt min/avg/max = 0.027/0.076/0.278 ms
This outcome verifies that end to end RD MA communications are working for user space
applications.
153
Red Hat Ent erprise Linux 7 Net working G uide
The following error may be encountered:
~]$ i bv_d evi nfo
libibverbs: Warning: no userspace device-specific driver found for
/sys/class/infiniband_verbs/uverbs0
No IB devices found
This error indicates that the necessary user-space library is not installed. The administrator will need
to install one of the user-space libraries (as appropriate for their hardware) listed in section
Section 9.2, “ InfiniBand and RD MA related software packages” . On rare occasions, this can happen
if a user installs the wrong arch type for the driver or for lib ib verb s. For example, if lib ib verb s is of
arch x86 _6 4 , and lib mlx4 is installed but is of type i 6 86 , then this error can result.
Note
Many sample applications prefer to use host names or addresses instead of LID s to open
communication between the server and client. For those applications, it is necessary to set up
IPoIB before attempting to test end-to-end RD MA communications. The ib p in g application is
unusual in that it will accept simple LID s as a form of addressing, and this allows it to be a
simple test that eliminates possible problems with IPoIB addressing from the test scenario and
therefore gives us a more isolated view of whether or not simple RD MA communications are
working.
9.6. Configuring IPoIB
9.6.1. Underst anding t he role of IPoIB
As mentioned in Section 1.2, “ IP Networks versus non-IP Networks” , most networks are IP networks.
InfiniBand is not. The role of IPoIB is to provide an IP network emulation layer on top of InfiniBand
RD MA networks. This allows existing applications to run over InfiniBand networks unmodified.
However, the performance of those applications is considerably lower than if the application were
written to use RD MA communication natively. Since most InfiniBand networks have some set of
applications that really must get all of the performance they can out of the network, and then some
other applications for which a degraded rate of performance is acceptable if it means that the
application does not need to be modified to use RD MA communications, IPoIB is there to allow those
less critical applications to run on the network as they are.
Because both iWARP and RoCE/IBoE networks are actually IP networks with RD MA layered on top
of their IP link layer, they have no need of IPoIB. As a result, the kernel will refuse to create any IPoIB
devices on top of iWARP or RoCE/IBoE RD MA devices.
9.6.2. Underst anding IPoIB communicat ion modes
IPoIB devices can be configured to run in either datagram or connected mode. The difference is in
what type of queue pair the IPoIB layer attempts to open with the machine at the other end of the
communication. For datagram mode, an unreliable, disconnected queue pair is opened. For
connected mode, a reliable, connected queue pair is opened.
When using datagram mode, the unreliable, disconnected queue pair type does not allow any
packets larger than the InfiniBand link-layer’s MTU. The IPoIB layer adds a 4 byte IPoIB header on
top of the IP packet being transmitted. As a result, the IPoIB MTU must be 4 bytes less than the
InfiniBand link-layer MTU. As 2048 is a common InfiniBand link-layer MTU, the common IPoIB device
154
⁠Chapt er 9 . Configure InfiniBand and RDMA Net works
MTU in datagram mode is 2044.
When using connected mode, the reliable, connected queue pair type allows messages that are
larger than the InfiniBand link-layer MTU and the host adapter handles packet segmentation and
reassembly at each end. As a result, there is no size limit imposed on the size of IPoIB messages that
can be sent by the InfiniBand adapters in connected mode. However, there is still the limitation that
an IP packet only has a 16 bit size field, and is therefore limited to 6 5535 as the maximum byte
count. The maximum allowed MTU is actually smaller than that because we have to account for
various TCP/IP headers that must also fit in that size. As a result, the IPoIB MTU in connected mode
is capped at 6 5520 in order to make sure there is sufficient room for all needed T C P headers.
The connected mode option generally has higher performance, but it also consumes more kernel
memory. Because most systems care more about performance than memory consumption, connected
mode is the most commonly used mode.
However, if a system is configured for connected mode, it must still send multicast traffic in datagram
mode (the InfiniBand switches and fabric cannot pass multicast traffic in connected mode) and it will
also fall back to datagram mode when communicating with any hosts not configured for connected
mode. Administrators should be aware that if they intend to run programs that send multicast data,
and those programs try to send multicast data up to the maximum MTU on the interface, then it is
necessary to configure the interface for datagram operation or find some way to configure the
multicast application to cap their packet send size at a size that will fit in datagram sized packets.
9.6.3. Underst anding IPoIB hardware addresses
IPoIB devices have a 20 byte hardware addresses. The deprecated utility if co n f ig is unable to read
all 20 bytes and should never be used to try and find the correct hardware address for an IPoIB
device. The ip utilities from the iproute package work properly.
The first 4 bytes of the IPoIB hardware address are flags and the queue pair number. The next 8
bytes are the subnet prefix. When the IPoIB device is first created, it will have the default subnet prefix
of 0 xfe: 80 : 0 0 : 0 0 : 0 0 : 0 0 : 0 0 : 0 0 . Once the system establishes contact with the subnet
manager, it is possible that the subnet manager will inform the system that the default subnet prefix is
not the prefix in use for this subnet, at which point the 8 byte subnet prefix will change to the subnet
prefix the subnet manager informs the system about. The final 8 bytes are the GUID address of the
InfiniBand port that the IPoIB device is attached to. Because both the first 4 bytes and the next 8
bytes can change from time to time, they are not used or matched against when specifying the
hardware address for an IPoIB interface. Section Section 9.3.3, “ Usage of 70-persistent-ipoib.rules”
explains how to derive the address by leaving the first 12 bytes out of the AT T R {ad d ress} field in
the u d ev rules file so that device matching will happen reliably. When configuring IPoIB interfaces,
the HWAD D R field of the configuration file can contain all 20 bytes, but only the last 8 bytes are
actually used to match against and find the hardware specified by a configuration file. However, if the
T Y P E= Infi ni Band entry is not spelled correctly in the device configuration file, and if u p - ib is not
the actual script used to bring up the IPoIB interface, then an error about the system being unable to
find the hardware specified by the configuration will be issued. For IPoIB interfaces, the T Y P E= field
of the configuration file must be either Infi ni Band or i nfi ni band (the entry is case sensitive, but
the scripts will accept these two specific spellings).
9.6.4 . Underst anding InfiniBand P_Key subnet s
An InfiniBand fabric can be logically segmented into virtual subnets by the use of different P _Key
subnets. This is highly analogous to using VLANs on Ethernet interfaces. All switches and hosts must
be a member of the default P _Key subnet, but administrators can create additional subnets and limit
members of those subnets to subsets of the hosts or switches in the fabric. A P _Key subnet must be
defined by the subnet manager before a host can use it. See section Section 9.4.4, “ Creating a P_Key
definition” for information on how to define a P _Key subnet using the o p en sm subnet manager. For
155
Red Hat Ent erprise Linux 7 Net working G uide
IPoIB interfaces, once a P _Key subnet has been created, we can create additional IPoIB
configuration files specifically for those P _Key subnets. Just like VLAN interfaces on Ethernet
devices, each IPoIB interface will behave as though it were on a completely different fabric from other
IPoIB interfaces that share the same link but have different P _Key values.
There are special requirements for the names of IPoIB P _Key interfaces. All IPoIB P _Keys range from
0 x0 0 0 0 to 0 x7fff, and the high bit, 0 x80 0 0 , denotes that membership in a P _Key is full
membership instead of partial membership. The Linux kernel’s IPoIB driver only supports full
membership in P _Key subnets, so for any subnet that Linux can connect to, the high bit of the P _Key
number will always be set. That means that if a Linux computer joins P _Key 0 x0 0 0 2, its actual
P _Key number once joined will be 0 x80 0 2, denoting that we are full members of P _Key 0 x0 0 0 2.
For this reason, when creating a P _Key definition in an o p en sm parti ti o ns. co nf file as
depicted in section Section 9.4.4, “ Creating a P_Key definition” , it is required to specify a P _Key
value without 0 x80 0 0 , but when defining the P _Key IPoIB interfaces on the Linux clients, add the
0 x80 0 0 value to the base P _Key value.
9.7. Configure InfiniBand Using t he T ext User Int erface, nmt ui
The text user interface tool n mt u i can be used to configure InfiniBand in a terminal window. Issue
the following command to start the tool:
~]$ nmtui
The text user interface appears. Any invalid command prints a usage message.
To navigate, use the arrow keys or press T ab to step forwards and press Shi ft+T ab to step back
through the options. Press Enter to select an option. The Space bar toggles the status of a check
box.
From the starting menu, select Ed i t a co nnecti o n. Select Ad d , the New C o nnecti o n screen
opens.
156
⁠Chapt er 9 . Configure InfiniBand and RDMA Net works
Fig u re 9 .1. T h e N et wo rkMan ag er T ext U ser In t erf ace Ad d an In f in iB an d C o n n ect io n
men u
Select Infi ni Band , the Ed i t co nnecti o n screen opens. Follow the on-screen prompts to
complete the configuration.
157
Red Hat Ent erprise Linux 7 Net working G uide
Fig u re 9 .2. T h e N et wo rkMan ag er T ext U ser In t erf ace C o n f ig u rin g a In f in iB an d
C o n n ect io n men u
See Section 9.11.1, “ Configuring the InfiniBand Tab” for definitions of the InfiniBand terms.
See Section 1.5, “ Network Configuration Using a Text User Interface (nmtui)” for information on
installing n mt u i.
9.8. Configure IPoIB using t he command-line t ool, nmcli
First determine if renaming the default IPoIB device(s) is required, and if so, follow the instructions in
section Section 9.3.3, “ Usage of 70-persistent-ipoib.rules” to rename the devices using u d ev
renaming rules. Users can force the IPoIB interfaces to be renamed without performing a reboot by
removing the i b_i po i b kernel module and then reloading it as follows:
~]$ rmmo d i b_i po i b
~]$ mo d pro be i b_i po i b
Once the devices have the name required, use the n mcli tool to create the IPoIB interface(s) as
follows:
~]$ nmcl i co n ad d type i nfi ni band co n-name ml x4 _i b0 i fname ml x4 _i b0
transpo rt-mo d e co nnected mtu 6 5520
Connection 'mlx4_ib0' (8029a0d7-8b05-49ff-a826-2a6d722025cc) successfully
added.
~]$ nmcl i co n ed i t ml x4 _i b0
===| nmcli interactive connection editor |===
Editing existing 'infiniband' connection: 'mlx4_ib0'
158
⁠Chapt er 9 . Configure InfiniBand and RDMA Net works
Type 'help' or '?' for available commands.
Type 'describe [>setting<.>prop<]' for detailed property description.
You may edit the following settings: connection, infiniband, ipv4, ipv6
nmcli> set infiniband.mac-address
80:00:02:00:fe:80:00:00:00:00:00:00:f4:52:14:03:00:7b:cb:a3
nmcli> save
Connection 'mlx4_ib3' (8029a0d7-8b05-49ff-a826-2a6d722025cc) successfully
updated.
nmcli> quit
At this point, an IPoIB interface named ml x4 _i b0 has been created and set to use connected mode,
with the maximum connected mode MTU, D HC P for IP v4 and IP v6 . If using IPoIB interfaces for
cluster traffic and an Ethernet interface for out-of-cluster communications, it is likely that disabling
default routes and any default name server on the IPoIB interfaces will be required. This is can be
done as follows:
~]$ nmcl i co n ed i t ml x4 _i b0
===| nmcli interactive connection editor |===
Editing existing 'infiniband' connection: 'mlx4_ib0'
Type 'help' or '?' for available commands.
Type 'describe [>setting<.>prop<]' for detailed property description.
You may edit the following settings: connection, infiniband, ipv4, ipv6
nmcli> set ipv4.ignore-auto-dns yes
nmcli> set ipv4.ignore-auto-routes yes
nmcli> set ipv4.never-default true
nmcli> set ipv6.ignore-auto-dns yes
nmcli> set ipv6.ignore-auto-routes yes
nmcli> set ipv6.never-default true
nmcli> save
Connection 'mlx4_ib0' (8029a0d7-8b05-49ff-a826-2a6d722025cc) successfully
updated.
nmcli> quit
If a P _Key interface is required, create one using n mcli as follows:
~]$ nmcl i co n ad d type i nfi ni band co n-name ml x4 _i b0 . 80 0 2 i fname
ml x4 _i b0 . 80 0 2 parent ml x4 _i b0 p-key 0 x80 0 2
Connection 'mlx4_ib0.8002' (4a9f5509-7bd9-4e89-87e9-77751a1c54b4)
successfully added.
~]$ nmcl i co n mo d i fy ml x4 _i b0 . 80 0 2 i nfi ni band . mtu 6 5520
i nfi ni band . transpo rt-mo d e co nnected i pv4 . i g no re-auto -d ns yes
i pv4 . i g no re-auto -ro utes yes i pv4 . never-d efaul t true i pv6 . i g no re-auto d ns yes i pv6 . i g no re-auto -ro utes yes i pv6 . never-d efaul t true
9.9. Configure IPoIB Using t he command line
159
Red Hat Ent erprise Linux 7 Net working G uide
First determine if renaming the default IPoIB device(s) is required, and if so, follow the instructions in
section Section 9.3.3, “ Usage of 70-persistent-ipoib.rules” to rename the devices using u d ev
renaming rules. Users can force the IPoIB interfaces to be renamed without performing a reboot by
removing the i b_i po i b kernel module and then reloading it as follows:
~]$ rmmo d i b_i po i b
~]$ mo d pro be i b_i po i b
Once the devices have the name required, administrators can create i fcfg files with their preferred
editor to control the devices. A typical IPoIB configuration file with static IP v4 addressing looks as
follows:
~]$ mo re i fcfg -ml x4 _i b0
DEVICE=mlx4_ib0
TYPE=InfiniBand
ONBOOT=yes
HWADDR=80:00:00:4c:fe:80:00:00:00:00:00:00:f4:52:14:03:00:7b:cb:a1
BOOTPROTO=none
IPADDR=172.31.0.254
PREFIX=24
NETWORK=172.31.0.0
BROADCAST=172.31.0.255
IPV4_FAILURE_FATAL=yes
IPV6INIT=no
MTU=65520
CONNECTED_MODE=yes
NAME=mlx4_ib0
The D EVICE field must match the custom name created in any u d ev renaming rules. The NAME entry
need not match the device name. If the GUI connection editor is started, the NAME field is what is used
to present a name for this connection to the user. The TYPE field must be InfiniBand in order for
InfiniBand options to be processed properly. CONNECTED _MOD E is either yes or no , where yes will
use connected mode and no will use datagram mode for communications (see section Section 9.6.2,
“ Understanding IPoIB communication modes” ).
For P _Key interfaces, this is a typical configuration file:
~]$ mo re i fcfg -ml x4 _i b0 . 80 0 2
DEVICE=mlx4_ib0.8002
PHYSDEV=mlx4_ib0
PKEY=yes
PKEY_ID=2
TYPE=InfiniBand
ONBOOT=yes
HWADDR=80:00:00:4c:fe:80:00:00:00:00:00:00:f4:52:14:03:00:7b:cb:a1
BOOTPROTO=none
IPADDR=172.31.2.254
PREFIX=24
NETWORK=172.31.2.0
BROADCAST=172.31.2.255
IPV4_FAILURE_FATAL=yes
IPV6INIT=no
MTU=65520
CONNECTED_MODE=yes
NAME=mlx4_ib0.8002
160
⁠Chapt er 9 . Configure InfiniBand and RDMA Net works
For all P _Key interface files, the PHYSD EV is required and must be the name of the parent device,
PKEY must be yes, and P KEY _ID must be the number of the interface (either with or without the
0 x80 0 0 membership bit added in). The device name, however, must be the four digit hexadecimal
representation of the P KEY _ID with the 0 x80 0 0 membership bit logically OR’ed into the number. By
default, the P KEY _ID in the file is treated as a decimal number and converted to hexadecimal and
then OR’ed with 0 x80 0 0 to arrive at the proper name for the device, but users may specify the
P KEY _ID in hexadecimal by prepending the standard 0 x prefix to the number.
9.10. T est ing an RDMA net work aft er IPoIB is configured
Once IPoIB is configured, it is possible to use IP addresses to specify RD MA devices. D ue to the
ubiquitous nature of using IP addresses and host names to specify machines, most RD MA
applications use this as their preferred, or in some cases only, way of specifying remote machines or
local devices to connect to.
To test the functionality of the IPoIB layer, it is possible to use any standard IP network test tool and
provide the IP address of the IPoIB devices to be tested. For example, the ping command between
the IP addresses of the IPoIB devices should now work.
There are two different RD MA performance testing packages included with Red Hat Enterprise Linux,
qperf and perftest. Either of these may be used to further test the performance of an RD MA network.
However, when using any of the applications that are part of the perftest package, or using the q p erf
application, there is a special note on address resolution. Even though the remote host is specified
using an IP address or host name of the IPoIB device, it is allowed for the test application to actually
connect through a different RD MA interface. The reason for this is because of the process of
converting from the host name or IP address to an RD MA address allows any valid RD MA address
pair between the two machines to be used. If there are multiple ways for the client to connect to the
server, then the programs might choose to use a different path if there is a problem with the path
specified. For example, if there are two ports on each machine connected to the same InfiniBand
subnet, and an IP address for the second port on each machine is given, it is likely that the program
will find the first port on each machine is a valid connection method and use them instead. In this
case, command-line options to any of the perftest programs can be used to tell them which card and
port to bind to, as was done with ib p in g in Section 9.5, “ Testing Early InfiniBand RD MA operation” ,
in order to ensure that testing occurs over the specific ports required to be tested. For q p erf , the
method of binding to ports is slightly different. The q p erf program operates as a server on one
machine, listening on all devices (including non-RD MA devices). The client may connect to q p erf
using any valid IP address or host name for the server. Q p erf will first attempt to open a data
connection and run the requested test(s) over the IP address or host name given on the client
command line, but if there is any problem using that address, q p erf will fall back to attempting to run
the test on any valid path between the client and server. For this reason, to force q p erf to test over a
specific link, use the -l o c_i d and -rem_i d options to the q p erf client in order to force the test to
run on a specific link.
9.11. Configure IPoIB Using a GUI
To configure an InfiniBand connection using a graphical tool, use the Netwo rk C o nnecti o ns
tool.
Pro ced u re 9 .1. Ad d in g a N ew In f in iB an d C o n n ect io n
1. To use the graphical N et wo rk C o n n ect io n s tool, press the Super key to enter the Activities
Overview, type Netwo rk C o nnecti o ns and then press Enter. The N et wo rk
C o n n ect io n s tool appears.
161
Red Hat Ent erprise Linux 7 Net working G uide
2. Click the Ad d button to open the selection list. Select Infi ni Band and then click C reate.
The Ed i ti ng Infi ni Band C o nnecti o n 1 window appears.
3. On the Infi ni Band tab, select the transport mode from the drop-down list you want to use
for the InfiniBand connection.
4. Enter the InfiniBand MAC address.
5. Review and confirm the settings and then click the Save button.
6. Edit the InfiniBand-specific settings by referring to Section 9.11.1, “ Configuring the InfiniBand
Tab” .
Pro ced u re 9 .2. Ed it in g an Exist in g In f in iB an d C o n n ect io n
Follow these steps to edit an existing InfiniBand connection.
1. Press the Super key to enter the Activities Overview, type Netwo rk C o nnecti o ns and then
press Enter. The N et wo rk C o n n ect io n s tool appears.
2. Select the connection you want to edit and click the Ed i t button.
3. Select the G eneral tab.
4. Configure the connection name, auto-connect behavior, and availability settings.
Five settings in the Ed i ti ng dialog are common to all connection types, see the G eneral
tab:
C o nnecti o n name — Enter a descriptive name for your network connection. This name
will be used to list this connection in the menu of the Netwo rk window.
Auto mati cal l y co nnect to thi s netwo rk when i t i s avai l abl e — Select
this box if you want N et wo rkMan ag er to auto-connect to this connection when it is
available. See Section 2.5.3, “ Connecting to a Network Automatically” for more
information.
Al l users may co nnect to thi s netwo rk — Select this box to create a connection
available to all users on the system. Changing this setting may require root privileges. See
Section 2.5.4, “ System-wide and Private Connection Profiles” for details.
Auto mati cal l y co nnect to VP N when usi ng thi s co nnecti o n — Select this
box if you want N et wo rkMan ag er to auto-connect to a VPN connection when it is
available. Select the VPN from the drop-down menu.
Fi rewal l Zo ne — Select the Firewall Z one from the drop-down menu. See the Red Hat
Enterprise Linux 7 Security Guide for more information on Firewall Z ones.
5. Edit the InfiniBand-specific settings by referring to the Section 9.11.1, “ Configuring the
InfiniBand Tab” .
Saving Your New (or Modified) Connect ion and Making Furt her Configurat ions
Once you have finished editing your InfiniBand connection, click the Save button to save your
customized configuration. If the profile was in use while being edited, power cycle the connection to
make N et wo rkMan ag er apply the changes. If the profile is OFF, set it to ON or select it in the
network connection icon's menu. See Section 2.5.1, “ Connecting to a Network Using a GUI” for
information on using your new or altered connection.
162
⁠Chapt er 9 . Configure InfiniBand and RDMA Net works
You can further configure an existing connection by selecting it in the Netwo rk C o nnecti o ns
window and clicking Ed i t to return to the Ed i ti ng dialog.
Then, to configure:
IPv4 settings for the connection, click the IP v4 Setti ng s tab and proceed to Section 2.5.10.4,
“ Configuring IPv4 Settings” ; or,
IPv6 settings for the connection, click the IP v6 Setti ng s tab and proceed to Section 2.5.10.5,
“ Configuring IPv6 Settings” .
9.11.1. Configuring t he InfiniBand T ab
If you have already added a new InfiniBand connection (refer to Procedure 9.1, “ Adding a New
InfiniBand Connection” for instructions), you can edit the Infi ni Band tab to set the parent interface
and the InfiniBand ID .
T ranspo rt mo d e
D atagram or Connected mode can be selected from the drop-down list. Select the same
mode the rest of your IPoIB network is using.
D evi ce MAC ad d ress
The MAC address of the InfiniBand capable device to be used for the InfiniBand network
traffic.This hardware address field will be pre-filled if you have InfiniBand hardware
installed.
MT U
Optionally sets a Maximum Transmission Unit (MTU) size to be used for packets to be sent
over the InfiniBand connection.
9.12. Addit ional Resources
The following sources of information provide additional resources regarding InfiniBand and RD MA
networking for Red Hat Enterprise Linux 7.
9.12.1. Inst alled Document at ion
/usr/share/d o c/i ni tscri pts-version/sysco nfi g . txt — D escribes configuration files
and their directives.
9.12.2. Online Document at ion
h t t p s://www.kern el.o rg /d o c/D o cu men t at io n /in f in ib an d /ip o ib .t xt
A description of the IPoIB driver. Includes references to relevant RFCs.
163
Red Hat Ent erprise Linux 7 Net working G uide
⁠Part III. Servers
This part discusses how to set up servers normally required for networking.
164
⁠Chapt er 1 0 . DHCP Servers
Chapter 10. DHCP Servers
D ynamic Host Configuration Protocol (D HCP) is a network protocol that automatically assigns
TCP/IP information to client machines. Each D HC P client connects to the centrally located D HC P
server, which returns the network configuration (including the IP address, gateway, and D NS
servers) of that client.
10.1. Why Use DHCP?
D HC P is useful for automatic configuration of client network interfaces. When configuring the client
system, you can choose D HC P instead of specifying an IP address, netmask, gateway, or D NS
servers. The client retrieves this information from the D HC P server. D HC P is also useful if you want to
change the IP addresses of a large number of systems. Instead of reconfiguring all the systems, you
can just edit one configuration file on the server for the new set of IP addresses. If the D NS servers for
an organization changes, the changes happen on the D HC P server, not on the D HC P clients. When
you restart the network or reboot the clients, the changes go into effect.
If an organization has a functional D HC P server correctly connected to a network, laptops and other
mobile computer users can move these devices from office to office.
Note that administrators of D NS and D HC P servers, as well as any provisioning applications, should
agree on the host name format used in an organization. See Section 3.1.1, “ Recommended Naming
Practices” for more information on the format of host names.
10.2. Configuring a DHCP Server
The dhcp package contains an Internet Systems Consortium (ISC) D HC P server. Install the package as
ro o t:
~]# yum i nstal l d hcp
Installing the dhcp package creates a file, /etc/d hcp/d hcpd . co nf, which is merely an empty
configuration file. As ro o t, issue the following command:
~]# cat /etc/d hcp/d hcpd . co nf
#
# DHCP Server Configuration file.
#
see /usr/share/doc/dhcp*/dhcpd.conf.example
#
see dhcpd.conf(5) man page
#
The example configuration file can be found at
/usr/share/d o c/d hcp-version;/d hcpd . co nf. exampl e. You should use this file to help you
configure /etc/d hcp/d hcpd . co nf, which is explained in detail below.
D HC P also uses the file /var/l i b/d hcpd /d hcpd . l eases to store the client lease database. Refer
to Section 10.2.2, “ Lease D atabase” for more information.
10.2.1. Configurat ion File
The first step in configuring a D HC P server is to create the configuration file that stores the network
information for the clients. Use this file to declare options for client systems.
165
Red Hat Ent erprise Linux 7 Net working G uide
The configuration file can contain extra tabs or blank lines for easier formatting. Keywords are caseinsensitive and lines beginning with a hash sign (#) are considered comments.
There are two types of statements in the configuration file:
Parameters — State how to perform a task, whether to perform a task, or what network
configuration options to send to the client.
D eclarations — D escribe the topology of the network, describe the clients, provide addresses for
the clients, or apply a group of parameters to a group of declarations.
The parameters that start with the keyword option are referred to as options. These options control
D HC P options; whereas, parameters configure values that are not optional or control how the D HC P
server behaves.
Parameters (including options) declared before a section enclosed in curly brackets ({ }) are
considered global parameters. Global parameters apply to all the sections below it.
Important
If the configuration file is changed, the changes do not take effect until the D HC P daemon is
restarted with the command systemctl restart d hcpd .
Note
Instead of changing a D HC P configuration file and restarting the service each time, using the
o mshel l command provides an interactive way to connect to, query, and change the
configuration of a D HC P server. By using o mshel l , all changes can be made while the server
is running. For more information on o mshel l , see the o mshel l man page.
In Example 10.1, “ Subnet D eclaration” , the ro uters, subnet-mask, d o mai n-search, d o mai nname-servers, and ti me-o ffset options are used for any ho st statements declared below it.
For every subnet which will be served, and for every subnet to which the D HC P server is connected,
there must be one subnet declaration, which tells the D HC P daemon how to recognize that an
address is on that subnet. A subnet declaration is required for each subnet even if no addresses will
be dynamically allocated to that subnet.
In this example, there are global options for every D HC P client in the subnet and a rang e declared.
Clients are assigned an IP address within the rang e.
Examp le 10.1. Su b n et D eclarat io n
subnet 192.168.1.0 netmask 255.255.255.0
option routers
option subnet-mask
option domain-search
option domain-name-servers
option time-offset
Time
range 192.168.1.10 192.168.1.100;
}
166
{
192.168.1.254;
255.255.255.0;
"example.com";
192.168.1.1;
-18000;
# Eastern Standard
⁠Chapt er 1 0 . DHCP Servers
To configure a D HC P server that leases a dynamic IP address to a system within a subnet, modify
the example values from Example 10.2, “ Range Parameter” . It declares a default lease time, maximum
lease time, and network configuration values for the clients. This example assigns IP addresses in
the rang e 19 2. 16 8. 1. 10 and 19 2. 16 8. 1. 10 0 to client systems.
Examp le 10.2. R an g e Paramet er
default-lease-time 600;
max-lease-time 7200;
option subnet-mask 255.255.255.0;
option broadcast-address 192.168.1.255;
option routers 192.168.1.254;
option domain-name-servers 192.168.1.1, 192.168.1.2;
option domain-search "example.com";
subnet 192.168.1.0 netmask 255.255.255.0 {
range 192.168.1.10 192.168.1.100;
}
To assign an IP address to a client based on the MAC address of the network interface card, use the
hard ware ethernet parameter within a ho st declaration. As demonstrated in Example 10.3, “ Static
IP Address Using D HCP” , the ho st apex declaration specifies that the network interface card with
the MAC address 0 0 : A0 : 78: 8E: 9 E: AA always receives the IP address 19 2. 16 8. 1. 4 .
Note that you can also use the optional parameter ho st-name to assign a host name to the client.
Examp le 10.3. St at ic IP Ad d ress U sin g D H C P
host apex {
option host-name "apex.example.com";
hardware ethernet 00:A0:78:8E:9E:AA;
fixed-address 192.168.1.4;
}
Red Hat Enterprise Linux 7 supports assigning static IP addresses to InfiniBand IPoIB interfaces.
However, as these interfaces do not have a normal hardware Ethernet address, a different method of
specifying a unique identifier for the IPoIB interface must be used. The standard is to use the option
d hcp-cl i ent-i d enti fi er= construct to specify the IPoIB interface’s d hcp-cl i enti d enti fi er field. The D HC P server host construct supports at most one hardware Ethernet and one
d hcp-cl i ent-i d enti fi er entry per host stanza. However, there may be more than one fixedaddress entry and the D HC P server will automatically respond with an address that is appropriate for
the network that the D HC P request was received on.
Examp le 10.4 . St at ic IP Ad d ress U sin g D H C P o n Mu lt ip le In t erf aces
If a machine has a complex configuration, for example two InfiniBand interfaces, and P _Key
interfaces on each physical interface, plus an Ethernet connection, the following static IP
construct could be used to serve this configuration:
Host apex.0 {
167
Red Hat Ent erprise Linux 7 Net working G uide
option host-name “apex.example.com”;
hardware ethernet 00:A0:78:8E:9E:AA;
option dhcp-clientidentifier=ff:00:00:00:00:00:02:00:00:02:c9:00:00:02:c9:03:00:31:7b:11;
fixed-address 172.31.0.50,172.31.2.50,172.31.1.50,172.31.3.50;
}
host apex.1 {
option host-name “apex.example.com”;
hardware ethernet 00:A0:78:8E:9E:AB;
option dhcp-clientidentifier=ff:00:00:00:00:00:02:00:00:02:c9:00:00:02:c9:03:00:31:7b:12;
fixed-address 172.31.0.50,172.31.2.50,172.31.1.50,172.31.3.50;
}
In order to find the right d hcp-cl i ent-i d enti fi er for your device, you can usually use the
prefix ff: 0 0 : 0 0 : 0 0 : 0 0 : 0 0 : 0 2: 0 0 : 0 0 : 0 2: c9 : 0 0 and then add the last 8 bytes of the
IPoIB interface (which happens to also be the 8 byte GUID of the InfiniBand port the IPoIB
interface is on). On some older controllers, this prefix is not correct. In that case, we recommend
using t cp d u mp on the D HC P server to capture the incoming IPoIB D HC P request and gather the
right d hcp-cl i ent-i d enti fi er from that capture. For example:
]$ tcpd ump -vv -i ml x4 _i b0
tcpdump: listening on mlx4_ib0, link-type LINUX_SLL (Linux cooked),
capture size 65535 bytes
23:42:44.131447 IP (tos 0x10, ttl 128, id 0, offset 0, flags [none],
proto UDP (17), length 328)
0.0.0.0.bootpc > 255.255.255.255.bootps: [udp sum ok] BOOTP/DHCP,
Request, length 300, htype 32, hlen 0, xid 0x975cb024, Flags
[Broadcast] (0x8000)
Vendor-rfc1048 Extensions
Magic Cookie 0x63825363
DHCP-Message Option 53, length 1: Discover
Hostname Option 12, length 10: "rdma-qe-03"
Parameter-Request Option 55, length 18:
Subnet-Mask, BR, Time-Zone, Classless-Static-Route
Domain-Name, Domain-Name-Server, Hostname, YD
YS, NTP, MTU, Option 119
Default-Gateway, Classless-Static-Route, ClasslessStatic-Route-Microsoft, Static-Route
Option 252, NTP
Client-ID Option 61, length 20: hardware-type 255,
00:00:00:00:00:02:00:00:02:c9:00:00:02:c9:02:00:21:ac:c1
The above dump shows the Client-ID field. The hardware-type 255 corresponds to the initial ff:
of the ID , the rest of the ID is then quoted exactly as it needs to appear in the D HC P configuration
file.
All subnets that share the same physical network should be declared within a shared -netwo rk
declaration as shown in Example 10.5, “ Shared-network D eclaration” . Parameters within the
shared -netwo rk, but outside the enclosed subnet declarations, are considered to be global
parameters. The name assigned to shared -netwo rk must be a descriptive title for the network, such
as using the title “ test-lab” to describe all the subnets in a test lab environment.
168
⁠Chapt er 1 0 . DHCP Servers
Examp le 10.5. Sh ared - n et wo rk D eclarat io n
shared-network name {
option domain-search
"test.redhat.com";
option domain-name-servers
ns1.redhat.com, ns2.redhat.com;
option routers
192.168.0.254;
#more parameters for EXAMPLE shared-network
subnet 192.168.1.0 netmask 255.255.252.0 {
#parameters for subnet
range 192.168.1.1 192.168.1.254;
}
subnet 192.168.2.0 netmask 255.255.252.0 {
#parameters for subnet
range 192.168.2.1 192.168.2.254;
}
}
As demonstrated in Example 10.6, “ Group D eclaration” , the g ro up declaration is used to apply
global parameters to a group of declarations. For example, shared networks, subnets, and hosts can
be grouped.
Examp le 10.6 . G ro u p D eclarat io n
group {
option routers
192.168.1.254;
option subnet-mask
255.255.255.0;
option domain-search
"example.com";
option domain-name-servers
192.168.1.1;
option time-offset
-18000;
# Eastern Standard Time
host apex {
option host-name "apex.example.com";
hardware ethernet 00:A0:78:8E:9E:AA;
fixed-address 192.168.1.4;
}
host raleigh {
option host-name "raleigh.example.com";
hardware ethernet 00:A1:DD:74:C3:F2;
fixed-address 192.168.1.6;
}
}
169
Red Hat Ent erprise Linux 7 Net working G uide
Note
You can use the provided example configuration file as a starting point and add custom
configuration options to it. To copy this file to the proper location, use the following command
as ro o t:
~]# cp /usr/share/d o c/d hcp-version_number/d hcpd . co nf. exampl e
/etc/d hcp/d hcpd . co nf
... where version_number is the D HC P version number.
For a complete list of option statements and what they do, see the d hcp-o pti o ns(5) man page.
10.2.2. Lease Dat abase
On the D HC P server, the file /var/l i b/d hcpd /d hcpd . l eases stores the D HC P client lease
database. D o not change this file. D HC P lease information for each recently assigned IP address is
automatically stored in the lease database. The information includes the length of the lease, to whom
the IP address has been assigned, the start and end dates for the lease, and the MAC address of the
network interface card that was used to retrieve the lease.
All times in the lease database are in Coordinated Universal Time (UTC), not local time.
The lease database is recreated from time to time so that it is not too large. First, all known leases are
saved in a temporary lease database. The d hcpd . l eases file is renamed d hcpd . l eases~ and
the temporary lease database is written to d hcpd . l eases.
The D HC P daemon could be killed or the system could crash after the lease database has been
renamed to the backup file but before the new file has been written. If this happens, the
d hcpd . l eases file does not exist, but it is required to start the service. D o not create a new lease
file. If you do, all old leases are lost which causes many problems. The correct solution is to rename
the d hcpd . l eases~ backup file to d hcpd . l eases and then start the daemon.
10.2.3. St art ing and St opping t he Server
Important
When the D HC P server is started for the first time, it fails unless the d hcpd . l eases file exists.
You can use the command to uch /var/l i b/d hcpd /d hcpd . l eases to create the file if it
does not exist. If the same server is also running BIND as a D NS server, this step is not
necessary, as starting the named service automatically checks for a d hcpd . l eases file.
D o not create a new lease file on a system that was previously running. If you do, all old
leases are lost which causes many problems. The correct solution is to rename the
d hcpd . l eases~ backup file to d hcpd . l eases and then start the daemon.
To start the D HC P service, use the following command:
systemctl start d hcpd . servi ce
170
⁠Chapt er 1 0 . DHCP Servers
To stop the D HC P server, type:
systemctl sto p d hcpd . servi ce
By default, the D HC P service does not start at boot time. For information on how to configure the
daemon to start automatically at boot time, see Red Hat Enterprise Linux 7 System Administrator's Guide.
If more than one network interface is attached to the system, but the D HC P server should only listen
for D HC P requests on one of the interfaces, configure the D HC P server to listen only on that device.
The D HC P daemon will only listen on interfaces for which it finds a subnet declaration in the
/etc/d hcp/d hcpd . co nf file.
This is useful for a firewall machine with two network cards. One network card can be configured as a
D HC P client to retrieve an IP address to the Internet. The other network card can be used as a D HC P
server for the internal network behind the firewall. Specifying only the network card connected to the
internal network makes the system more secure because users can not connect to the daemon via the
Internet.
To specify command-line options, copy and then edit the d hcpd . servi ce file as the ro o t user. For
example, as follows:
~]# cp /usr/l i b/systemd /system/d hcpd . servi ce /etc/systemd /system/
~]# vi /etc/systemd /system/d hcpd . servi ce
Edit the line under section [Service]:
ExecStart=/usr/sbin/dhcpd -f -cf /etc/dhcp/dhcpd.conf -user dhcpd -group
dhcpd --no-pid your_interface_name(s)
Then, as the ro o t user, restart the service:
~]# systemctl --system d aemo n-rel o ad
~]# systemctl restart d hcpd
Command line options can be appended to ExecStart= /usr/sbi n/d hcpd in the
/etc/systemd /system/d hcpd . servi ce unit file under section [Service]. They include:
-p portnum — Specifies the UD P port number on which d hcpd should listen. The default is port
67. The D HC P server transmits responses to the D HC P clients at a port number one greater than
the UD P port specified. For example, if the default port 67 is used, the server listens on port 67 for
requests and responds to the client on port 68. If a port is specified here and the D HC P relay
agent is used, the same port on which the D HC P relay agent should listen must be specified. See
Section 10.3, “ D HCP Relay Agent” for details.
-f — Runs the daemon as a foreground process. This is mostly used for debugging.
-d — Logs the D HC P server daemon to the standard error descriptor. This is mostly used for
debugging. If this is not specified, the log is written to /var/l o g /messag es.
-cf filename — Specifies the location of the configuration file. The default location is
/etc/d hcp/d hcpd . co nf.
-l f filename — Specifies the location of the lease database file. If a lease database file
already exists, it is very important that the same file be used every time the D HC P server is started.
It is strongly recommended that this option only be used for debugging purposes on nonproduction machines. The default location is /var/l i b/d hcpd /d hcpd . l eases.
171
Red Hat Ent erprise Linux 7 Net working G uide
-q — D o not print the entire copyright message when starting the daemon.
10.3. DHCP Relay Agent
The D HCP Relay Agent (d h crelay) enables the relay of D HC P and BO O T P requests from a subnet
with no D HC P server on it to one or more D HC P servers on other subnets.
When a D HC P client requests information, the D HCP Relay Agent forwards the request to the list of
D HC P servers specified when the D HCP Relay Agent is started. When a D HC P server returns a reply,
the reply is broadcast or unicast on the network that sent the original request.
The D HCP Relay Agent for IP v4 , d h crelay, listens for D HC P v4 and BO O T P requests on all
interfaces unless the interfaces are specified in /etc/sysco nfi g /d hcrel ay with the INT ER FAC ES
directive. See Section 10.3.2, “ Configure dhcrelay as a D HCPv6 relay agent” . The D HCP Relay Agent
for IP v6 , d h crelay6 , does not have this default behavior and interfaces to listen for D HC P v6
requests must be specified. See Section 10.3.2, “ Configure dhcrelay as a D HCPv6 relay agent” .
d h crelay can either be run as a D HC P v4 and BO O T P relay agent (by default) or as a D HC P v6 relay
agent (with -6 argument). To see the usage message, issue the command d hcrel ay -h.
10.3.1. Configure dhcrelay as a DHCPv4 and BOOT P relay agent
To run d h crelay in D HC P v4 and BO O T P mode specify the servers to which the requests should be
forwarded to. Copy and then edit the d hcrel ay. servi ce file as the ro o t user:
~]# cp /l i b/systemd /system/d hcrel ay. servi ce /etc/systemd /system/
~]# vi /etc/systemd /system/d hcrel ay. servi ce
Edit the ExecStart option under section [Service] and add one or more server IP v4 addresses to
the end of the line, for example:
ExecStart=/usr/sbin/dhcrelay -d --no-pid 192.168.1.1
If you also want to specify interfaces where the D HCP Relay Agent listens for D HC P requests, add
them to the ExecStart option with -i argument (otherwise it will listen on all interfaces), for example:
ExecStart=/usr/sbin/dhcrelay -d --no-pid 192.168.1.1 -i em1
For other options see the d hcrel ay(8) man page.
To activate the changes made, as the ro o t user, restart the service:
~]# systemctl --system d aemo n-rel o ad
~]# systemctl restart d hcrel ay
10.3.2. Configure dhcrelay as a DHCPv6 relay agent
To run d h crelay in D HC P v6 mode add the -6 argument and specify the “ lower interface” (on which
queries will be received from clients or from other relay agents) and the “ upper interface” (to which
queries from clients and other relay agents should be forwarded). Copy d hcrel ay. servi ce to
d hcrel ay6 . servi ce and edit it as the ro o t user:
172
⁠Chapt er 1 0 . DHCP Servers
~]# cp /l i b/systemd /system/d hcrel ay. servi ce
/etc/systemd /system/d hcrel ay6 . servi ce
~]# vi /etc/systemd /system/d hcrel ay6 . servi ce
Edit the ExecStart option under section [Service] add -6 argument and add the “ lower interface”
and “ upper interface” interface, for example:
ExecStart=/usr/sbin/dhcrelay -d --no-pid -6 -l em1 -u em2
For other options see the d hcrel ay(8) man page.
To activate the changes made, as the ro o t user, restart the service:
~]# systemctl --system d aemo n-rel o ad
~]# systemctl restart d hcrel ay6
10.4 . Configuring a Mult ihomed DHCP Server
A multihomed D HC P server serves multiple networks, that is, multiple subnets. The examples in these
sections detail how to configure a D HC P server to serve multiple networks, select which network
interfaces to listen on, and how to define network settings for systems that move networks.
Before making any changes, back up the existing /etc/d hcp/d hcpd . co nf file.
The D HC P daemon will only listen on interfaces for which it finds a subnet declaration in the
/etc/d hcp/d hcpd . co nf file.
The following is a basic /etc/d hcp/d hcpd . co nf file, for a server that has two network interfaces,
eth0 in a 10 . 0 . 0 . 0 /24 network, and eth1 in a 172. 16 . 0 . 0 /24 network. Multiple subnet
declarations allow you to define different settings for multiple networks:
default-lease-time 600;
max-lease-time 7200;
subnet 10.0.0.0 netmask 255.255.255.0 {
option subnet-mask 255.255.255.0;
option routers 10.0.0.1;
range 10.0.0.5 10.0.0.15;
}
subnet 172.16.0.0 netmask 255.255.255.0 {
option subnet-mask 255.255.255.0;
option routers 172.16.0.1;
range 172.16.0.5 172.16.0.15;
}
subnet 10.0.0.0 netmask 255.255.255.0;
A subnet declaration is required for every network your D HC P server is serving. Multiple
subnets require multiple subnet declarations. If the D HC P server does not have a network
interface in a range of a subnet declaration, the D HC P server does not serve that network.
If there is only one subnet declaration, and no network interfaces are in the range of that
subnet, the D HC P daemon fails to start, and an error such as the following is logged to
/var/l o g /messag es:
173
Red Hat Ent erprise Linux 7 Net working G uide
dhcpd: No subnet declaration for eth0 (0.0.0.0).
dhcpd: ** Ignoring requests on eth0. If this is not what
dhcpd:
you want, please write a subnet declaration
dhcpd:
in your dhcpd.conf file for the network segment
dhcpd:
to which interface eth1 is attached. **
dhcpd:
dhcpd:
dhcpd: Not configured to listen on any interfaces!
o pti o n subnet-mask 255.255.255.0;
The o pti o n subnet-mask option defines a subnet mask, and overrides the netmask
value in the subnet declaration. In simple cases, the subnet and netmask values are the
same.
o pti o n ro uters 10.0.0.1;
The o pti o n ro uters option defines the default gateway for the subnet. This is required
for systems to reach internal networks on a different subnet, as well as external networks.
rang e 10.0.0.5 10.0.0.15;
The rang e option specifies the pool of available IP addresses. Systems are assigned an
address from the range of specified IP addresses.
For further information, see the d hcpd . co nf(5) man page.
10.4 .1. Host Configurat ion
Before making any changes, back up the existing /etc/sysco nfi g /d hcpd and
/etc/d hcp/d hcpd . co nf files.
C o n f ig u rin g a Sin g le Syst em f o r Mu lt ip le N et wo rks
The following /etc/d hcp/d hcpd . co nf example creates two subnets, and configures an IP
address for the same system, depending on which network it connects to:
default-lease-time 600;
max-lease-time 7200;
subnet 10.0.0.0 netmask 255.255.255.0 {
option subnet-mask 255.255.255.0;
option routers 10.0.0.1;
range 10.0.0.5 10.0.0.15;
}
subnet 172.16.0.0 netmask 255.255.255.0 {
option subnet-mask 255.255.255.0;
option routers 172.16.0.1;
range 172.16.0.5 172.16.0.15;
}
host example0 {
hardware ethernet 00:1A:6B:6A:2E:0B;
fixed-address 10.0.0.20;
}
174
⁠Chapt er 1 0 . DHCP Servers
host example1 {
hardware ethernet 00:1A:6B:6A:2E:0B;
fixed-address 172.16.0.20;
}
ho st example0
The ho st declaration defines specific parameters for a single system, such as an IP
address. To configure specific parameters for multiple hosts, use multiple ho st
declarations.
Most D HC P clients ignore the name in ho st declarations, and as such, this name can be
anything, as long as it is unique to other ho st declarations. To configure the same system
for multiple networks, use a different name for each ho st declaration, otherwise the D HC P
daemon fails to start. Systems are identified by the hard ware ethernet option, not the
name in the ho st declaration.
hard ware ethernet 00:1A:6B:6A:2E:0B;
The hard ware ethernet option identifies the system. To find this address, run the i p
l i nk command.
fi xed -ad d ress 10.0.0.20;
The fi xed -ad d ress option assigns a valid IP address to the system specified by the
hard ware ethernet option. This address must be outside the IP address pool specified
with the rang e option.
If o pti o n statements do not end with a semicolon, the D HC P daemon fails to start, and an error such
as the following is logged to /var/l o g /messag es:
/etc/dhcp/dhcpd.conf line 20: semicolon expected.
dhcpd: }
dhcpd: ^
dhcpd: /etc/dhcp/dhcpd.conf line 38: unexpected end of file
dhcpd:
dhcpd: ^
dhcpd: Configuration file errors encountered -- exiting
C o n f ig u rin g Syst ems wit h Mu lt ip le N et wo rk In t erf aces
The following ho st declarations configure a single system, which has multiple network interfaces, so
that each interface receives the same IP address. This configuration will not work if both network
interfaces are connected to the same network at the same time:
host interface0 {
hardware ethernet 00:1a:6b:6a:2e:0b;
fixed-address 10.0.0.18;
}
host interface1 {
hardware ethernet 00:1A:6B:6A:27:3A;
fixed-address 10.0.0.18;
}
For this example, i nterface0 is the first network interface, and i nterface1 is the second interface.
The different hard ware ethernet options identify each interface.
175
Red Hat Ent erprise Linux 7 Net working G uide
If such a system connects to another network, add more ho st declarations, remembering to:
assign a valid fi xed -ad d ress for the network the host is connecting to.
make the name in the ho st declaration unique.
When a name given in a ho st declaration is not unique, the D HC P daemon fails to start, and an error
such as the following is logged to /var/l o g /messag es:
dhcpd:
dhcpd:
dhcpd:
dhcpd:
/etc/dhcp/dhcpd.conf line 31: host interface0: already exists
}
^
Configuration file errors encountered -- exiting
This error was caused by having multiple ho st i nterface0 declarations defined in
/etc/d hcp/d hcpd . co nf.
10.5. DHCP for IPv6 (DHCPv6)
The ISC D HC P includes support for IP v6 (D HC P v6 ) since the 4.x release with a D HC P v6 server,
client, and relay agent functionality. The agents support both IP v4 and IP v6 , however the agents
can only manage one protocol at a time; for dual support they must be started separately for IP v4
and IP v6 . For example, configure both D HC P v4 and D HC P v6 by editing their respective
configuration files /etc/d hcp/d hcpd . co nf and /etc/d hcp/d hcpd 6 . co nf and then issue the
following commands:
~]# systemctl start d hcpd
~]# systemctl start d hcpd 6
The D HC P v6 server configuration file can be found at /etc/d hcp/d hcpd 6 . co nf.
The example server configuration file can be found at
/usr/share/d o c/d hcp-version/d hcpd 6 . co nf. exampl e.
A simple D HC P v6 server configuration file can look like this:
subnet6 2001:db8:0:1::/64 {
range6 2001:db8:0:1::129 2001:db8:0:1::254;
option dhcp6.name-servers fec0:0:0:1::1;
option dhcp6.domain-search "domain.example";
}
10.6. Addit ional Resources
The following sources of information provide additional resources regarding D HC P .
10.6.1. Inst alled Document at ion
d hcpd (8) man page — D escribes how the D HC P daemon works.
d hcpd . co nf(5) man page — Explains how to configure the D HC P configuration file; includes
some examples.
d hcpd . l eases(5) man page — D escribes a persistent database of leases.
176
⁠Chapt er 1 0 . DHCP Servers
d hcp-o pti o ns(5) man page — Explains the syntax for declaring D HC P options in
d hcpd . co nf; includes some examples.
d hcrel ay(8) man page — Explains the D HC P Relay Agent and its configuration options.
/usr/share/d o c/d hcp-version/ — Contains example files, READ ME files, and release notes
for current versions of the D HC P service.
177
Red Hat Ent erprise Linux 7 Net working G uide
Chapter 11. DNS Servers
D NS (D omain Name System), is a distributed database system that is used to associate host names
with their respective IP addresses. For users, this has the advantage that they can refer to machines
on the network by names that are usually easier to remember than the numerical network addresses.
For system administrators, using a D NS server, also known as a nameserver, enables changing the
IP address for a host without ever affecting the name-based queries. The use of the D NS databases
is not only for resolving IP addresses to domain names and their use is becoming broader and
broader as D NSSEC is deployed.
11.1. Int roduct ion t o DNS
D NS is usually implemented using one or more centralized servers that are authoritative for certain
domains. When a client host requests information from a nameserver, it usually connects to port 53.
The nameserver then attempts to resolve the name requested. If the nameserver is configured to be a
recursive name servers and it does not have an authoritative answer, or does not already have the
answer cached from an earlier query, it queries other nameservers, called root nameservers, to
determine which nameservers are authoritative for the name in question, and then queries them to get
the requested name. Nameservers configured as purely authoritative, with recursion disabled, will not
do lookups on behalf of clients.
11.1.1. Nameserver Zones
In a D NS server, all information is stored in basic data elements called resource records (RR).
Resource records are defined in RFC 1034. The domain names are organized into a tree structure.
Each level of the hierarchy is divided by a period (. ). For example: The root domain, denoted by . , is
the root of the D NS tree, which is at level zero. The domain name co m, referred to as the top-level
domain (TLD ) is a child of the root domain (. ) so it is the first level of the hierarchy. The domain name
exampl e. co m is at the second level of the hierarchy.
Examp le 11.1. A Simp le R eso u rce R eco rd
An example of a simple resource record (RR):
example.com.
86400
IN
A
192.0.2.1
The domain name, exampl e. co m, is the owner for the RR. The value 86 4 0 0 is the time to live
(TTL). The letters IN, meaning “ the Internet system” , indicate the class of the RR. The letter A
indicates the type of RR (in this example, a host address). The host address 19 2. 0 . 2. 1 is the
data contained in the final section of this RR. This one line example is a RR. A set of RRs with the
same type, owner, and class is called a resource record set (RRSet).
Z ones are defined on authoritative nameservers through the use of zone files, which contain
definitions of the resource records in each zone. Z one files are stored on primary nameservers (also
called master nameservers), where changes are made to the files, and secondary nameservers (also
called slave nameservers), which receive zone definitions from the primary nameservers. Both primary
and secondary nameservers are authoritative for the zone and look the same to clients. D epending
on the configuration, any nameserver can also serve as a primary or secondary server for multiple
zones at the same time.
178
⁠Chapt er 1 1 . DNS Servers
Note that administrators of D NS and D HC P servers, as well as any provisioning applications, should
agree on the host name format used in an organization. See Section 3.1.1, “ Recommended Naming
Practices” for more information on the format of host names.
11.1.2. Nameserver T ypes
There are two nameserver configuration types:
au t h o rit at ive
Authoritative nameservers answer to resource records that are part of their zones only. This
category includes both primary (master) and secondary (slave) nameservers.
recu rsive
Recursive nameservers offer resolution services, but they are not authoritative for any zone.
Answers for all resolutions are cached in a memory for a fixed period of time, which is
specified by the retrieved resource record.
Although a nameserver can be both authoritative and recursive at the same time, it is recommended
not to combine the configuration types. To be able to perform their work, authoritative servers should
be available to all clients all the time. On the other hand, since the recursive lookup takes far more
time than authoritative responses, recursive servers should be available to a restricted number of
clients only, otherwise they are prone to distributed denial of service (D D oS) attacks.
11.1.3. BIND as a Nameserver
BIND consists of a set of D NS-related programs. It contains a nameserver called named , an
administration utility called rnd c, and a debugging tool called d i g . See Red Hat Enterprise Linux 7
System Administrator's Guide for more information on how to run a service in Red Hat Enterprise Linux.
11.2. BIND
This section covers BIND (Berkeley Internet Name D omain), the D NS server included in Red Hat
Enterprise Linux. It focuses on the structure of its configuration files, and describes how to administer
it both locally and remotely.
11.2.1. Empt y Zones
BIND configures a number of “ empty zones” to prevent recursive servers from sending unnecessary
queries to Internet servers that cannot handle them (thus creating delays and SERVFAIL responses to
clients who query for them). These empty zones ensure that immediate and authoritative NXD OMAIN
responses are returned instead. The configuration option empty-zo nes-enabl e controls whether
or not empty zones are created, whilst the option d i sabl e-empty-zo ne can be used in addition to
disable one or more empty zones from the list of default prefixes that would be used.
The number of empty zones created for RFC 1918 prefixes has been increased, and users of BIND
9 . 9 and above will see the RFC 1918 empty zones both when empty-zo nes-enabl e is unspecified
(defaults to yes), and when it is explicitly set to yes.
11.2.2. Configuring t he named Service
When the named service is started, it reads the configuration from the files as described in Table 11.1,
“ The named Service Configuration Files” .
179
Red Hat Ent erprise Linux 7 Net working G uide
T ab le 11.1. T h e n amed Service C o n f ig u rat io n Files
Pat h
D escrip t io n
/etc/named . co nf
/etc/named /
The main configuration file.
An auxiliary directory for configuration files that are included
in the main configuration file.
The configuration file consists of a collection of statements with nested options surrounded by
opening and closing curly brackets ({ and }). Note that when editing the file, you have to be careful
not to make any syntax error, otherwise the named service will not start. A typical /etc/named . co nf
file is organized as follows:
statement-1 ["statement-1-name"] [statement-1-class] {
option-1;
option-2;
option-N;
};
statement-2 ["statement-2-name"] [statement-2-class] {
option-1;
option-2;
option-N;
};
statement-N ["statement-N-name"] [statement-N-class] {
option-1;
option-2;
option-N;
};
180
⁠Chapt er 1 1 . DNS Servers
Note
If you have installed the bind-chroot package, the BIND service will run in the chro o t
environment. In that case, the initialization script will mount the above configuration files using
the mo unt --bi nd command, so that you can manage the configuration outside this
environment. There is no need to copy anything into the /var/named /chro o t/ directory
because it is mounted automatically. This simplifies maintenance since you do not need to
take any special care of BIND configuration files if it is run in a chro o t environment. You can
organize everything as you would with BIND not running in a chro o t environment.
The following directories are automatically mounted into /var/named /chro o t/ if they are
empty in the /var/named /chro o t/ directory. They must be kept empty if you want them to be
mounted into /var/named /chro o t/:
/etc/named
/etc/pki /d nssec-keys
/run/named
/var/named
/usr/l i b6 4 /bi nd or /usr/l i b/bi nd (architecture dependent).
The following files are also mounted if the target file does not exist in /var/named /chro o t/:
/etc/named . co nf
/etc/rnd c. co nf
/etc/rnd c. key
/etc/named . rfc19 12. zo nes
/etc/named . d nssec. keys
/etc/named . i scd l v. key
/etc/named . ro o t. key
Important
Editing files which have been mounted in a chro o t environment requires creating a backup
copy and then editing the original file. Alternatively, use an editor with “ edit-a-copy” mode
disabled. For example, to edit the BIND 's configuration file, /etc/named . co nf, with Vim
while it is running in a chro o t environment, issue the following command as ro o t:
~]# vi m -c "set backupco py= yes" /etc/named . co nf
1 1 .2 .2 .1 . Inst alling BIND in a chro o t Enviro nm e nt
To install B IN D to run in a chro o t environment, issue the following command as ro o t:
~]# yum i nstal l bi nd -chro o t
To enable the named -chro o t service, first check if the named service is running by issuing the
following command:
~]$ systemctl status named
181
Red Hat Ent erprise Linux 7 Net working G uide
If it is running, it must be disabled.
To disable named , issue the following commands as ro o t:
~]# systemctl sto p named
~]# systemctl d i sabl e named
Then, to enable the named -chro o t service, issue the following commands as ro o t:
~]# systemctl enabl e named -chro o t
~]# systemctl start named -chro o t
To check the status of the named -chro o t service, issue the following command as ro o t:
~]# systemctl status named -chro o t
1 1 .2 .2 .2 . Co m m o n St at e m e nt T ype s
The following types of statements are commonly used in /etc/named . co nf:
acl
The acl (Access Control List) statement allows you to define groups of hosts, so that they
can be permitted or denied access to the nameserver. It takes the following form:
acl acl-name {
match-element;
...
};
The acl-name statement name is the name of the access control list, and the match-element
option is usually an individual IP address (such as 10 . 0 . 1. 1) or a Classless InterDomain Routing (CID R) network notation (for example, 10 . 0 . 1. 0 /24 ). For a list of already
defined keywords, see Table 11.2, “ Predefined Access Control Lists” .
T ab le 11.2. Pred ef in ed Access C o n t ro l List s
K eywo rd
D escrip t io n
any
l o cal ho st
l o cal nets
Matches every IP address.
Matches any IP address that is in use by the local system.
Matches any IP address on any network to which the local system
is connected.
D oes not match any IP address.
no ne
The acl statement can be especially useful in conjunction with other statements such as
o pti o ns. Example 11.2, “ Using acl in Conjunction with Options” defines two access
control lists, bl ack-hats and red -hats, and adds bl ack-hats on the blacklist while
granting red -hats normal access.
Examp le 11.2. U sin g acl in C o n ju n ct io n wit h O p t io n s
182
⁠Chapt er 1 1 . DNS Servers
acl black-hats {
10.0.2.0/24;
192.168.0.0/24;
1234:5678::9abc/24;
};
acl red-hats {
10.0.1.0/24;
};
options {
blackhole { black-hats; };
allow-query { red-hats; };
allow-query-cache { red-hats; };
};
i ncl ud e
The i ncl ud e statement allows you to include files in the /etc/named . co nf, so that
potentially sensitive data can be placed in a separate file with restricted permissions. It
takes the following form:
include "file-name"
The file-name statement name is an absolute path to a file.
Examp le 11.3. In clu d in g a File t o /et c/n amed .co n f
include "/etc/named.rfc1912.zones";
o pti o ns
The o pti o ns statement allows you to define global server configuration options as well as
to set defaults for other statements. It can be used to specify the location of the named
working directory, the types of queries allowed, and much more. It takes the following form:
options {
option;
...
};
For a list of frequently used option directives, see Table 11.3, “ Commonly Used
Configuration Options” below.
T ab le 11.3. C o mmo n ly U sed C o n f ig u rat io n O p t io n s
O p t io n
D escrip t io n
al l o w-q uery
Specifies which hosts are allowed to query the nameserver for
authoritative resource records. It accepts an access control list, a
collection of IP addresses, or networks in the CID R notation. All
hosts are allowed by default.
Specifies which hosts are allowed to query the nameserver for
non-authoritative data such as recursive queries. Only
l o cal ho st and l o cal nets are allowed by default.
al l o w-q uerycache
183
Red Hat Ent erprise Linux 7 Net working G uide
O p t io n
D escrip t io n
bl ackho l e
Specifies which hosts are not allowed to query the nameserver.
This option should be used when a particular host or network
floods the server with requests. The default option is no ne.
Specifies a working directory for the named service. The default
option is /var/named /.
Used to disable one or more empty zones from the list of default
prefixes that would be used. Can be specified in the options
statement and also in view statements. It can be used multiple
times.
Specifies whether to return D NSSEC related resource records. The
default option is yes.
Specifies whether to prove that resource records are authentic via
D NSSEC. The default option is yes.
Controls whether or not empty zones are created. Can be specified
only in the options statement.
Specifies a list of valid IP addresses for nameservers to which the
requests should be forwarded for resolution.
Specifies the behavior of the fo rward ers directive. It accepts the
following options:
d i recto ry
d i sabl e-emptyzo ne
d nssec-enabl e
d nssecval i d ati o n
empty-zo nesenabl e
fo rward ers
fo rward
fi rst — The server will query the nameservers listed in the
fo rward ers directive before attempting to resolve the name on
its own.
o nl y — When unable to query the nameservers listed in the
fo rward ers directive, the server will not attempt to resolve the
name on its own.
l i sten-o n
l i sten-o n-v6
max-cache-si ze
no ti fy
Specifies the IP v4 network interface on which to listen for queries.
On a D NS server that also acts as a gateway, you can use this
option to answer queries originating from a single network only.
All IP v4 interfaces are used by default.
Specifies the IP v6 network interface on which to listen for queries.
On a D NS server that also acts as a gateway, you can use this
option to answer queries originating from a single network only.
All IP v6 interfaces are used by default.
Specifies the maximum amount of memory to be used for server
caches. When the limit is reached, the server causes records to
expire prematurely so that the limit is not exceeded. In a server with
multiple views, the limit applies separately to the cache of each
view. The default option is 32M.
Specifies whether to notify the secondary nameservers when a
zone is updated. It accepts the following options:
yes — The server will notify all secondary nameservers.
no — The server will not notify any secondary nameserver.
master-o nl y — The server will notify primary server for the
zone only.
expl i ci t — The server will notify only the secondary servers
that are specified in the al so -no ti fy list within a zone
statement.
pi d -fi l e
184
Specifies the location of the process ID file created by the named
service.
⁠Chapt er 1 1 . DNS Servers
O p t io n
D escrip t io n
recursi o n
Specifies whether to act as a recursive server. The default option is
yes.
Specifies an alternate location for statistics files. The
/var/named /named . stats file is used by default.
stati sti cs-fi l e
Note
The directory used by named for runtime data has been moved from the BIND default
location, /var/run/named /, to a new location /run/named /. As a result, the PID
file has been moved from the default location /var/run/named /named . pi d to the
new location /run/named /named . pi d . In addition, the session-key file has been
moved to /run/named /sessi o n. key. These locations need to be specified by
statements in the options section. See Example 11.4, “ Using the options Statement” .
Important
To prevent distributed denial of service (D D oS) attacks, it is recommended that you
use the al l o w-q uery-cache option to restrict recursive D NS services for a
particular subset of clients only.
Refer to the BIND 9 Administrator Reference Manual referenced in Section 11.2.8.1, “ Installed
D ocumentation” , and the named . co nf manual page for a complete list of available
options.
Examp le 11.4 . U sin g t h e o p t io n s St at emen t
options {
allow-query
listen-on port
listen-on-v6 port
max-cache-size
directory
statistics-file
{ localhost; };
53 { 127.0.0.1; };
53 { ::1; };
256M;
"/var/named";
"/var/named/data/named_stats.txt";
recursion
yes;
dnssec-enable
yes;
dnssec-validation yes;
pid-file
session-keyfile
"/run/named/named.pid";
"/run/named/session.key";
};
zo ne
The zo ne statement allows you to define the characteristics of a zone, such as the location
of its configuration file and zone-specific options, and can be used to override the global
o pti o ns statements. It takes the following form:
185
Red Hat Ent erprise Linux 7 Net working G uide
zone zone-name [zone-class] {
option;
...
};
The zone-name attribute is the name of the zone, zone-class is the optional class of the zone,
and option is a zo ne statement option as described in Table 11.4, “ Commonly Used Options
in Z one Statements” .
The zone-name attribute is particularly important, as it is the default value assigned for the
$O R IG IN directive used within the corresponding zone file located in the /var/named /
directory. The named daemon appends the name of the zone to any non-fully qualified
domain name listed in the zone file. For example, if a zo ne statement defines the
namespace for exampl e. co m, use exampl e. co m as the zone-name so that it is placed at
the end of host names within the exampl e. co m zone file.
For more information about zone files, refer to Section 11.2.3, “ Editing Z one Files” .
T ab le 11.4 . C o mmo n ly U sed O p t io n s in Z o n e St at emen t s
O p t io n
D escrip t io n
al l o w-q uery
Specifies which clients are allowed to request information about
this zone. This option overrides global al l o w-q uery option. All
query requests are allowed by default.
Specifies which secondary servers are allowed to request a
transfer of the zone's information. All transfer requests are allowed
by default.
Specifies which hosts are allowed to dynamically update
information in their zone. The default option is to deny all dynamic
update requests.
al l o w-transfer
al l o w-upd ate
Note that you should be careful when allowing hosts to update
information about their zone. D o not set IP addresses in this
option unless the server is in the trusted network. Instead, use
TSIG key as described in Section 11.2.6.3, “ Transaction
SIGnatures (TSIG)” .
fi l e
masters
no ti fy
Specifies the name of the file in the named working directory that
contains the zone's configuration data.
Specifies from which IP addresses to request authoritative zone
information. This option is used only if the zone is defined as type
sl ave.
Specifies whether to notify the secondary nameservers when a
zone is updated. It accepts the following options:
yes — The server will notify all secondary nameservers.
no — The server will not notify any secondary nameserver.
master-o nl y — The server will notify primary server for the
zone only.
expl i ci t — The server will notify only the secondary servers
that are specified in the al so -no ti fy list within a zone
statement.
186
⁠Chapt er 1 1 . DNS Servers
O p t io n
D escrip t io n
type
Specifies the zone type. It accepts the following options:
d el eg ati o n-o nl y — Enforces the delegation status of
infrastructure zones such as COM, NET, or ORG. Any answer
that is received without an explicit or implicit delegation is
treated as NXD O MAIN. This option is only applicable in TLD s
(Top-Level D omain) or root zone files used in recursive or
caching implementations.
fo rward — Forwards all requests for information about this
zone to other nameservers.
hi nt — A special type of zone used to point to the root
nameservers which resolve queries when a zone is not
otherwise known. No configuration beyond the default is
necessary with a hi nt zone.
master — D esignates the nameserver as authoritative for this
zone. A zone should be set as the master if the zone's
configuration files reside on the system.
sl ave — D esignates the nameserver as a slave server for this
zone. Master server is specified in masters directive.
Most changes to the /etc/named . co nf file of a primary or secondary nameserver involve
adding, modifying, or deleting zo ne statements, and only a small subset of zo ne statement
options is usually needed for a nameserver to work efficiently.
In Example 11.5, “ A Z one Statement for a Primary nameserver” , the zone is identified as
exampl e. co m, the type is set to master, and the named service is instructed to read the
/var/named /exampl e. co m. zo ne file. It also allows only a secondary nameserver
(19 2. 16 8. 0 . 2) to transfer the zone.
Examp le 11.5. A Z o n e St at emen t f o r a Primary n ameserver
zone "example.com" IN {
type master;
file "example.com.zone";
allow-transfer { 192.168.0.2; };
};
A secondary server's zo ne statement is slightly different. The type is set to sl ave, and the
masters directive is telling named the IP address of the master server.
In Example 11.6, “ A Z one Statement for a Secondary nameserver” , the named service is
configured to query the primary server at the 19 2. 16 8. 0 . 1 IP address for information
about the exampl e. co m zone. The received information is then saved to the
/var/named /sl aves/exampl e. co m. zo ne file. Note that you have to put all slave
zones in the /var/named /sl aves/ directory, otherwise the service will fail to transfer the
zone.
Examp le 11.6 . A Z o n e St at emen t f o r a Seco n d ary n ameserver
187
Red Hat Ent erprise Linux 7 Net working G uide
zone "example.com" {
type slave;
file "slaves/example.com.zone";
masters { 192.168.0.1; };
};
1 1 .2 .2 .3. Ot he r St at e m e nt T ype s
The following types of statements are less commonly used in /etc/named . co nf:
co ntro l s
The co ntro l s statement allows you to configure various security requirements necessary
to use the rnd c command to administer the named service.
Refer to Section 11.2.4, “ Using the rndc Utility” for more information on the rnd c utility and
its usage.
key
The key statement allows you to define a particular key by name. Keys are used to
authenticate various actions, such as secure updates or the use of the rnd c command.
Two options are used with key:
al g o ri thm algorithm-name — The type of algorithm to be used (for example,
hmac-md 5).
secret "key-value" — The encrypted key.
Refer to Section 11.2.4, “ Using the rndc Utility” for more information on the rnd c utility and
its usage.
l o g g i ng
The l o g g i ng statement allows you to use multiple types of logs, so called channels. By
using the channel option within the statement, you can construct a customized type of log
with its own file name (fi l e), size limit (si ze), version number (versi o n), and level of
importance (severi ty). Once a customized channel is defined, a categ o ry option is
used to categorize the channel and begin logging when the named service is restarted.
By default, named sends standard messages to the rsysl o g daemon, which places them
in /var/l o g /messag es. Several standard channels are built into BIND with various
severity levels, such as d efaul t_sysl o g (which handles informational logging
messages) and d efaul t_d ebug (which specifically handles debugging messages). A
default category, called d efaul t, uses the built-in channels to do normal logging without
any special configuration.
Customizing the logging process can be a very detailed process and is beyond the scope
of this chapter. For information on creating custom BIND logs, refer to the BIND 9
Administrator Reference Manual referenced in Section 11.2.8.1, “ Installed D ocumentation” .
server
The server statement allows you to specify options that affect how the named service
should respond to remote nameservers, especially with regard to notifications and zone
transfers.
The transfer-fo rmat option controls the number of resource records that are sent with
188
⁠Chapt er 1 1 . DNS Servers
each message. It can be either o ne-answer (only one resource record), or many-answers
(multiple resource records). Note that while the many-answers option is more efficient, it is
not supported by older versions of BIND .
trusted -keys
The trusted -keys statement allows you to specify assorted public keys used for secure
D NS (D NSSEC). Refer to Section 11.2.6.4, “ D NS Security Extensions (D NSSEC)” for more
information on this topic.
vi ew
The vi ew statement allows you to create special views depending upon which network the
host querying the nameserver is on. This allows some hosts to receive one answer
regarding a zone while other hosts receive totally different information. Alternatively, certain
zones may only be made available to particular trusted hosts while non-trusted hosts can
only make queries for other zones.
Multiple views can be used as long as their names are unique. The match-cl i ents option
allows you to specify the IP addresses that apply to a particular view. If the o pti o ns
statement is used within a view, it overrides the already configured global options. Finally,
most vi ew statements contain multiple zo ne statements that apply to the match-cl i ents
list.
Note that the order in which the vi ew statements are listed is important, as the first
statement that matches a particular client's IP address is used. For more information on
this topic, refer to Section 11.2.6.1, “ Multiple Views” .
1 1 .2 .2 .4 . Co m m e nt T ags
Additionally to statements, the /etc/named . co nf file can also contain comments. Comments are
ignored by the named service, but can prove useful when providing additional information to a user.
The following are valid comment tags:
//
Any text after the // characters to the end of the line is considered a comment. For example:
notify yes;
// notify all secondary nameservers
#
Any text after the # character to the end of the line is considered a comment. For example:
notify yes;
# notify all secondary nameservers
/* an d */
Any block of text enclosed in /* and */ is considered a comment. For example:
notify yes;
/* notify all secondary nameservers */
11.2.3. Edit ing Zone Files
As outlined in Section 11.1.1, “ Nameserver Z ones” , zone files contain information about a
namespace. They are stored in the named working directory located in /var/named / by default.
189
Red Hat Ent erprise Linux 7 Net working G uide
Each zone file is named according to the fi l e option in the zo ne statement, usually in a way that
relates to the domain in and identifies the file as containing zone data, such as
exampl e. co m. zo ne.
T ab le 11.5. T h e n amed Service Z o n e Files
Pat h
D escrip t io n
/var/named /
The working directory for the named service. The
nameserver is not allowed to write to this
directory.
The directory for secondary zones. This
directory is writable by the named service.
The directory for other files, such as dynamic
D NS (D D NS) zones or managed D NSSEC keys.
This directory is writable by the named service.
The directory for various statistics and
debugging files. This directory is writable by the
named service.
/var/named /sl aves/
/var/named /d ynami c/
/var/named /d ata/
A zone file consists of directives and resource records. D irectives tell the nameserver to perform tasks
or apply special settings to the zone, resource records define the parameters of the zone and assign
identities to individual hosts. While the directives are optional, the resource records are required in
order to provide name service to a zone.
All directives and resource records should be entered on individual lines.
1 1 .2 .3.1 . Co m m o n Dire ct ive s
D irectives begin with the dollar sign character ($) followed by the name of the directive, and usually
appear at the top of the file. The following directives are commonly used in zone files:
$INC LUD E
The $INC LUD E directive allows you to include another file at the place where it appears, so
that other zone settings can be stored in a separate zone file.
Examp le 11.7. U sin g t h e $IN C LU D E D irect ive
$INCLUDE /var/named/penguin.example.com
$O R IG IN
The $O R IG IN directive allows you to append the domain name to unqualified records,
such as those with the host name only. Note that the use of this directive is not necessary if
the zone is specified in /etc/named . co nf, since the zone name is used by default.
In Example 11.8, “ Using the $ORIGIN D irective” , any names used in resource records that
do not end in a trailing period (the . character) are appended with exampl e. co m.
Examp le 11.8. U sin g t h e $O R IG IN D irect ive
$ORIGIN example.com.
190
⁠Chapt er 1 1 . DNS Servers
$T T L
The $T T L directive allows you to set the default Time to Live (TTL) value for the zone, that is,
how long is a zone record valid. Each resource record can contain its own TTL value,
which overrides this directive.
Increasing this value allows remote nameservers to cache the zone information for a longer
period of time, reducing the number of queries for the zone and lengthening the amount of
time required to propagate resource record changes.
Examp le 11.9 . U sin g t h e $T T L D irect ive
$TTL 1D
1 1 .2 .3.2 . Co m m o n Re so urce Re co rds
The following resource records are commonly used in zone files:
A
The Address record specifies an IP address to be assigned to a name. It takes the following
form:
hostname IN A IP-address
If the hostname value is omitted, the record will point to the last specified hostname.
In Example 11.10, “ Using the A Resource Record” , the requests for
server1. exampl e. co m are pointed to 10 . 0 . 1. 3 or 10 . 0 . 1. 5.
Examp le 11.10. U sin g t h e A R eso u rce R eco rd
server1
IN
IN
A
A
10.0.1.3
10.0.1.5
C NAME
The Canonical Name record maps one name to another. Because of this, this type of record
is sometimes referred to as an alias record. It takes the following form:
alias-name IN CNAME real-name
C NAME records are most commonly used to point to services that use a common naming
scheme, such as www for Web servers. However, there are multiple restrictions for their
usage:
CNAME records should not point to other CNAME records. This is mainly to avoid
possible infinite loops.
CNAME records should not contain other resource record types (such as A, NS, MX,
etc.). The only exception are D NSSEC related records (RRSIG, NSEC, etc.) when the
zone is signed.
191
Red Hat Ent erprise Linux 7 Net working G uide
Other resource records that point to the fully qualified domain name (FQD N) of a host
(NS, MX, PTR) should not point to a CNAME record.
In Example 11.11, “ Using the CNAME Resource Record” , the A record binds a host name to
an IP address, while the C NAME record points the commonly used www host name to it.
Examp le 11.11. U sin g t h e C N AME R eso u rce R eco rd
server1
www
IN
IN
A
CNAME
10.0.1.5
server1
MX
The Mail Exchange record specifies where the mail sent to a particular namespace controlled
by this zone should go. It takes the following form:
IN MX preference-value email-server-name
The email-server-name is a fully qualified domain name (FQD N). The preference-value allows
numerical ranking of the email servers for a namespace, giving preference to some email
systems over others. The MX resource record with the lowest preference-value is preferred
over the others. However, multiple email servers can possess the same value to distribute
email traffic evenly among them.
In Example 11.12, “ Using the MX Resource Record” , the first mai l . exampl e. co m email
server is preferred to the mai l 2. exampl e. co m email server when receiving email destined
for the exampl e. co m domain.
Examp le 11.12. U sin g t h e MX R eso u rce R eco rd
example.com.
IN
IN
MX
MX
10
20
mail.example.com.
mail2.example.com.
NS
The Nameserver record announces authoritative nameservers for a particular zone. It takes
the following form:
IN NS nameserver-name
The nameserver-name should be a fully qualified domain name (FQD N). Note that when two
nameservers are listed as authoritative for the domain, it is not important whether these
nameservers are secondary nameservers, or if one of them is a primary server. They are
both still considered authoritative.
Examp le 11.13. U sin g t h e N S R eso u rce R eco rd
IN
IN
NS
NS
dns1.example.com.
dns2.example.com.
PTR
The Pointer record points to another part of the namespace. It takes the following form:
192
⁠Chapt er 1 1 . DNS Servers
The Pointer record points to another part of the namespace. It takes the following form:
last-IP-digit IN PTR FQDN-of-system
The last-IP-digit directive is the last number in an IP address, and the FQDN-of-system is a
fully qualified domain name (FQD N).
P T R records are primarily used for reverse name resolution, as they point IP addresses
back to a particular name. Refer to Section 11.2.3.4.2, “ A Reverse Name Resolution Z one
File” for examples of P T R records in use.
SO A
The Start of Authority record announces important authoritative information about a
namespace to the nameserver. Located after the directives, it is the first resource record in a
zone file. It takes the following form:
@
IN
SOA primary-name-server hostmaster-email (
serial-number
time-to-refresh
time-to-retry
time-to-expire
minimum-TTL )
The directives are as follows:
The @ symbol places the $O R IG IN directive (or the zone's name if the $O R IG IN
directive is not set) as the namespace being defined by this SO A resource record.
The primary-name-server directive is the host name of the primary nameserver that is
authoritative for this domain.
The hostmaster-email directive is the email of the person to contact about the namespace.
The serial-number directive is a numerical value incremented every time the zone file is
altered to indicate it is time for the named service to reload the zone.
The time-to-refresh directive is the numerical value secondary nameservers use to
determine how long to wait before asking the primary nameserver if any changes have
been made to the zone.
The time-to-retry directive is a numerical value used by secondary nameservers to
determine the length of time to wait before issuing a refresh request in the event that the
primary nameserver is not answering. If the primary server has not replied to a refresh
request before the amount of time specified in the time-to-expire directive elapses, the
secondary servers stop responding as an authority for requests concerning that
namespace.
In BIND 4 and 8, the minimum-TTL directive is the amount of time other nameservers
cache the zone's information. In BIND 9, it defines how long negative answers are
cached for. Caching of negative answers can be set to a maximum of 3 hours (3H).
When configuring BIND , all times are specified in seconds. However, it is possible to use
abbreviations when specifying units of time other than seconds, such as minutes (M), hours
(H), days (D ), and weeks (W). Table 11.6, “ Seconds compared to other time units” shows an
amount of time in seconds and the equivalent time in another format.
T ab le 11.6 . Seco n d s co mp ared t o o t h er t ime u n it s
193
Red Hat Ent erprise Linux 7 Net working G uide
Seco n d s
O t h er T ime U n it s
60
1800
3600
10800
21600
43200
86400
259200
604800
31536000
1M
30 M
1H
3H
6H
12H
1D
3D
1W
36 5D
Examp le 11.14 . U sin g t h e SO A R eso u rce R eco rd
@
IN
SOA dns1.example.com. hostmaster.example.com. (
2001062501 ; serial
21600
; refresh after 6 hours
3600
; retry after 1 hour
604800
; expire after 1 week
86400 )
; minimum TTL of 1 day
1 1 .2 .3.3. Co m m e nt T ags
Additionally to resource records and directives, a zone file can also contain comments. Comments
are ignored by the named service, but can prove useful when providing additional information to the
user. Any text after the semicolon character to the end of the line is considered a comment. For
example:
604800
; expire after 1 week
1 1 .2 .3.4 . Exam ple Usage
The following examples show the basic usage of zone files.
11.2.3.4 .1. A Simp le Z o n e File
Example 11.15, “ A simple zone file” demonstrates the use of standard directives and SO A values.
Examp le 11.15. A simp le z o n e f ile
$ORIGIN example.com.
$TTL 86400
@
IN SOA dns1.example.com. hostmaster.example.com. (
2001062501 ; serial
21600
; refresh after 6 hours
3600
; retry after 1 hour
604800
; expire after 1 week
86400 )
; minimum TTL of 1 day
;
;
194
⁠Chapt er 1 1 . DNS Servers
dns1
dns2
;
;
@
mail
mail2
IN
IN
IN
IN
IN
IN
NS
NS
A
AAAA
A
AAAA
dns1.example.com.
dns2.example.com.
10.0.1.1
aaaa:bbbb::1
10.0.1.2
aaaa:bbbb::2
IN
IN
IN
IN
IN
IN
MX
MX
A
AAAA
A
AAAA
10 mail.example.com.
20 mail2.example.com.
10.0.1.5
aaaa:bbbb::5
10.0.1.6
aaaa:bbbb::6
;
;
; This sample zone file illustrates sharing the same IP addresses
; for multiple services:
;
services IN A
10.0.1.10
IN AAAA
aaaa:bbbb::10
IN A
10.0.1.11
IN AAAA
aaaa:bbbb::11
ftp
www
;
;
IN
IN
CNAME
CNAME
services.example.com.
services.example.com.
In this example, the authoritative nameservers are set as d ns1. exampl e. co m and
d ns2. exampl e. co m, and are tied to the 10 . 0 . 1. 1 and 10 . 0 . 1. 2 IP addresses respectively
using the A record.
The email servers configured with the MX records point to mai l and mai l 2 via A records. Since
these names do not end in a trailing period, the $O R IG IN domain is placed after them, expanding
them to mai l . exampl e. co m and mai l 2. exampl e. co m.
Services available at the standard names, such as www. exampl e. co m (WWW), are pointed at the
appropriate servers using the C NAME record.
This zone file would be called into service with a zo ne statement in the /etc/named . co nf similar to
the following:
zone "example.com" IN {
type master;
file "example.com.zone";
allow-update { none; };
};
11.2.3.4 .2. A R everse N ame R eso lu t io n Z o n e File
A reverse name resolution zone file is used to translate an IP address in a particular namespace into
a fully qualified domain name (FQD N). It looks very similar to a standard zone file, except that the
P T R resource records are used to link the IP addresses to a fully qualified domain name as shown
in Example 11.16, “ A reverse name resolution zone file” .
195
Red Hat Ent erprise Linux 7 Net working G uide
Examp le 11.16 . A reverse n ame reso lu t io n z o n e f ile
$ORIGIN 1.0.10.in-addr.arpa.
$TTL 86400
@ IN SOA dns1.example.com. hostmaster.example.com. (
2001062501 ; serial
21600
; refresh after 6 hours
3600
; retry after 1 hour
604800
; expire after 1 week
86400 )
; minimum TTL of 1 day
;
@ IN NS
dns1.example.com.
;
1 IN PTR dns1.example.com.
2 IN PTR dns2.example.com.
;
5 IN PTR server1.example.com.
6 IN PTR server2.example.com.
;
3 IN PTR ftp.example.com.
4 IN PTR ftp.example.com.
In this example, IP addresses 10 . 0 . 1. 1 through 10 . 0 . 1. 6 are pointed to the corresponding
fully qualified domain name.
This zone file would be called into service with a zo ne statement in the /etc/named . co nf file
similar to the following:
zone "1.0.10.in-addr.arpa" IN {
type master;
file "example.com.rr.zone";
allow-update { none; };
};
There is very little difference between this example and a standard zo ne statement, except for the
zone name. Note that a reverse name resolution zone requires the first three blocks of the IP address
reversed followed by . i n-ad d r. arpa. This allows the single block of IP numbers used in the
reverse name resolution zone file to be associated with the zone.
11.2.4 . Using t he rndc Ut ilit y
The rnd c utility is a command-line tool that allows you to administer the named service, both locally
and from a remote machine. Its usage is as follows:
rnd c [option...] command [command-option]
1 1 .2 .4 .1 . Co nfiguring t he Ut ilit y
To prevent unauthorized access to the service, named must be configured to listen on the selected
port (9 53 by default), and an identical key must be used by both the service and the rnd c utility.
T ab le 11.7. R elevan t f iles
196
⁠Chapt er 1 1 . DNS Servers
Pat h
D escrip t io n
/etc/named . co nf
/etc/rnd c. co nf
/etc/rnd c. key
The default configuration file for the named service.
The default configuration file for the rnd c utility.
The default key location.
The rnd c configuration is located in /etc/rnd c. co nf. If the file does not exist, the utility will use
the key located in /etc/rnd c. key, which was generated automatically during the installation
process using the rnd c-co nfg en -a command.
The named service is configured using the co ntro l s statement in the /etc/named . co nf
configuration file as described in Section 11.2.2.3, “ Other Statement Types” . Unless this statement is
present, only the connections from the loopback address (127. 0 . 0 . 1) will be allowed, and the key
located in /etc/rnd c. key will be used.
For more information on this topic, refer to manual pages and the BIND 9 Administrator Reference
Manual listed in Section 11.2.8, “ Additional Resources” .
Important
To prevent unprivileged users from sending control commands to the service, make sure only
ro o t is allowed to read the /etc/rnd c. key file:
~]# chmo d o -rwx /etc/rnd c. key
1 1 .2 .4 .2 . Che cking t he Se rvice St at us
To check the current status of the named service, use the following command:
~]# rnd c status
version: 9.7.0-P2-RedHat-9.7.0-5.P2.el6
CPUs found: 1
worker threads: 1
number of zones: 16
debug level: 0
xfers running: 0
xfers deferred: 0
soa queries in progress: 0
query logging is OFF
recursive clients: 0/0/1000
tcp clients: 0/100
server is up and running
1 1 .2 .4 .3. Re lo ading t he Co nfigurat io n and Zo ne s
To reload both the configuration file and zones, type the following at a shell prompt:
~]# rnd c rel o ad
server reload successful
This will reload the zones while keeping all previously cached responses, so that you can make
changes to the zone files without losing all stored name resolutions.
197
Red Hat Ent erprise Linux 7 Net working G uide
To reload a single zone, specify its name after the rel o ad command, for example:
~]# rnd c rel o ad l o cal ho st
zone reload up-to-date
Finally, to reload the configuration file and newly added zones only, type:
~]# rnd c reco nfi g
Note
If you intend to manually modify a zone that uses D ynamic D NS (D D NS), make sure you run
the freeze command first:
~]# rnd c freeze l o cal ho st
Once you are finished, run the thaw command to allow the D D NS again and reload the zone:
~]# rnd c thaw l o cal ho st
The zone reload and thaw was successful.
1 1 .2 .4 .4 . Updat ing Zo ne Ke ys
To update the D NSSEC keys and sign the zone, use the si g n command. For example:
~]# rnd c si g n l o cal ho st
Note that to sign a zone with the above command, the auto -d nssec option has to be set to
mai ntai n in the zone statement. For example:
zone "localhost" IN {
type master;
file "named.localhost";
allow-update { none; };
auto-dnssec maintain;
};
1 1 .2 .4 .5 . Enabling t he DNSSEC Validat io n
To enable the D NSSEC validation, issue the following command as ro o t:
~]# rnd c val i d ati o n o n
Similarly, to disable this option, type:
~]# rnd c val i d ati o n o ff
Refer to the o pti o ns statement described in Section 11.2.2.2, “ Common Statement Types” for
information on how to configure this option in /etc/named . co nf.
198
⁠Chapt er 1 1 . DNS Servers
The Red Hat Enterprise Linux 7 Security Guide has a comprehensive section on D NSSEC.
1 1 .2 .4 .6 . Enabling t he Que ry Lo gging
To enable (or disable in case it is currently enabled) the query logging, issue the following command
as ro o t:
~]# rnd c q ueryl o g
To check the current setting, use the status command as described in Section 11.2.4.2, “ Checking
the Service Status” .
11.2.5. Using t he dig Ut ilit y
The d i g utility is a command-line tool that allows you to perform D NS lookups and debug a
nameserver configuration. Its typical usage is as follows:
d i g [@ server] [option...] name type
Refer to Section 11.2.3.2, “ Common Resource Records” for a list of common values to use for type.
1 1 .2 .5 .1 . Lo o king Up a Nam e se rve r
To look up a nameserver for a particular domain, use the command in the following form:
d i g name NS
In Example 11.17, “ A sample nameserver lookup” , the d i g utility is used to display nameservers for
exampl e. co m.
Examp le 11.17. A samp le n ameserver lo o ku p
~]$ d i g exampl e. co m NS
; <<>> DiG 9.7.1-P2-RedHat-9.7.1-2.P2.fc13 <<>> example.com NS
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 57883
;; flags: qr rd ra; QUERY: 1, ANSWER: 2, AUTHORITY: 0, ADDITIONAL: 0
;; QUESTION SECTION:
;example.com.
;; ANSWER SECTION:
example.com.
example.com.
;;
;;
;;
;;
99374
99374
IN
NS
IN
IN
NS
NS
a.iana-servers.net.
b.iana-servers.net.
Query time: 1 msec
SERVER: 10.34.255.7#53(10.34.255.7)
WHEN: Wed Aug 18 18:04:06 2010
MSG SIZE rcvd: 77
199
Red Hat Ent erprise Linux 7 Net working G uide
1 1 .2 .5 .2 . Lo o king Up an IP Addre ss
To look up an IP address assigned to a particular domain, use the command in the following form:
d i g name A
In Example 11.18, “ A sample IP address lookup” , the d i g utility is used to display the IP address of
exampl e. co m.
Examp le 11.18. A samp le IP ad d ress lo o ku p
~]$ d i g exampl e. co m A
; <<>> DiG 9.7.1-P2-RedHat-9.7.1-2.P2.fc13 <<>> example.com A
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 4849
;; flags: qr rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 2, ADDITIONAL: 0
;; QUESTION SECTION:
;example.com.
IN
A
;; ANSWER SECTION:
example.com.
155606
IN
A
192.0.32.10
;; AUTHORITY SECTION:
example.com.
example.com.
99175
99175
IN
IN
NS
NS
a.iana-servers.net.
b.iana-servers.net.
;;
;;
;;
;;
Query time: 1 msec
SERVER: 10.34.255.7#53(10.34.255.7)
WHEN: Wed Aug 18 18:07:25 2010
MSG SIZE rcvd: 93
1 1 .2 .5 .3. Lo o king Up a Ho st Nam e
To look up a host name for a particular IP address, use the command in the following form:
d i g -x address
In Example 11.19, “ A Sample Host Name Lookup” , the d i g utility is used to display the host name
assigned to 19 2. 0 . 32. 10 .
Examp le 11.19 . A Samp le H o st N ame Lo o ku p
~]$ d i g -x 19 2. 0 . 32. 10
; <<>> DiG 9.7.1-P2-RedHat-9.7.1-2.P2.fc13 <<>> -x 192.0.32.10
;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 29683
;; flags: qr rd ra; QUERY: 1, ANSWER: 1, AUTHORITY: 5, ADDITIONAL: 6
200
⁠Chapt er 1 1 . DNS Servers
;; QUESTION SECTION:
;10.32.0.192.in-addr.arpa.
IN
PTR
;; ANSWER SECTION:
10.32.0.192.in-addr.arpa. 21600 IN
PTR
www.example.com.
;; AUTHORITY SECTION:
32.0.192.in-addr.arpa.
32.0.192.in-addr.arpa.
32.0.192.in-addr.arpa.
32.0.192.in-addr.arpa.
32.0.192.in-addr.arpa.
21600
21600
21600
21600
21600
IN
IN
IN
IN
IN
NS
NS
NS
NS
NS
b.iana-servers.org.
c.iana-servers.net.
d.iana-servers.net.
ns.icann.org.
a.iana-servers.net.
;; ADDITIONAL SECTION:
a.iana-servers.net.
b.iana-servers.org.
b.iana-servers.org.
c.iana-servers.net.
c.iana-servers.net.
ns.icann.org.
13688
5844
5844
12173
12173
12884
IN
IN
IN
IN
IN
IN
A
A
AAAA
A
AAAA
A
192.0.34.43
193.0.0.236
2001:610:240:2::c100:ec
139.91.1.10
2001:648:2c30::1:10
192.0.34.126
;;
;;
;;
;;
Query time: 156 msec
SERVER: 10.34.255.7#53(10.34.255.7)
WHEN: Wed Aug 18 18:25:15 2010
MSG SIZE rcvd: 310
11.2.6. Advanced Feat ures of BIND
Most BIND implementations only use the named service to provide name resolution services or to act
as an authority for a particular domain. However, BIND version 9 has a number of advanced features
that allow for a more secure and efficient D NS service.
Important
Before attempting to use advanced features like D NSSEC, TSIG, or IXFR (Incremental Z one
Transfer), make sure that the particular feature is supported by all nameservers in the network
environment, especially when you use older versions of BIND or non-BIND servers.
All of the features mentioned are discussed in greater detail in the BIND 9 Administrator Reference
Manual referenced in Section 11.2.8.1, “ Installed D ocumentation” .
1 1 .2 .6 .1 . Mult iple Vie ws
Optionally, different information can be presented to a client depending on the network a request
originates from. This is primarily used to deny sensitive D NS entries from clients outside of the local
network, while allowing queries from clients inside the local network.
To configure multiple views, add the vi ew statement to the /etc/named . co nf configuration file.
Use the match-cl i ents option to match IP addresses or entire networks and give them special
options and zone data.
201
Red Hat Ent erprise Linux 7 Net working G uide
1 1 .2 .6 .2 . Incre m e nt al Zo ne T ransfe rs (IXFR)
Incremental Zone Transfers (IXFR) allow a secondary nameserver to only download the updated
portions of a zone modified on a primary nameserver. Compared to the standard transfer process,
this makes the notification and update process much more efficient.
Note that IXFR is only available when using dynamic updating to make changes to master zone
records. If manually editing zone files to make changes, Automatic Zone Transfer (AXFR) is used.
1 1 .2 .6 .3. T ransact io n SIGnat ure s (T SIG)
Transaction SIGnatures (TSIG) ensure that a shared secret key exists on both primary and secondary
nameservers before allowing a transfer. This strengthens the standard IP address-based method of
transfer authorization, since attackers would not only need to have access to the IP address to
transfer the zone, but they would also need to know the secret key.
Since version 9, BIND also supports TKEY, which is another shared secret key method of authorizing
zone transfers.
Important
When communicating over an insecure network, do not rely on IP address-based
authentication only.
1 1 .2 .6 .4 . DNS Se curit y Ext e nsio ns (DNSSEC)
Domain Name System Security Extensions (DNSSEC) provide origin authentication of D NS data,
authenticated denial of existence, and data integrity. When a particular domain is marked as secure,
the SER VFAIL response is returned for each resource record that fails the validation.
Note that to debug a D NSSEC-signed domain or a D NSSEC-aware resolver, you can use the d i g
utility as described in Section 11.2.5, “ Using the dig Utility” . Useful options are + d nssec (requests
D NSSEC-related resource records by setting the D NSSEC OK bit), + cd (tells recursive nameserver
not to validate the response), and + bufsi ze= 512 (changes the packet size to 512B to get through
some firewalls).
1 1 .2 .6 .5 . Int e rne t Pro t o co l ve rsio n 6 (IPv6 )
Internet Protocol version 6 (IPv6) is supported through the use of AAAA resource records, and the
l i sten-o n-v6 directive as described in Table 11.3, “ Commonly Used Configuration Options” .
11.2.7. Common Mist akes t o Avoid
The following is a list of recommendations on how to avoid common mistakes users make when
configuring a nameserver:
U se semico lo n s an d cu rly b racket s co rrect ly
An omitted semicolon or unmatched curly bracket in the /etc/named . co nf file can
prevent the named service from starting.
U se p erio d ( t h e . ch aract er) co rrect ly
In zone files, a period at the end of a domain name denotes a fully qualified domain name.
If omitted, the named service will append the name of the zone or the value of $O R IG IN to
202
⁠Chapt er 1 1 . DNS Servers
If omitted, the named service will append the name of the zone or the value of $O R IG IN to
complete it.
In cremen t t h e serial n u mb er wh en ed it in g a z o n e f ile
If the serial number is not incremented, the primary nameserver will have the correct, new
information, but the secondary nameservers will never be notified of the change, and will
not attempt to refresh their data of that zone.
C o n f ig u re t h e f irewall
If a firewall is blocking connections from the named service to other nameservers, the
recommended practice is to change the firewall settings.
Warning
Using a fixed UD P source port for D NS queries is a potential security vulnerability
that could allow an attacker to conduct cache-poisoning attacks more easily. To
prevent this, by default D NS sends from a random ephemeral port. Configure your
firewall to allow outgoing queries from a random UD P source port. The range 10 24
to 6 5535 is used by default.
11.2.8. Addit ional Resources
The following sources of information provide additional resources regarding BIND .
1 1 .2 .8 .1 . Inst alle d Do cum e nt at io n
BIND features a full range of installed documentation covering many different topics, each placed in
its own subject directory. For each item below, replace version with the version of the bind package
installed on the system:
/usr/share/d o c/bi nd -version/
The main directory containing the most recent documentation. The directory contains the
BIND 9 Administrator Reference Manual in HTML and PD F formats, which details BIND
resource requirements, how to configure different types of nameservers, how to perform load
balancing, and other advanced topics.
/usr/share/d o c/bi nd -version/sampl e/etc/
The directory containing examples of named configuration files.
rnd c(8)
The manual page for the rnd c name server control utility, containing documentation on its
usage.
named (8)
The manual page for the Internet domain name server named , containing documentation
on assorted arguments that can be used to control the BIND nameserver daemon.
l wresd (8)
The manual page for the lightweight resolver daemon l wresd , containing documentation
on the daemon and its usage.
203
Red Hat Ent erprise Linux 7 Net working G uide
named . co nf(5)
The manual page with a comprehensive list of options available within the named
configuration file.
rnd c. co nf(5)
The manual page with a comprehensive list of options available within the rnd c
configuration file.
1 1 .2 .8 .2 . Online Re so urce s
h t t p s://access.red h at .co m/sit e/art icles/770133
A Red Hat Knowledgebase article about running BIND in a chro o t environment, including
the differences compared to Red Hat Enterprise Linux 6.
h t t p s://access.red h at .co m/d o cu men t at io n /en U S/R ed _H at _En t erp rise_Lin u x/7/h t ml/Secu rit y_G u id e/
The Red Hat Enterprise Linux 7 Security Guide has a comprehensive section on D NSSEC.
h t t p ://www.ican n .o rg /en /h elp /n ame- co llisio n /f aq s
The ICANN FAQ on domain name collision.
204
Revision Hist ory
Revision History
R evisio n 0.9 - 16
Version for 7.1 GA release
T u e 17 Feb 2015
C h rist ian H u f f man
R evisio n 0.9 - 14
Fri D ec 05 2014
C h rist ian H u f f man
Updated the nmtui and NetworkManager GUI sections in Bridging, Bonding, and Teaming.
R evisio n 0.9 - 12
Wed N o v 05 2014
Improved IP Networking, 802.1Q VLAN tagging, and Teaming.
St ep h en Wad eley
R evisio n 0.9 - 11
T u es O ct 21 2014
Improved Bonding, Bridging, and Teaming.
St ep h en Wad eley
R evisio n 0.9 - 9
T u e Sep 2 2014
Improved Bonding and Consistent Network Device Naming.
St ep h en Wad eley
R evisio n 0.9 - 8
T u e Ju ly 8 2014
St ep h en Wad eley
Red Hat Enterprise Linux 7.0 GA release of the Networking Guide.
R evisio n 0- 0
Wed D ec 12 2012
St ep h en Wad eley
Initialization of the Red Hat Enterprise Linux 7 Networking Guide.
A.1. Acknowledgment s
Certain portions of this text first appeared in the Red Hat Enterprise Linux 6 Deployment Guide,
copyright © 2014 Red Hat, Inc., available at https://access.redhat.com/documentation/enUS/Red_Hat_Enterprise_Linux/6/html/D eployment_Guide/index.html.
Index
Symbols
/et c/n amed .co n f ( see B IN D )
A
au t h o rit at ive n ameserver ( see B IN D )
B
B erkeley In t ern et N ame D o main ( see B IN D )
B IN D
- additional resources, Online Resources
- installed documentation, Installed D ocumentation
- common mistakes, Common Mistakes to Avoid
- configuration
- acl statement, Common Statement Types
- comment tags, Comment Tags
- controls statement, Other Statement Types
- include statement, Common Statement Types
205
Red Hat Ent erprise Linux 7 Net working G uide
-
key statement, Other Statement Types
logging statement, Other Statement Types
options statement, Common Statement Types
server statement, Other Statement Types
trusted-keys statement, Other Statement Types
view statement, Other Statement Types
zone statement, Common Statement Types
- directories
- /etc/named/, Configuring the named Service
- /var/named/, Editing Z one Files
- /var/named/data/, Editing Z one Files
- /var/named/dynamic/, Editing Z one Files
- /var/named/slaves/, Editing Z one Files
- features
-
Automatic Z one Transfer (AXFR), Incremental Z one Transfers (IXFR)
D NS Security Extensions (D NSSEC), D NS Security Extensions (D NSSEC)
Incremental Z one Transfer (IXFR), Incremental Z one Transfers (IXFR)
Internet Protocol version 6 (IPv6), Internet Protocol version 6 (IPv6)
multiple views, Multiple Views
Transaction SIGnature (TSIG), Transaction SIGnatures (TSIG)
- files
- /etc/named.conf, Configuring the named Service, Configuring the Utility
- /etc/rndc.conf, Configuring the Utility
- /etc/rndc.key, Configuring the Utility
- resource record, Nameserver Z ones
- types
- authoritative nameserver, Nameserver Types
- primary (master) nameserver, Nameserver Z ones, Nameserver Types
- recursive nameserver, Nameserver Types
- secondary (slave) nameserver, Nameserver Z ones, Nameserver Types
- utilities
- dig, BIND as a Nameserver, Using the dig Utility, D NS Security Extensions
(D NSSEC)
- named, BIND as a Nameserver, Configuring the named Service
- rndc, BIND as a Nameserver, Using the rndc Utility
- zones
-
$INCLUD E directive, Common D irectives
$ORIGIN directive, Common D irectives
$TTL directive, Common D irectives
A (Address) resource record, Common Resource Records
CNAME (Canonical Name) resource record, Common Resource Records
comment tags, Comment Tags
description, Nameserver Z ones
example usage, A Simple Z one File, A Reverse Name Resolution Z one File
MX (Mail Exchange) resource record, Common Resource Records
NS (Nameserver) resource record, Common Resource Records
PTR (Pointer) resource record, Common Resource Records
SOA (Start of Authority) resource record, Common Resource Records
b o n d in g ( see ch an n el b o n d in g )
C
206
Revision Hist ory
ch an n el b o n d in g
- configuration, Using Channel Bonding
- description, Using Channel Bonding
- parameters to bonded interfaces, Bonding Module D irectives
ch an n el b o n d in g in t erf ace ( see kern el mo d u le)
D
d ef au lt g at eway, St at ic R o u t es an d t h e D ef au lt G at eway
D H C P, D H C P Servers
- additional resources, Additional Resources
- command-line options, Starting and Stopping the Server
- dhcpd.conf, Configuration File
- dhcpd.leases, Starting and Stopping the Server
- dhcpd6.conf, D HCP for IPv6 (D HCPv6)
- D HCPv6, D HCP for IPv6 (D HCPv6)
- dhcrelay, D HCP Relay Agent
- global parameters, Configuration File
- group, Configuration File
- options, Configuration File
- reasons for using, Why Use D HCP?
- Relay Agent, D HCP Relay Agent
- server configuration, Configuring a D HCP Server
- shared-network, Configuration File
- starting the server, Starting and Stopping the Server
- stopping the server, Starting and Stopping the Server
- subnet, Configuration File
d h cp d .co n f , C o n f ig u rat io n File
d h cp d .leases, St art in g an d St o p p in g t h e Server
d h crelay, D H C P R elay Ag en t
d ig ( see B IN D )
DNS
- definition, D NS Servers
- (see also BIND )
D yn amic H o st C o n f ig u rat io n Pro t o co l ( see D H C P)
K
kern el mo d u le
- bonding module, Using Channel Bonding
- description, Using Channel Bonding
- parameters to bonded interfaces, Bonding Module D irectives
- module parameters
- bonding module parameters, Bonding Module D irectives
M
Mu lt ih o med D H C P
- host configuration, Host Configuration
- server configuration, Configuring a Multihomed D HCP Server
N
207
Red Hat Ent erprise Linux 7 Net working G uide
n amed ( see B IN D )
n ameserver ( see D N S)
N IC
- binding into single channel, Using Channel Bonding
P
p rimary n ameserver ( see B IN D )
R
recu rsive n ameserver ( see B IN D )
reso u rce reco rd ( see B IN D )
rn d c ( see B IN D )
ro o t n ameserver ( see B IN D )
S
seco n d ary n ameserver ( see B IN D )
st at ic ro u t e, St at ic R o u t es an d t h e D ef au lt G at eway
RED HAT CONFID ENTIAL
208
Download PDF
Similar pages