Qlogic Fast Fabric User guide

Qlogic Fast Fabric User guide
User Guide
FastFabric 7.0
IB0054607-01 A
FastFabric 7.0 User Guide
Information furnished in this manual is believed to be accurate and reliable. However, QLogic Corporation assumes no
responsibility for its use, nor for any infringements of patents or other rights of third parties which may result from its
use. QLogic Corporation reserves the right to change product specifications at any time without notice. Applications
described in this document for any of these products are for illustrative purposes only. QLogic Corporation makes no
representation nor warranty that such applications are suitable for the specified use without further testing or
modification. QLogic Corporation assumes no responsibility for any errors that may appear in this document.
Document Revision History
Revision A, September, 2011
Changes
ii
Sections Affected
IB0054607-01 A
Table of Contents
Preface
Intended Audience . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Related Materials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Documentation Conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
License Agreements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Technical Support. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Contact Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Knowledge Database . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1
FastFabric Overview
Feature Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
FastFabric Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
How FastFabric Works . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2
xi
xi
xi
xii
xiii
xiii
xiii
xiv
1-1
1-3
1-4
FastFabric TUI Menu
FastFabric TUI Menu Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
TUI Menu Usage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
QLogic InfiniBand Software Main Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Menu Items Description. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Show Installed Software . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Reconfigure OFED IP over IB . . . . . . . . . . . . . . . . . . . . . . . . . .
Reconfigure Driver Autostart . . . . . . . . . . . . . . . . . . . . . . . . . . .
Update HCA Firmware . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Generate Supporting Information for Problem Report . . . . . . . .
Fast Fabric (Host/Chassis/Switch Setup/Admin) . . . . . . . . . . . .
Uninstall Software. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
FastFabric Main Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Menu Items Description. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Chassis Setup/Admin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Externally Managed Switch Setup/Admin. . . . . . . . . . . . . . . . . .
Host Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Host Verification/Admin . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Fabric Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
IB0054607-01 A
2-1
2-1
2-3
2-3
2-3
2-3
2-3
2-3
2-4
2-4
2-4
2-5
2-5
2-5
2-5
2-5
2-5
2-6
iii
FastFabric 7.0 User Guide
Fast Fabric IB Chassis Setup/Admin Menu . . . . . . . . . . . . . . . . . . . . . . . . .
Menu Items Description. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Edit the Configuration and Select/Edit Chassis Files . . . . . . . . .
Verify Chassis via Ethernet ping . . . . . . . . . . . . . . . . . . . . . . . . .
Update Chassis Firmware . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Set up Chassis Basic Configuration . . . . . . . . . . . . . . . . . . . . . .
Setup Password-less SSH/SCP . . . . . . . . . . . . . . . . . . . . . . . . .
Reboot Chassis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Configure Chassis Fabric Manager . . . . . . . . . . . . . . . . . . . . . .
Get Basic Chassis Configuration . . . . . . . . . . . . . . . . . . . . . . . .
Check Status of IB Ports. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Control Chassis Fabric Manager . . . . . . . . . . . . . . . . . . . . . . . .
Generate all Chassis Problem Report Information . . . . . . . . . . .
Run a command on all chassis. . . . . . . . . . . . . . . . . . . . . . . . . .
View iba_chassis_admin results files . . . . . . . . . . . . . . . . . . . . .
Fast Fabric InfiniBand Externally Managed
Switch Setup/Admin Menu. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Menu Items Description. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Edit Config and Select/Edit Switch Files. . . . . . . . . . . . . . . . . . .
Test for Switch Presence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Verify Switch Firmware . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Update Switch Firmware . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Setup Switch Basic Configuration . . . . . . . . . . . . . . . . . . . . . . .
Reboot Switch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Report Switch Firmware & Hardware Info . . . . . . . . . . . . . . . . .
Get Basic Switch configuration . . . . . . . . . . . . . . . . . . . . . . . . . .
Report Switch VPD Information . . . . . . . . . . . . . . . . . . . . . . . . .
Generate All Switch Problem Report Info . . . . . . . . . . . . . . . . . .
View iba_switch_admin result files . . . . . . . . . . . . . . . . . . . . . . .
Fast Fabric IB Host Setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
iv
2-6
2-7
2-7
2-7
2-8
2-8
2-9
2-9
2-10
2-11
2-11
2-12
2-13
2-13
2-13
2-13
2-14
2-14
2-15
2-15
2-15
2-16
2-17
2-17
2-17
2-18
2-18
2-18
2-19
IB0054607-01 A
FastFabric 7.0 User Guide
Menu Items Description. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Edit Configuration and Select/Edit Hosts Files . . . . . . . . . . . . . .
Verify Hosts via Ethernet ping. . . . . . . . . . . . . . . . . . . . . . . . . . .
Setup Password-less SSH/SCP . . . . . . . . . . . . . . . . . . . . . . . . .
Copy /etc/hosts to all hosts. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Show uname -a for all hosts . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Install/Upgrade QLogic IB Software . . . . . . . . . . . . . . . . . . . . . .
Configure IPoIB IP Address . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Build MPI Test Apps and Copy to Hosts . . . . . . . . . . . . . . . . . . .
Reboot Hosts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Refresh SSH Known Hosts. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Rebuild MPI Library and Tools . . . . . . . . . . . . . . . . . . . . . . . . . .
Run a command on all hosts . . . . . . . . . . . . . . . . . . . . . . . . . . .
Copy a file to all hosts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
View iba_host_admin result files . . . . . . . . . . . . . . . . . . . . . . . .
Fast Fabric IB Host Verification/Admin Menu. . . . . . . . . . . . . . . . . . . . . . . .
Menu Items Description. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Edit Config and Select/Edit Hosts Files . . . . . . . . . . . . . . . . . . .
Verify Hosts via Ethernet ping. . . . . . . . . . . . . . . . . . . . . . . . . . .
Summary of Fabric Components . . . . . . . . . . . . . . . . . . . . . . . .
Check Status of IB Ports. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Verify Hosts see each other . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Verify Hosts ping via IPoIB . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Refresh SSH Known Hosts. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Check MPI Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Check Overall Fabric Health. . . . . . . . . . . . . . . . . . . . . . . . . . . .
Generate all Hosts Problem Report Info. . . . . . . . . . . . . . . . . . .
Run a command on all hosts . . . . . . . . . . . . . . . . . . . . . . . . . . .
View iba_host_admin result files . . . . . . . . . . . . . . . . . . . . . . . .
Fabric Monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Menu Items Description. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Fabric Performance Monitoring . . . . . . . . . . . . . . . . . . . . . . . . .
3
2-19
2-20
2-20
2-20
2-20
2-20
2-21
2-21
2-22
2-22
2-22
2-22
2-22
2-22
2-23
2-23
2-24
2-24
2-25
2-25
2-25
2-26
2-26
2-27
2-27
2-28
2-28
2-29
2-29
2-29
2-29
2-30
iba_top Fabric Performance Monitor
Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
iba_top TUI. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Image Identification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Screen Specific Information. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
IB0054607-01 A
3-1
3-1
3-2
3-2
v
FastFabric 7.0 User Guide
Command Entry. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Common Input Commands. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Screen-Specific Input Commands . . . . . . . . . . . . . . . . . . . . . . .
Access to Live and Recent PM Historical Data . . . . . . . . . . . . . . . . . .
iba_top TUI Screens. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Summary Screen. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Fabric configuration information . . . . . . . . . . . . . . . . . . . . . . . . .
Performance and Error Statistics for Each Port Group . . . . . . . .
Screen-Specific Input Commands . . . . . . . . . . . . . . . . . . . . . . .
Additional Screens. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
PM Configuration Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Image Information Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Group Information Select Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Bandwidth Statistics Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Error Statistics Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Group Configuration Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Group Focus Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Port Statistics Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Command Line Options. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4
Real-Time Fabric Monitor
Real-time Fabric Monitor Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
RFM Screen Layout and User Interaction . . . . . . . . . . . . . . . . . . . . . . . . . .
Main Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Command-line Segment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Menu Segment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Fabric Segment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Selection Segment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Submenu Segment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
View Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Menu Section. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
View Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Context Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Fabric Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Information Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Sub-Menu Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Infrastructure View Sub-Screens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
vi
3-2
3-3
3-3
3-3
3-4
3-4
3-4
3-5
3-6
3-6
3-6
3-7
3-7
3-8
3-9
3-12
3-13
3-14
3-15
4-1
4-2
4-4
4-5
4-5
4-5
4-6
4-6
4-7
4-8
4-9
4-9
4-10
4-10
4-14
4-15
IB0054607-01 A
FastFabric 7.0 User Guide
Switch Node Screens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Switch Node Selection Screen (500+ Switch Nodes). . . . . . . . .
Switch Node Selection Screen (26-500 Switch Nodes) . . . . . . .
Switch Node Selection Screen (1-25 Switch Nodes) . . . . . . . . .
Switch Node Information Selection Screen . . . . . . . . . . . . . . . .
Switch Node Device Information Screen . . . . . . . . . . . . . . . . . .
Switch Node Port Selection Screen . . . . . . . . . . . . . . . . . . . . . .
Switch Node Port Information Selection Screen. . . . . . . . . . . . .
Switch Node General Port Information Screen. . . . . . . . . . . . . .
Switch Node Port Statistics Selection Screen. . . . . . . . . . . . . . .
Switch Node Port Performance Screen . . . . . . . . . . . . . . . . . . .
Channel Adapter (CA) Screens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
CA Selection Screen (26+ CAs) . . . . . . . . . . . . . . . . . . . . . . . . .
CA Selection Screen (1-25 CAs) . . . . . . . . . . . . . . . . . . . . . . . .
CA Information Selection Screen . . . . . . . . . . . . . . . . . . . . . . . .
CA Device Information Screen . . . . . . . . . . . . . . . . . . . . . . . . . .
CA Port Selection Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
CA Port Information Selection Screen . . . . . . . . . . . . . . . . . . . .
CA Port General Information Screen . . . . . . . . . . . . . . . . . . . . .
CA Port Statistics Selection Screen . . . . . . . . . . . . . . . . . . . . . .
CA Port Performance Screen . . . . . . . . . . . . . . . . . . . . . . . . . . .
Fabric View Sub-Screens. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Link Screens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Link Category Selection Screen . . . . . . . . . . . . . . . . . . . . . . . . .
Link Selection Screen (24000+ Links) . . . . . . . . . . . . . . . . . . . .
Link Selection Screen (1200-24000 Links) . . . . . . . . . . . . . . . . .
Link Selection Screen (60-1200 Links) . . . . . . . . . . . . . . . . . . . .
Link Selection Screen (1-60 Links) . . . . . . . . . . . . . . . . . . . . . . .
Link Information Selection Screen . . . . . . . . . . . . . . . . . . . . . . .
Link End-Node Information Selection Screen. . . . . . . . . . . . . . .
Link End-Port Information Selection Screen. . . . . . . . . . . . . . . .
Slow Link Selection Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . .
SM Screens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
SM Selection Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
SM Detailed Information Screen. . . . . . . . . . . . . . . . . . . . . . . . .
Performance View Sub-Screens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Bandwidth Utilization Screens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Switch Group Bandwidth Utilization Selection Screen . . . . . . . .
Low Bandwidth Utilization Selection Screen . . . . . . . . . . . . . . .
IB0054607-01 A
4-15
4-15
4-17
4-18
4-19
4-20
4-21
4-22
4-23
4-24
4-25
4-26
4-26
4-27
4-28
4-29
4-30
4-31
4-32
4-33
4-34
4-35
4-35
4-35
4-36
4-37
4-38
4-39
4-40
4-41
4-43
4-44
4-45
4-45
4-46
4-47
4-47
4-47
4-48
vii
FastFabric 7.0 User Guide
Error Condition Screens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Switch Group Error Condition Selection Screen . . . . . . . . . . . .
Integrity Error Selection Screen . . . . . . . . . . . . . . . . . . . . . . . . .
Link Selection Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Admin Menu Screens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Main Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Fabric Discovery Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5
Configuration of IPoIB Name Mapping
A
FastFabric Configuration Files
FastFabric Configuration File . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Port Statistics Thresholds Configuration File . . . . . . . . . . . . . . . . . . . . . . . .
Signal Integrity Thresholds Configuration File . . . . . . . . . . . . . . . . . . . . . . .
Host List Files. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Chassis List Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Selection of slots within a chassis . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Externally Managed Switch List File. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Port List File . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Fabric Topology Input File . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
viii
4-49
4-49
4-50
4-51
4-52
4-52
4-53
A-2
A-7
A-9
A-11
A-12
A-12
A-13
A-17
A-18
IB0054607-01 A
FastFabric 7.0 User Guide
List of Figures
Figure
Page
1-1
FastFabric Architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1-3
2-2
QLogic InfiniBand Software Main Menu (Example) . . . . . . . . . . . . . . . . . . . . . . . . .
2-3
2-3
QLogic Fast Fabric InfiniBand Tools Menu (Example) . . . . . . . . . . . . . . . . . . . . . . .
2-5
2-4
Fast Fabric IB Chassis Setup/Admin Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2-6
2-5
Fast Fabric IB Switch Setup/Admin Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-14
2-6
Fast Fabric IB Host Setup Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-19
2-7
Fast Fabric IB Host Verification/Admin Menu. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-24
2-8
Fast Fabric IB Fabric Monitoring Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2-29
3-9
iba_top TUI Screen Layout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3-2
3-10 iba_top Summary Screen. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3-4
3-11 PM Configuration Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3-6
3-12 Image Information Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3-7
3-13 Group Information Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3-8
3-14 Bandwidth Statistics Screen. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3-8
3-15 Error Statistics Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-10
3-16 Group Configuration Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-12
3-17 Group Focus Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-14
3-18 Port Statistics Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3-15
4-19 RFM Main Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-3
4-20 Main Menu Segments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-4
4-21 View Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-8
4-22 Menu Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-8
4-23 View Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-9
4-24 Context Section . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-10
4-25 Infrastructure and Fabric View Information Section . . . . . . . . . . . . . . . . . . . . . . . . . 4-11
4-26 Performance View Error Conditions Information Section. . . . . . . . . . . . . . . . . . . . . 4-12
4-27 Performance View Bandwidth Utilization Information Section . . . . . . . . . . . . . . . . . 4-13
4-28 Infrastructure and Fabric View Sub-Menu Section. . . . . . . . . . . . . . . . . . . . . . . . . . 4-14
4-29 Performance View Bandwidth Utilization Sub-Menu Section . . . . . . . . . . . . . . . . . . 4-14
4-30 Switch Node Selection Screen (500+ Switch Nodes) . . . . . . . . . . . . . . . . . . . . . . . 4-16
4-31 Switch Node Selection Screen (26-500 Switch Nodes) . . . . . . . . . . . . . . . . . . . . . . 4-17
4-32 Switch Node Selection Screen (1-25 Switch Nodes) . . . . . . . . . . . . . . . . . . . . . . . . 4-18
4-33 Switch Node Information Selection Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-19
4-34 Switch Node Device Information Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-20
4-35 Switch Node Port Selection Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-21
4-36 Switch Node Port Information Selection Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-22
4-37 Switch Node General Port Information Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-23
4-38 Switch Node Port Statistics Selection Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-24
4-39 Switch Node Port Performance Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-25
4-40 CA Selection Screen (26+ CAs). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-26
4-41 CA Selection Screen (1-25 CAs) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-27
4-42 CA Information Selection Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-28
4-43 CA Device Information Screen. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4-29
IB0054607-01 A
ix
FastFabric 7.0 User Guide
4-44
4-45
4-46
4-47
4-48
4-49
4-50
4-51
4-52
4-53
4-54
4-55
4-56
4-57
4-58
4-59
4-60
4-61
4-62
4-63
4-64
4-65
4-66
CA Port Selection Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
CA Port Information Selection Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
CA Port General Information Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
CA Port Statistics Selection Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
CA Port Performance Screen. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Link Category Selection Screen. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Link Selection Screen (24000+ Links) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Link Selection Screen (1200-24000 Links) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Link Selection Screen (60-1200 Links) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Link Selection Screen (1-60 Links) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Link Information Selection Screen (60-1200 Links) . . . . . . . . . . . . . . . . . . . . . . . . .
Link End-Node Information Selection Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Link End-Port Information Selection Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Slow Link Selection Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
SM Selection Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
SM Detailed Information Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Switch Group Bandwidth Utilization Selection Screen . . . . . . . . . . . . . . . . . . . . . . .
Low Bandwidth Utilization Selection Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Switch Group Error Condition Selection Screen . . . . . . . . . . . . . . . . . . . . . . . . . . .
Integrity Error Selection Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Link Selection Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Admin Menu Main Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Fabric Discovery Screen . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4-30
4-31
4-32
4-33
4-34
4-35
4-36
4-37
4-38
4-39
4-40
4-41
4-43
4-44
4-45
4-46
4-47
4-48
4-49
4-50
4-51
4-52
4-53
List of Tables
Table
Page
1-1
FastFabric Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1-4
A-2
FastFabric Configuration Files . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
A-1
x
IB0054607-01 A
Preface
Intended Audience
This manual is intended to provide network administrators and other qualified
personnel a reference for installation, configuration and administration task
information for the FastFabric Toolset.
Related Materials

QLogic FastFabric Command Line Interface Reference Guide

QLogic Fabric Software Installation Guide
Documentation Conventions
This guide uses the following documentation conventions:

NOTE: provides additional information.

CAUTION! indicates the presence of a hazard that has the potential of
causing damage to data or equipment.

WARNING!! indicates the presence of a hazard that has the potential of
causing personal injury.

Text in blue font indicates a hyperlink (jump) to a figure, table, or section in
this guide, and links to Web sites are shown in underlined blue. For
example:


IB0054607-01 A

Table 9-2 lists problems related to the user interface and remote agent.

See “Installation Checklist” on page 3-6.

For more information, visit www.qlogic.com.
Text in bold font indicates user interface elements such as a menu items,
buttons, check boxes, or column headings. For example:

Click the Start button, point to Programs, point to Accessories, and
then click Command Prompt.

Under Notification Options, select the Warning Alarms check box.
Text in Courier font indicates a file name, directory path, or command line
text. For example:
xi




To return to the root directory from anywhere in the file structure:
Type cd /root and press ENTER.

Enter the following command: sh ./install.bin
Key names and key strokes are indicated with UPPERCASE:

Press CTRL+P.

Press the UP ARROW key.
Text in italics indicates terms, emphasis, variables, or document titles. For
example:

For a complete listing of license agreements, refer to the QLogic
Software End User License Agreement.

What are shortcut keys?

To enter the date type mm/dd/yyyy (where mm is the month, dd is the
day, and yyyy is the year).
Topic titles between quotation marks identify related topics either within this
manual or in the online help, which is also referred to as the help system
throughout this document.
License Agreements
Refer to the QLogic Software End User License Agreement for a complete listing
of all license agreements affecting this product.
xii
IB0054607-01 A
Technical Support
Customers should contact their authorized maintenance provider for technical
support of their QLogic products. QLogic-direct customers may contact QLogic
Technical Support; others will be redirected to their authorized maintenance
provider. Visit the QLogic support Web site listed in Contact Information for the
latest firmware and software updates.
For details about available service plans, or for information about renewing and
extending your service, visit the Service Program web page at
http://www.qlogic.com/services.
Training
QLogic offers training for technical professionals for all iSCSI, InfiniBand, and
Fibre Channel products. From the main QLogic web page at www.qlogic.com,
click the Support tab at the top, and then click Training and Certification on the
left. The QLogic Global Training portal offers online courses, certification exams,
and scheduling of in-person training.
Technical Certification courses include installation, maintenance and
troubleshooting QLogic products. Upon demonstrating knowledge using live
equipment, QLogic awards a certificate identifying the student as a certified
professional. You can reach the training professionals at QLogic by e-mail at
[email protected]
Contact Information
QLogic Technical Support for products under warranty is available during local
standard working hours excluding QLogic Observed Holidays. For customers with
extended service, consult your plan for available hours.For Support phone
numbers, see the Contact Support link at [email protected]
Support Headquarters
QLogic Corporation
4601 Dean Lakes Blvd.
Shakopee, MN 55379 USA
QLogic Web Site
www.qlogic.com
Technical Support Web Site
http://support.qlogic.com
Technical Support E-mail
[email protected]
Technical Training E-mail
[email protected]
IB0054607-01 A
xiii
Knowledge Database
The QLogic knowledge database is an extensive collection of QLogic product
information that you can search for specific solutions. We are constantly adding to
the collection of information in our database to provide answers to your most
urgent questions. Access the database from the QLogic Support Center:
http://support.qlogic.com.
xiv
IB0054607-01 A
1
FastFabric Overview
Feature Overview
The FastFabric Toolset is designed to both simplify and expedite common
InfiniBand (IB) cluster management tasks. FastFabric can assist in generic
management tasks as well as InfiniBand installation, upgrade, configuration and
verification tasks.
FastFabric has the following key capabilities:



Accelerates initial fabric installation

Performs chassis and switch configuration

Verifies host management network connectivity

Verifies host OS levels

Sets up host and chassis ssh keys

Performs initial InfiniBand software installation

Configures Internet Protocol over InfiniBand (IPoIB) IP addresses

Performs InfiniBand driver upgrades or the installation of additional
InfiniBand drivers
Verifies key fabric installation matrixes:

Components in fabric

Link error counters

Link widths and speeds

InfiniBand and PCI bus bandwidth

InfiniBand end-to-end latency

IPoIB connectivity

Subnet Agent (SA) visibility of all nodes

InfiniBand connectivity of all switches and nodes
Aids in diagnosis of fabric problems

IB0054607-01 A
Fabric error isolation
1-1
1–FastFabric Overview
Feature Overview





Fabric topology analysis and verification

Fabric route analysis
Aids in ongoing fabric status and configuration monitoring

Fabric Performance, Error and congestion monitoring

Automated fabric health checks and configuration baseline compare

Automated chassis health checks and configuration baseline compare

Automated Subnet Management (SM) health checks and configuration
baseline compare
Provides tools to accelerate common host administration tasks

Executes commands across many hosts

Copies files to and from many hosts

Edits host-specific files across many hosts
Provides tools to accelerate common chassis and switch administration
tasks

Manage firmware levels on switches and chassis

Execute commands across many chassis

Administer switches
Assists in the initial benchmarking and tuning of High Performance
Computing (HPC) fabrics.
FastFabric includes both a Textual User Interface (TUI) menu system and
command line tools. The TUI presents the menus in a typical order of execution
for a new fabric install to simplify fabric installation for new users. All operations
available in the TUI can also be accomplished using the command line. The
command line tools are designed to permit customer-specific scripts to invoke the
command line tools.
1-2
IB0054607-01 A
1–FastFabric Overview
FastFabric Architecture
FastFabric Architecture
Figure 1-1. FastFabric Architecture
FastFabric is typically installed on one or more InfiniBand Management Nodes.
The InfiniBand Management Node must be connected to the rest of the cluster
through InfiniBand and a management network. The management network may
be the primary InfiniBand network (IPoIB) or Ethernet. The management network
will be used for FastFabric host setup and administration tasks. It may also be
used for other aspects of server administration or operation.
Depending on cluster size and design, the InfiniBand Management node may also
be used as the master node for starting Message Passing Interface (MPI) jobs. It
may also be used to run a QLogic Host Fabric Manager (including the Subnet
Manager) and other management software. Consult the QLogic Fabric Manager
User Guide for details and what combinations are valid.
NOTE:
When InfiniBand is used as the management network, FastFabric will not be
able to install host InfiniBand software or configure IPoIB. However in this
configuration, FastFabric is able to support host InfiniBand software
upgrades, verification and all its other features.
IB0054607-01 A
1-3
1–FastFabric Overview
FastFabric Architecture
If remote access to FastFabric is desired, set up remote access to the InfiniBand
Management Node using ssh, telnet, X-Windows, VNC or any other mechanism
which will allow the remote user to access a Linux Command Line shell. Typically
FastFabric is used only by cluster administrators.
How FastFabric Works
FastFabric consists of a variety of tools to administer hosts, chassis and externally
managed switches. Depending on the tool, the method of accessing and
administering the target devices may differ.
The following methods are used by FastFabric:
Table 1-1. FastFabric Methods
Method
Examples
Inband access through InfiniBand
Fabric Performance, Error and congestion monitoring. Fabric topology reports, SA database queries, fabric error and link speed analysis, tools for
externally managed switches, etc.
Login through a management network
Host setup and installation, tools for internally
managed chassis, etc.
MPI job startup (can be inband or
through a management network)
Verify MPI performance, running sample MPI
benchmarks.
Tools which log in to other hosts will do so in a password-less manner using ssh or
telnet (configurable). Tools which log in to internally managed chassis can use ssh
or telnet (configurable). Chassis tools can prompt for a single password for all
chassis, use password-less ssh, or can be preconfigured with the password.
These approaches permit the tools to operate with minimal user interaction and
for this reason reduce the time to perform operations against many hosts or
chassis.
After initial installation, FastFabric can be configured to use IPoIB instead of the
management network.
NOTE:
IPoIB can not be used to reconfigure IPoIB or install new InfiniBand hosts.
1-4
IB0054607-01 A
2
FastFabric TUI Menu
FastFabric TUI Menu Overview
FastFabric is easiest to use from the textual user interface (TUI) menu system.
The menu system provides a way to perform all common tasks and presents
common options. Additional less common options are available directly, using the
Command Line Tools, documented in the QLogic FastFabric Command Line
Interface Reference Guide.
In the sections that follow, the menu system will be discussed. The majority of
menu items directly invoke various FastFabric command tools. As such the
section on each menu item will indicate what command tool it invokes and a
summary of the operation performed. For further details about the given command
tool, consult the relevant section in the FastFabric Command Line Interface
Reference Guide.
Some of the menu items are only applicable when Linux is being used. They will
be marked with (Linux). Similarly some of the menu items are only applicable
when QLogic OFED+, the QLogicIB packaging of OFED software, is being used
on the hosts and will be marked with (Host). All menu items which are applicable
only when QLogic InfiniBand Switches or Chassis are being used will be marked
with (Switch). All remaining menu items are generally applicable to all
environments and will be marked with (All)
.
NOTE:
Some of the Linux menu items may be applicable to other Unix-like
operating systems.These may be used to enable the use of non-InfiniBand
specific FastFabric tools (such as cmdall) against the given hosts.
TUI Menu Usage
The TUI menus are set up for ease of use. The submenus typically present
operations in the typical order they would be used during an installation. Pressing
the keys corresponding to menu items (0-9, a-d) will toggle the Skip/Perform
selection for the given item. More than one item may be selected. Once the
desired set of items have been selected, enter P. To unselect all items, enter N.
Entering X or pressing ESC will exit this menu and return to the Main Menu.
IB0054607-01 A
2-1
2–FastFabric TUI Menu
FastFabric TUI Menu Overview
If more than one item is selected, the items will be performed in the order shown
in the menu. This is the typical order desired during fabric setup. If it's desired to
perform items in a different order, select a single item and enter P to perform it by
itself. Then repeat for the next item to be performed. An opportunity will be
presented after each item is selected to abort as follows:
Hit any key to continue (or ESC to abort)...
If ESC is pressed, the sequence of operations will be aborted and return to the
previous menu. Any other key will result in the next selected menu item being
performed. This prompt is also shown after the last selected item completes
permitting an opportunity to review the results before the screen is cleared to
display the menu.
At the top of each FastFabric menu, the file listing the components to operate on
is shown. For example:
Host List: /etc/sysconfig/iba/hosts
On each FastFabric menu, item 0 will permit a different file to be selected and will
permit the editing of the file (using the editor selected by the EDITOR environment
variable). In addition it will also permit review and editing of the fastfabric.conf file.
The fastfabric.conf file guides the overall configuration of FastFabric and
describes cluster specific attributes of how FastFabric will operate. It is discussed
in greater detail in Appendix A.
During the execution of each menu selection, the actual FastFabric command line
tool being used will be shown. This can be used as an educational aid to learn the
command line tools.
2-2
IB0054607-01 A
2–FastFabric TUI Menu
QLogic InfiniBand Software Main Menu
QLogic InfiniBand Software Main Menu
The QLogic InfiniBand Software main menu is the top level menu for the QLogic
InfiniBand Software. It can be activated using the iba_config command. This
menu is not part of the FastFabric TUI. However, since it is one way of getting to
the FastFabric Main Menu it will be summarized here. Figure 2-2 is a example of
the QLogic Infiniband Software main menu.
QLogic Inc. InfiniBand VERSION Software
1) Show Installed Software
2) Reconfigure OFED IP over IB
3) Reconfigure Driver Autostart
4) Update HCA Firmware
5) Generate Supporting Information for Problem Report
6) Fast Fabric (Host/Chassis/Switch Setup/Admin)
7) Uninstall Software
X) Exit
Figure 2-2. QLogic InfiniBand Software Main Menu (Example)
Menu Items Description
Selecting items 1 through 7 will display the given submenu. Selecting X will exit
the menu system. The submenus are described below.
Show Installed Software
Menu item 1) Show Installed Software when selected displays the QLogic Inc.
IB Installed Software list and shows what is Installed and Not Installed.
Reconfigure OFED IP over IB
Menu item 2) Reconfigure OFED IP over IB when selected proceeds through
the reconfiguration of the OFED IP over IB configuration.
Reconfigure Driver Autostart
Menu item 3) Reconfigure Driver Autostart when selected proceeds through
the reconfiguring of the drivers autostart configuration.
Update HCA Firmware
Menu item 4) Update HCA Firmware when selected proceeds through the
updating the HCA Firmware.
IB0054607-01 A
2-3
2–FastFabric TUI Menu
QLogic InfiniBand Software Main Menu
Generate Supporting Information for Problem Report
Menu item 5) Generate Supporting Information for Problem Report when
selected proceeds through the process of generating a report and saving it to a
user specified file.
Fast Fabric (Host/Chassis/Switch Setup/Admin)
Menu item 6) Fast Fabric (Host/Chassis/Switch Setup/Admin) when selected
displays the Fast Fabric InfiniBand Tools menu. Refer to FastFabric Main Menu
section below.
Uninstall Software
Menu item 7) Uninstall Software when selected proceeds to the QLogic Inc. IB
Uninstall Menu.
2-4
IB0054607-01 A
2–FastFabric TUI Menu
FastFabric Main Menu
FastFabric Main Menu
The FastFabric main menu is the starting point to manage the fabric using the
TUI. Selecting 6 from the above menu or executing the fastfabric command
at a prompt, displays the Fast Fabric InfiniBand Tools menu (Figure 2-3)
QLogic Inc. Fast Fabric InfiniBand Tools
Version: VERSION
1) Chassis Setup/Admin
2) Externally Managed Switch Setup/Admin
3) Host Setup
4) Host Verification/Admin
5) Fabric Monitoring
X) Exit
Figure 2-3. QLogic Fast Fabric InfiniBand Tools Menu (Example)
Menu Items Description
Selecting items 1 through 5 will display the given submenu. Selecting X will exit
the menu system. The submenus are described below.
Chassis Setup/Admin
Menu item 1) Chassis Setup/Admin when selected displays the Fast Fabric IB
Chassis Setup/Admin Menu. Refer to “Fast Fabric IB Chassis Setup/Admin
Menu” on page 2-6 for detailed information.
Externally Managed Switch Setup/Admin
Menu item 2) Externally Managed Switch Setup/Admin when selected
displays the Fast Fabric IB Switch Setup/Admin Menu. Refer to “Fast Fabric
InfiniBand Externally Managed Switch Setup/Admin Menu” on page 2-13 for
detailed information.
Host Setup
Menu item 3) Host Setup when selected displays the Fast Fabric IB Host
Setup Menu. Refer to “Fast Fabric IB Host Setup” on page 2-19 for detailed
information.
Host Verification/Admin
Menu item 4) Host Verification/Admin when selected displays the Fast Fabric
IB Host Verification/Admin Menu. Refer to “Fast Fabric IB Host
Verification/Admin Menu” on page 2-23 for detailed information.
IB0054607-01 A
2-5
2–FastFabric TUI Menu
Fast Fabric IB Chassis Setup/Admin Menu
Fabric Monitoring
Menu item 5) Fabric Monitoring when selected displays the Fast Fabric IB
Fabric Monitoring Menu. Refer to “Fabric Monitoring” on page 2-29 for detailed
information.
Fast Fabric IB Chassis Setup/Admin Menu
This menu is focused on initial setup and administration of QLogic 12000
internally-managed InfiniBand switches. Pressing the keys corresponding to
menu items (0-9) will toggle the Skip/Perform selection for the given item. More
than one item may be selected. Once the desired set of items have been selected,
enter P. To unselect all items, enter N. Entering X or pressing ESC will exit this
menu and return to the Main Menu.
Selecting 1 from the Fast Fabric InfiniBand Tools menu (Figure 2-3), displays
the Fast Fabric IB Chassis Setup/Admin Menu (Figure 2-4)
Fast Fabric IB Chassis Setup/Admin Menu
Chassis List: /etc/sysconfig/iba/chassis
Setup:
0) Edit Config and Select/Edit Chassis Files [ Skip
]
1) Verify Chassis via Ethernet ping
[ Skip
]
2) Update Chassis Firmware
[ Skip
]
3) Setup Chassis Basic Configuration
[ Skip
]
4) Setup Password-less ssh/scp
[ Skip
]
5) Reboot Chassis
[ Skip
]
6) Configure Chassis Fabric Manager
[ Skip
]
7) Get Basic Chassis Configuration
[ Skip
]
8) Check Status of IB Ports
[ Skip
]
9) Control Chassis Fabric Manager
[ Skip
]
a) Generate all Chassis Problem Report Info
[ Skip
]
b) Run a command on all chassis
[ Skip
]
c) View iba_chassis_admin result files
[ Skip
]
P) Perform the selected actions
N) Select None
Admin:
Review:
X) Return to Previous Menu (or ESC)
Figure 2-4. Fast Fabric IB Chassis Setup/Admin Menu
2-6
IB0054607-01 A
2–FastFabric TUI Menu
Fast Fabric IB Chassis Setup/Admin Menu
Menu Items Description
Selecting items 0 through c will change the item from skip to perform. Selecting
N will unselect all items and X will exit the menu system. The items are described
below.
Edit the Configuration and Select/Edit Chassis Files
(Switch) This will permit the chassis, ports and fastfabric.conf files to be
edited. The chassis file selected and created by the menu should list the internally
managed QLogic switching chassis which are to be operated on. After editing the
files, an opportunity is given to edit them again or continue forward. The first file to
review and edit is the FastFabric configuration file,
/etc/sysconfig/fastfabric.conf:
About to: vi /etc/sysconfig/fastfabric.conf
Hit any key to continue (or ESC to abort)...
The next file to be reviewed is the FastFabric PORTS_FILE,
/etc/sysconfig/iba/ports:
About to: vi /etc/sysconfig/iba/ports
Hit any key to continue (or ESC to abort)...
The next file to be reviewed is the chassis file, It first ask the user to select the
chassis file to use and gives a default. If a different chassis file needs to be edited
the user would need to type in the file location and name. If not pressing ENTER
selects the default chassis file:
Select Chassis File to Use/Edit [/etc/sysconfig/iba/chassis]:
About to: vi /etc/sysconfig/iba/chassis
Hit any key to continue (or ESC to abort)...
After exiting the vi editor, the TUI will ask if you want to edit the chassis file again.
Answer n to continue forward and return to the Fast Fabric IB Chassis
Setup/Admin Menu.
Refer to “FastFabric Configuration File” on page A-2 for more details about the
format of the FastFabric configuration file. Refer to “Port List File” on page A-17
for more details about the format of the FastFabric ports file. Refer to “Chassis List
Files” on page A-12 for more details about the format of the chassis list file and
about the iba_gen_chassis command which can help generate the chassis file.
Verify Chassis via Ethernet ping
(Switch) This will run the pingall -C -p -F command to verify the existence of
each selected chassis listed in the chassis file using a ping over the
management network.
IB0054607-01 A
2-7
2–FastFabric TUI Menu
Fast Fabric IB Chassis Setup/Admin Menu
Update Chassis Firmware
(Switch) This will run the iba_chassis_admin update command to permit the
chassis firmware version to be verified and updated as needed.
NOTE:
Any QLogic or SilverStorm 9000 chassis must be running firmware version
4.0.0.4.3 or later to perform this function. If the chassis is not up to this level,
it will need to be manually updated using the chassis GUI. See the
SilverStorm 9000 Users Guide for more information.
NOTE:
Consult the relevant chassis firmware release notes to ensure any
prerequisites for the upgrade to the new firmware level have been met prior
to performing the upgrade using FastFabric.
Prompts will guide the user through the options:

run - Ensures given firmware is in primary image and running. As needed
push firmware to each chassis, select it for use and/or if its not the presently
running firmware, reboot the chassis

select - Ensures given firmware is in primary image. As needed push
firmware to each chassis and/or select it for use on next reboot

push - Ensures given firmware is in primary or alternate image. As needed
push firmware to each chassis but do not change selected nor running
firmware
Additional options prompted for:

selection of firmware files or directory containing .pkg files

parallel vs serial update

prompting for chassis password (default is to have password in
fastfabric.conf or to use password-less ssh)
If any chassis fails to be updated, use the View iba_chassis_admin result files
option to review the result files from the update. Refer to FastFabric Command
Line Interface Reference Guide for more details.
Set up Chassis Basic Configuration
(Switch) This will run the iba_chassis_admin configure command which will
prompt for chassis configuration settings and then configure all the selected
chassis accordingly. The following aspects of chassis configuration may be set:

2-8
Syslog Server IP Address, and Facility code
IB0054607-01 A
2–FastFabric TUI Menu
Fast Fabric IB Chassis Setup/Admin Menu

NTP Server IP Address

Time zone and Daylight Savings Time (DST)

Maximum MTU size, VL Capability and Link Layer Credit Distribution

Link Width Supported

IB Node Description (configured to match chassis ethernet name)
The IB node description must be a string consisting of the characters A–Z,
a–z,0–9, and underscore. No spaces are allowed in the node description
string, and it may not begin with a digit.,

IB Node Description format (concise format or verbose format)

Disable chassis auto clear of port counters
NOTE:
It is required to set the chassis IP address using the chassis serial port CLI
during initial chassis installation and setup.
Setup Password-less SSH/SCP
(Switch) This will run the setup_ssh -C command. This will set up secure
password-less SSH such that the InfiniBand Management Node can securely
login to all the other chassis as admin through the management network without
requiring a password.
Password-less SSH avoids the need to enter the chassis password for each
FastFabric chassis operation. It also avoids the need to put the chassis password
in /etc/sysconfig/fastfabric.conf. Once password-less ssh is set up, the
password in the chassis may be changed without impacting the ability to use
password-less ssh.
As part of this operation, the user will be prompted:
Would you like to override the default Chassis password? [n]:
If y is answered, the user will be prompted for the current password to use to login
into each chassis. Otherwise the password specified in
/etc/sysconfig/fastfabric.conf will be used.
Reboot Chassis
(Switch) This will run the iba_chassis_admin reboot command to reboot each
chassis listed in the /etc/sysconfig/iba/chassis file that was created in an
earlier step. it also ensures that they are rebooted (as verified using ping over the
management network).
IB0054607-01 A
2-9
2–FastFabric TUI Menu
Fast Fabric IB Chassis Setup/Admin Menu
Configure Chassis Fabric Manager
(Switch) The Configure Chassis Fabric Manager selection will assist in
configuring the Fabric Manager for any QLogic 12000 chassis with appropriate
license keys. This operation will be skipped for other chassis models.
Prompts will first guide the user through selection or generation of a
qlogic_fm.xml file. When generate is selected, the config_generate
command is used to guide the user through selecting Fabric Manager
configuration options. See the QLogic Fabric Manager User Guide for more
information about config_generate.
Prompts will guide the user through selecting the config update options:

run - Ensures given Fabric Manager configuration is on the chassis and
running. As needed push Fabric Manager configuration to each chassis. It
will unconditionally restart the Fabric Manager on master Management
Modules (MM), it will make sure it is not running on slave MMs.

runall - Ensures given Fabric Manager configuration is on the chassis and
running. As needed push Fabric Manager configuration to each chassis. It
will unconditionally restart the FM on master and slave MMs

push - Ensures given Fabric Manager configuration is on the chassis. As
needed push Fabric Manager configuration to each chassis.
Prompts will also guide the user through selecting the Fabric Manager autostart
options:

enable - Enables Fabric Manager start on master MM upon chassis
boot/reboot, disabled Fabric Manager autostart on any slave MMs in
selected chassis.

enableall - Enables Fabric Manager start on master and any slave MMs in
selected chassis upon boot/reboot.

disable - Disables Fabric Manager start on master and any slave MMs in
selected chassis upon boot/reboot.
Additional options prompted for:


parallel vs serial update
prompting for chassis password (default is to have password in
fastfabric.conf or to use password-less ssh)
If any chassis fails to be updated, use the View iba_chassis_admin result files
option to review the result files from the update. Refer to FastFabric Command
Line Interface Reference Guide for more details.
2-10
IB0054607-01 A
2–FastFabric TUI Menu
Fast Fabric IB Chassis Setup/Admin Menu
Get Basic Chassis Configuration
(Switch) The Get Basic Chassis Configuration supports a new feature to
retrieve basic information from chassis such as syslog, NTP configuration, time
zone information, MTU Capability, VL Capability, VL Credit Distribution, Link Width
and node description. The following is an example of the information retrieved:
TEST SUITE getconfig CASE (getconfig.i12k71f.getconfig) get i12k71f
i12k71f:
Syslog Configuration
: Syslog host set to: 192.168.2.15 port 514
facility 22
NTP
: Configured to use NTP server: 10.32.2.3
time zone
: Current time zone offset is: -5
MTU Capability
: 2048 Bytes
VL Capability
: 2 VLs
VL Credit Distribution : 4
LinkWidth Support
: 4X
Node Description
: i12k71f
Auto clear status
: Auto clear is disabled
PASSED
Check Status of IB Ports
(Switch or All): Check Status of IB Ports allows the state and error counts of all
ports to be checked and reviewed.
Based on the answers to the prompts shown below either the showallports -C
or the iba_report command will be used.
Would you like to perform fabric error analysis? [y]:
Clear error counters after generating report? [y]:
Would you like to perform fabric link speed error analysis?
[y]:
Check for links configured to run slower than supported? [n]:
Check for links connected with mismatched speed potential?
[n]:
(All): The answer to Would you like to perform fabric error
analysis selects whether iba_report -o errors should be run. If the user
answers y, the Clear error counters after processing report
question is asked. If the user answers y, the -C option will also be used on
iba_report to clear the port error counters after doing the error analysis.
IB0054607-01 A
2-11
2–FastFabric TUI Menu
Fast Fabric IB Chassis Setup/Admin Menu
(All): The answer to Would you like to perform fabric link speed
error analysis indicates whether iba_report -o slowlinks should be
run. If the user answers y to this question, the Check for links
configured to run slower than supported question is asked. If the
user answers y, the -o misconfiglinks option will also be used for
iba_report. Additionally, if the user answers y to the Would you like to
perform fabric link speed error analysis question, the Check for
links connected with mismatched speed potential question will also
be asked. If the user answers y, the -o misconnlinks option is also be
used for iba_report.
Answering y to all but the last 2 questions is recommended (these are the
defaults for each prompt). This checks all the ports in the fabric for any links
that have high error rates or are running at a lower speed than expected.
Any identified links should be diagnosed and corrected.
NOTE:
If the fabric is homogeneous and all links are expected to be running at
full speed, answer y to the last two questions as well.
(Switch): If the user responds n to all of the prompts, the showallports -C
command is run to allow the state and symbol error counts of all chassis ports to
be manually reviewed.
Control Chassis Fabric Manager
(Switch) The Configure Chassis Fabric Manager selection will assist in
controlling the Fabric Manager for any QLogic 12000 chassis with appropriate
license keys. This operation will be skipped for other chassis models.
Prompts will guide the user through selecting the control options:

restart - Unconditionally restart the Fabric Manager on master Management
Modules (MM), it will make sure it is not running on slave MMs.

restartall - Unconditionally restart the FM on master and slave MMs

run - make sure FM is running on master Management Modules (MM), it will
make sure it is not running on slave MMs.

runall - make sure FM is running on master and slave MMs

stop - Stop FM on master and slave MMs.
Prompts will also guide the user through selecting the Fabric Manager autostart
options:

enable - Enables Fabric Manager start on master MM upon chassis
boot/reboot, disabled Fabric Manager autostart on any slave MMs in
selected chassis.
2-12
IB0054607-01 A
2–FastFabric TUI Menu
Fast Fabric InfiniBand Externally Managed Switch Setup/Admin Menu

enableall - Enables Fabric Manager start on master and any slave MMs in
selected chassis upon boot/reboot.

disable - Disables Fabric Manager start on master and any slave MMs in
selected chassis upon boot/reboot.
Additional options prompted for:

parallel vs serial operation

prompting for chassis password (default is to have password in
fastfabric.conf or to use password-less ssh)
If any chassis fails the operation, use the View iba_chassis_admin result files
option to review the result files from the update. Refer to QLogic FastFabric
Command Line Interface Reference Guide for more details.
Generate all Chassis Problem Report Information
(Switch) This will run the captureall -C command to collect configuration and
status information from all selected chassis and generate a single *.tgz file that
can be sent to the Support Representative.
Run a command on all chassis
(Switch) This will run the cmdall -C command. A Chassis CLI command may be
specified to be executed against all selected chassis.
View iba_chassis_admin results files
(All) This permits viewing of the test.log and test.res files which reflect the
results from iba_chassis_admin runs (such as for updating Chassis Firmware or
rebooting all chassis per menu items above). The user is also given the option to
remove these files after viewing them.
If not removed, subsequent runs of iba_chassis_admin, iba_host_admin or
iba_switch_admin from within the current directory will continue to append to
these files.
Fast Fabric InfiniBand Externally Managed
Switch Setup/Admin Menu
This menu is focused on administration of QLogic 12200 and SilverStorm 9024FC
externally managed switches. Pressing the keys corresponding to menu items
(0-8) will toggle the Skip/Perform selection for the given item. More than one item
may be selected. Once the desired set of items have been selected, enter P. To
unselect all items, enter N. Entering X or pressing ESC will exit this menu and
return to the Main Menu.
Selecting 2 from the Fast Fabric InfiniBand Tools menu (Figure 2-3), displays
the Fast Fabric IB Externally Managed Switch Setup/Admin Menu (Figure 2-5)
IB0054607-01 A
2-13
2–FastFabric TUI Menu
Fast Fabric InfiniBand Externally Managed Switch Setup/Admin Menu
Fast Fabric IB Switch Setup/Admin Menu
Externally Managed Switch List: /etc/sysconfig/iba/ibnodes
Setup:
0) Edit Config and Select/Edit Switch Files
[ Skip
]
1) Test for Switch Presence
[ Skip
]
2) Verify Switch Firmware
[ Skip
]
3) Update Switch Firmware
[ Skip
]
4) Setup Switch Basic Configuration
[ Skip
]
5) Reboot Switch
[ Skip
]
6) Report Switch Firmware & Hardware Info
[ Skip
]
7) Get Basic Switch configuration
[ Skip
]
8) Report Switch VPD Information
[ Skip
]
9) Generate all Switch Problem Report Info
[ Skip
]
a) View iba_switch_admin result files
[ Skip
]
P) Perform the selected actions
N) Select None
Admin:
Review:
X) Return to Previous Menu (or ESC)
Figure 2-5. Fast Fabric IB Switch Setup/Admin Menu
Menu Items Description
Selecting items 0 through 9 or a will change the item from skip to perform.
Selecting N will unselect all items and X will exit the menu system. The items are
described below.
Edit Config and Select/Edit Switch Files
(Switch) This will permit the fastfabric.conf, ports, and ibnodes files to be
edited. The fastfabric.conf file controls the default operation of the FastFabric
Tools. The values specified in the fastfabric.conf specify the defaults that will
be used for all FastFabric operations performed using this o access the fabric(s)
being used. Existing environment variables will override the values in this ports
file. The ibnodes file selected and created using this menu should list Externally
Managed 12000 and 9024FC Switches that are to be used. After editing the three
files, an opportunity is given to edit them again or continue forward.
Do you want to edit/review/change the files? [y]:
The default will repeat the editing process, answer n to continue forward.
2-14
IB0054607-01 A
2–FastFabric TUI Menu
Fast Fabric InfiniBand Externally Managed Switch Setup/Admin Menu
Refer to FastFabric Command Line Interface Reference Guide for more details
about the format of the ibnodes and ports file, and about the iba_gen_ibnodes
command which can help generate the ibnodes file.
Test for Switch Presence
(Switch) This will run the iba_switch_admin ping command to test for the
presence of the selected switches in the fabric.
Verify Switch Firmware
(Switch) This will run the iba_switch_admin fwverify command to verify the
integrity of the present firmware in the switch. If this operation fails, prior to any
switch reboots or power-offs of the switch, perform Update Switch Firmware to
correct the firmware in the switch.
Update Switch Firmware
(Switch) This will run the iba_switch_admin upgrade command to permit the
switch firmware version to be updated and the switch node name to be set.
NOTE:
Consult the relevant switch firmware release notes to ensure any
prerequisites for the upgrade to the new firmware level have been met prior
to performing the upgrade using FastFabric.
Prompts will guide the user through options:

Selection of firmware files or directory containing .emfw files

Reboot switch after update (needed to run new firmware)

Parallel vs serial update

Prompting for switch password
IB0054607-01 A
2-15
2–FastFabric TUI Menu
Fast Fabric InfiniBand Externally Managed Switch Setup/Admin Menu
NOTE:
Since the InfiniBand fabric itself is used to update externally managed
switches, updating multiple switches with the reboot option may disrupt
parallel update operations. If there are no selected externally managed
switches in the path from the InfiniBand Management Node to any other
externally managed switch, parallel operations may be used. For example if
an InfiniBand Management node is connected directly to a core switch and
the externally managed switches are only at the edges. If in doubt, do not
use parallel update. Be aware that non-parallel operation for a fabric with
many externally managed switches could take a significant amount of time.
Another alternative is to perform the update in parallel without a reboot and
then perform the reboot separately using the Reboot Switch menu
selection. To control the order of the rebooting of externally managed
switches by FastFabric, see the discussion of the distance value in
“Externally Managed Switch List File” on page A-13.
If any switches fail to be updated, use the View iba_switch_admin result files
option to review the result files from the update. Refer to FastFabric Command
Line Interface Reference Guide for more details.
Setup Switch Basic Configuration
(Switch) This will run the iba_switch_admin configure command which will
prompt for switch configuration settings and then configure all the selected QLogic
12200 switches accordingly. The following aspects of switch configuration may be
set:

Maximum MTU size

VL Capability

Link Layer Credit Distribution

Link Width Supported

Link Speed Supported

IB Node Description
NOTE:
Normally, the IB Node Description is updated as part of a firmware upgrade;
however, the user is given the option to update the node description outside
of an upgrade procedure.
2-16
IB0054607-01 A
2–FastFabric TUI Menu
Fast Fabric InfiniBand Externally Managed Switch Setup/Admin Menu
NOTE:
This only operates on QLogic 12200 switches. Any 9024FC switches
selected will be skipped without change.
Reboot Switch
(Switch) This will run the iba_switch_admin reboot command to reboot all the
switches listed in the /etc/sysconfig/iba/ibnodes file that was created in a
previous step.
Report Switch Firmware & Hardware Info
(Switch) This will run the iba_switch_admin info command to provide a
summary of the present state for all the selected switches.
The information reported by this option includes:

Firmware Version

Hardware Version

Hardware Part Number

Switch Capability (SDR, DDR, or QDR)

Present Fan Status

Present status of both power supplies (in some models the second power
supply is optional)
If information fails to be reported for any switches, use the View
iba_switch_admin result files option to review the result files from the update.
Refer to FastFabric Command Line Interface Reference Guide for more details.
Get Basic Switch configuration
(Switch) This will run the iba_switch_admin -S -L
/etc/sysconfig/iba/ibnodes getconfig command to retrieve basic
information from externally managed switch such as: MTU, VL Cap, Credit
Distribution, Link Width, Link Speed and node description.
The following files are produced from this selection:

test.res - appended with summary results of run

test.log - appended with detailed results of run

save_tmp/ - contains a directory per failed operation with detailed logs

test_tmp*/ - intermediate result files while operation is running
IB0054607-01 A
2-17
2–FastFabric TUI Menu
Fast Fabric InfiniBand Externally Managed Switch Setup/Admin Menu
Report Switch VPD Information
(Switch) This will run the iba_switch_admin hwvpd command to provide the
Virtual Product Data (VPD) for all the selected switches. This information can be
useful for inventory and asset control as well as to provide details about the
product to customer support.
The information reported by this option includes:

Serial Number

Part Number

Model Number

Hardware Version

Manufacturer

Product description

Manufacturer ID code

Manufacture date

Manufacture time of day
If information fails to be reported for any switches, use the View
iba_switch_admin result files option to review the result files from the update.
Refer to FastFabric Command Line Interface Reference Guide for more details.
Generate All Switch Problem Report Info
(Switch) This menu item runs the captureall -n -S -p command to collect
configuration and status information from all selected switches, then generates a
single *.tgz file that can be sent to Technical Support.
View iba_switch_admin result files
(All) This permits viewing of the test.log and test.res files that reflect the
results from iba_switch_admin runs (such as those for updating Switch
Firmware or rebooting all switches per menu items above). The user is also given
the option to remove these files after viewing them.
If not removed, subsequent runs of iba_chassis_admin, iba_host_admin or
iba_switch_admin from within the current directory will continue to append to
these files.
2-18
IB0054607-01 A
2–FastFabric TUI Menu
Fast Fabric IB Host Setup
Fast Fabric IB Host Setup
This menu is focused on initial host setup and installation of InfiniBand software
on all the hosts. Pressing the keys corresponding to menu items (0-9, a-d) will
toggle the Skip/Perform selection for the given item. More than one item may be
selected. Once the desired set of items have been selected, enter P. To unselect
all items, enter N. Entering X or pressing ESC will exit this menu and return to the
Main Menu.
Selecting 3 from the Fast Fabric InfiniBand Tools menu (Figure 2-3), displays
the Fast Fabric IB Host Setup Menu (Figure 2-6)
Fast Fabric IB Host Setup Menu
Host List: /etc/sysconfig/iba/hosts
Setup:
0) Edit Config and Select/Edit Hosts Files
[ Skip
]
1) Verify Hosts via Ethernet ping
[ Skip
]
2) Setup Password-less ssh/scp
[ Skip
]
3) Copy /etc/hosts to all hosts
[ Skip
]
4) Show uname -a for all hosts
[ Skip
]
5) Install/Upgrade QLogic IB Software
[ Skip
]
6) Configure IPoIB IP Address
[ Skip
]
7) Build MPI Test Apps and Copy to Hosts
[ Skip
]
8) Reboot Hosts
[ Skip
]
9) Refresh ssh Known Hosts
[ Skip
]
a) Rebuild MPI Library and Tools
[ Skip
]
b) Run a command on all hosts
[ Skip
]
c) Copy a file to all hosts
[ Skip
]
d) View iba_host_admin result files
[ Skip
]
P) Perform the selected actions
N) Select None
Admin:
Review:
X) Return to Previous Menu (or ESC)
Figure 2-6. Fast Fabric IB Host Setup Menu
Menu Items Description
Selecting items 0 through 9 and a through d will change the item from skip to
perform. Selecting N will unselect all items and X will exit the menu system. The
items are described below.
IB0054607-01 A
2-19
2–FastFabric TUI Menu
Fast Fabric IB Host Setup
Edit Configuration and Select/Edit Hosts Files
(All) This will permit the hosts and fastfabric.conf files to be edited. The hosts
file selected and created using this menu should not list the FastFabric host itself.
After editing the two files, an opportunity is given to edit them again or continue
forward.
Selected Host File: /etc/sysconfig/iba/hosts
Do you want to edit/review/change the files? [y]:
The default will repeat the editing process, answer n to continue forward.
Refer to FastFabric Command Line Interface Reference Guide for more details
about the format of the hosts file.
Verify Hosts via Ethernet ping
(All) This will run the pingall command. All the hosts listed will be pinged
through the Management Network.
Setup Password-less SSH/SCP
(Linux) This will run the setup_ssh -s -S -i “” command. This will setup
secure password-less SSH such that the InfiniBand Management Node can
securely login to all the other hosts as root through the management network
without requiring a password. The user will be prompted for the present password
of the hosts, the same password will be used to login to all selected hosts. Once
password-less ssh is setup, the password in the hosts may be changed without
impacting the ability to use password-less ssh.
Password-less SSH is required by FastFabric, MPI test applications and most
versions of MPI (including QuickSilver MPI, OFED openmpi, OFED mvapich, and
OFED mvapich2).
Copy /etc/hosts to all hosts
(Linux) This will run the scpall /etc/hosts /etc/hosts command to copy
the /etc/hosts file on this host to all the other selected hosts. This is not
necessary when using a DNS server to resolve hostnames for the cluster.
Show uname -a for all hosts
(Linux) This run the cmdall “uname -a” command to show the OS version on
all the hosts. Review the results carefully to verify all the hosts have the expected
OS version. In typical clusters all hosts will be running the same OS and kernel
version.
2-20
IB0054607-01 A
2–FastFabric TUI Menu
Fast Fabric IB Host Setup
Install/Upgrade QLogic IB Software
(Host) This will run the iba_host_admin load or iba_host_admin update
command to install the QLogicIB software on all the hosts. By default it will look in
the current directory for the FF_PRODUCT.FF_PRODUCT_VERSION.tgz file. If it is
not found in the current directory, it will prompt for input of a directory name where
this file can be found.
Prompts will guide the user through options:

upgrade - updates all servers with new release. Only components
previously installed are upgraded. Will fail for any hosts which have no
QLogicIB software currently installed

initial install/load - uninstalls any existing QLogicIB software and
installs the given release based on fastfabric.conf installation options
specified.
After the install is completed, the hosts will still need to be rebooted to bring up the
new InfiniBand drivers. This can be performed using Reboot Hosts option.
If any hosts fail to be updated, use the View iba_host_admin result files option to
review the result files from the update. For more details, see FastFabric
Command Line Interface Reference Guide.
NOTE:
When using the QLogicIB packaging of OFED, the entire InfiniBand stack
may be installed using FastFabric, which is the recommended approach.
When using other packagings of OFED, FastFabric may be used to install
the QLogicIB Tools (InfiniServTools.FF_PRODUCT_VERSION.tgz) on the
remaining hosts. To do so, FF_PRODUCT_VERSION must be set to the desired
InfiniServTools release level and OFED must be installed on each host
manually.
NOTE:
The InfiniBand software selected for installation must be appropriate for the
OS version and distribution installed on the destination hosts.
Configure IPoIB IP Address
(Host) This will run the iba_host_admin configipoib command to create the
ifcfg-ib0 files on each host. The file will be created with a statically assigned
IPv4 address. The IPoIB IP address for each host will be determined by the
resolver (Linux host command). If not found using the resolver, /etc/hosts on
the given host will be checked.
IB0054607-01 A
2-21
2–FastFabric TUI Menu
Fast Fabric IB Host Setup
Build MPI Test Apps and Copy to Hosts
(Host) This will build the MPI sample benchmarks on the InfiniBand Management
Node and copy the resulting object files to all the hosts. This is in preparation for
execution of MPI performance tests and benchmarks in a later step.
NOTE:
This option is available for the QLogicIB packaging of OFED, but is not
presently available for other packagings of OFED.
Reboot Hosts
(Linux) This will run the iba_host_admin reboot command to reboot all the
selected hosts and ensure they go down and come back up (as verified using ping
over the management network). When the hosts come back up, they will be
running the InfiniBand software installed.
Refresh SSH Known Hosts
(Linux) This will run the setup_ssh -U “” command to refresh the ssh known
hosts list on this server for the Management Network. This may be used to update
security for this host if hosts are replaced, reinstalled, renamed, or repaired.
Rebuild MPI Library and Tools
(Host) This will rebuild the MPI Library itself and related tools (such as mpirun).
This will be performed using the do_build tool supplied with the MPI Source.
When rebuilding MPI, do_build will prompt the user for selection of which MPI
(openmpi, mvapich or mvapich2) to rebuild and provide choices as to which
available compiler to use. Consult the QLogic HCA and QLogic OFED Software
Users Guide for more information.
NOTE:
This option is available for the QLogicIB packaging of OFED, but is not
presently available for other packagings of OFED.
Run a command on all hosts
(Linux) This will run the cmdall command. A Linux shell command (or sequence
of commands separated by semicolons) may be specified to be executed against
all selected hosts.
Copy a file to all hosts
(Linux) This will run the scpall command. A file on the local host may be
specified to be copied to all selected hosts.
2-22
IB0054607-01 A
2–FastFabric TUI Menu
Fast Fabric IB Host Verification/Admin Menu
View iba_host_admin result files
(All) This permits viewing of the test.log and test.res files that reflect the
results from iba_host_admin runs (such as for installing InfiniBand software or
rebooting all hosts per menu items above). The user is also given the option to
remove these files after viewing them.
If not removed, subsequent runs of iba_chassis_admin, iba_host_admin or
iba_switch_admin from within the current directory will continue to append to
these files.
Fast Fabric IB Host Verification/Admin Menu
This menu is focused on verifying hosts and the fabric as well as administration of
all the hosts. Pressing the keys corresponding to menu items (0-9, a-b) will toggle
the Skip/Perform selection for the given item. More than one item may be
selected. Once the desired set of items have been selected, enter P. To unselect
all items, enter N. Entering X or pressing ESC will exit this menu and return to the
Main Menu.
Selecting 4 from the Fast Fabric InfiniBand Tools menu (Figure 2-3), displays
the Fast Fabric IB Host Verification/Admin Menu (Figure 2-7)
IB0054607-01 A
2-23
2–FastFabric TUI Menu
Fast Fabric IB Host Verification/Admin Menu
Fast Fabric IB Host Verification/Admin Menu
Host List: /etc/sysconfig/iba/allhosts
Validation:
0) Edit Config and Select/Edit Hosts Files
[ Skip
]
1) Verify Hosts via Ethernet ping
[ Skip
]
2) Summary of Fabric Components
[ Skip
]
3) Check Status of IB Ports
[ Skip
]
4) Verify Hosts see each other
[ Skip
]
5) Verify Hosts ping via IPoIB
[ Skip
]
6) Refresh ssh Known Hosts
[ Skip
]
7) Check MPI Performance
[ Skip
]
8) Check Overall Fabric Health
[ Skip
]
9) Generate all Hosts Problem Report Info
[ Skip
]
a) Run a command on all hosts
[ Skip
]
b) View iba_host_admin result files
[ Skip
]
P) Perform the selected actions
N) Select None
Admin:
Review:
X) Return to Previous Menu (or ESC)
Figure 2-7. Fast Fabric IB Host Verification/Admin Menu
Menu Items Description
Selecting items 0 through 9 and a through b will change the item from skip to
perform. Selecting N will unselect all items and X will exit the menu system. The
items are described below.
Edit Config and Select/Edit Hosts Files
(All) This will permit the allhosts, ports, and fastfabric.conf files to be
edited. The allhosts file selected and created using this menu should list the
FastFabric host itself. After editing the two files, an opportunity is given to edit
them again or continue forward.
Selected Host File: /etc/sysconfig/iba/allhosts
Do you want to edit/review/change the files? [y]:
The default will repeat the editing process, enter n to continue forward.
Refer to FastFabric Command Line Interface Reference Guide for more details
about the format of the allhosts and ports files.
2-24
IB0054607-01 A
2–FastFabric TUI Menu
Fast Fabric IB Host Verification/Admin Menu
Verify Hosts via Ethernet ping
(All) This will run the pingall command. All the hosts listed will be pinged
through the Management Network.
Summary of Fabric Components
(All) This will run the fabric_info command to provide a brief summary of the
counts of components in the fabric including how many switch chips, hosts, and
links are in the fabric. It will also indicate if any 1x links were found (that could
indicate a poorly seated or bad cable). Review the results against the expected
configuration of the cluster.
NOTE:
The link count includes some internal links within the switch boxes. This
means that the count displayed will be greater than the actual number of
cables.
Check Status of IB Ports
(Host or All): Check Status of IB Ports allows the state and error counts of all
ports to be checked and reviewed.
Based on the answers to the prompts shown below either the showallports or
the iba_report command will be used.
Would you like to perform fabric error analysis? [y]:
Clear error counters after generating report? [y]:
Would you like to perform fabric link speed error analysis?
[y]:
Check for links configured to run slower than supported? [n]:
Check for links connected with mismatched speed potential?
[n]:
(All): The answer to Would you like to perform fabric error
analysis selects whether iba_report -o errors should be run. If the user
enters y, the Clear error counters after processing report question
is asked. If the user enters y, the -C option will also be used on iba_report
to clear the error counters after doing the error analysis.
IB0054607-01 A
2-25
2–FastFabric TUI Menu
Fast Fabric IB Host Verification/Admin Menu
(All): The answer to Would you like to perform fabric link speed
error analysis indicates when iba_report -o slowlinks should be
run. If the user enters y to this question, the Check for links configured
to run slower than supported question is asked. If the user enters y, the
-o misconfiglinks option will also be used for iba_report. Additionally, if
the user enters y to the Would you like to perform fabric link speed
error analysis question, the Check for links connected with
mismatched speed potential question will also be asked. If the user
enters y, the -o misconnlinks option is also be used for iba_report.
Entering y to all but the last 2 questions is recommended (these are the
defaults for each prompt). This checks all the ports in the fabric for any links
that have high error rates or are running at a lower speed than expected.
Any identified links should be diagnosed and corrected.
NOTE:
If the fabric is homogeneous and all links are expected to be running at full
speed, enter y to the last two questions as well.
(Host): If the user enters n to all of the prompts, the showallports command is
run to allow the state and symbol error counts of all host ports to be manually
reviewed. Selection of this option requires that QLogicIB OFED+ software be
installed on all hosts being checked.
Verify Hosts see each other
(Host) This will run the iba_host_admin sacache command to verify that each
host can see all the others through queries to the Subnet Administrator. For the
OFED+ stack this also verifies the Distributed SA on each host has been fully
populated.
NOTE:
This operation requires that the hosts being queried be specified by a
resolvable TCP/IP host name. This operation will FAIL if the selected hosts
are specified by IP address. See the FastFabric Command Line Interface
Reference Guide, Selection of Hosts section for more information.
Verify Hosts ping via IPoIB
(Host) This will run the iba_host_admin ipoibping command to verify that
IPoIB is properly configured and running on all the hosts. This is accomplished
through the InfiniBand management node pinging each host using IPoIB.
2-26
IB0054607-01 A
2–FastFabric TUI Menu
Fast Fabric IB Host Verification/Admin Menu
Refresh SSH Known Hosts
(Linux) This will run the setup_ssh -U command to refresh the SSH known
hosts list on this server for the IPoIB and Management Networks. This may be
used to update security for this host if hosts are replaced, reinstalled, renamed, or
repaired.
Check MPI Performance
(Host) This will do a quick check of PCI and MPI performance using end to end
latency and bandwidth tests.
NOTE:
This option is available for the QLogicIB packaging of OFED, but is not
presently available for other packagings of OFED.
Based on the answer to the prompt shown below either the iba_host_admin
mpiperfdeviation or the iba_host_admin mpiperf command will be used.
Test Latency and Bandwidth deviation between all hosts? [y]:
Answering y is recommended, this will run the iba_host_admin
mpiperfdeviation command to do pairwise analysis of latency and bandwidth for
the selected hosts and report pairs outside an acceptable tolerance range. By
default performance is compared relative to other hosts in the fabric (with the
assumption that all hosts selected for a given run should have comparable
InfiniBand performance). Failing hosts will be clearly indicated.
Answering n will run the iba_host_admin mpiperf command. This displays the
MPI latency and bandwidth between pairs of hosts (1-2, 3-4, 5-6, etc). The numbers
reported should be checked against the practical PCI speeds in the Performance
Impact table in the QLogic Fabric Software Installation Guide. If any pairs are not
in the expected performance range, it should be considered a failure for those pairs
of hosts.
For either test, If any hosts fail, carefully examine the failing hosts to verify the HCA
models, PCI slot used, BIOS settings and any motherboard jumpers related to
devices on PCI buses or slot speeds. Also verify the HCA and riser cards are
properly seated.
The results of either test are also written to the test.res file which may be
viewed using the View iba_host_admin result files. Refer to FastFabric
Command Line Interface Reference Guide for more details.
The bandwidth reported should also be checked against the practical PCI speeds
in the Performance Impact section. If all pairs are not in the expected performance
range, carefully examine all hosts to verify the HCA models, PCI slot used, BIOS
IB0054607-01 A
2-27
2–FastFabric TUI Menu
Fast Fabric IB Host Verification/Admin Menu
settings and any motherboard jumpers related to devices on PCI buses or slot
speeds.
Check Overall Fabric Health
(Host) This will run the all_analysis command to check the overall fabric
health.
The user will be prompted:
Baseline present configuration? [n]:
If the user enters y, a new baseline will be created using the present fabric
configuration. If the user enters n, the present fabric state is checked against the
baseline and the general health of the fabric is also checked.
Generate all Hosts Problem Report Info
(Host) This will run the captureall command to collect configuration and status
information from all hosts and generate a single *.tgz file which can be sent to
the Support Representative.
Based on the answer to the prompt shown below, various levels of detail about the
fabric can be included in the capture.
Capture detail level (1=Normal, 2-Fabric, 3-Fabric+FDB,
4-Analysis):
The Details levels are:
1-Normal - obtains local information from each host
2-Fabric - in addition to “Normal”, also obtains basic fabric information by
queries to the SM and fabric error analysis using iba_report.
3-Fabric+FDB - in addition to “Fabric”, also obtains all the switch forwarding
tables and InfiniBand multicast membership lists from the SM.
4-Analysis - in addition to “Fabric+FDB”, also obtains all_analysis
results. If all_analysis has not yet been run, it is run as part of the
capture.
NOTE:
Detail levels 2-4 can be used when fabric operational problems occur. If the
problem is most likely node specific, detail level 1 should be sufficient. Detail
levels 2-4 require an operational Fabric Manager. Typically your support
representative will request a given detail level. If a given detail level takes
excessively long or fails to be gathered, try a lower detail level.
2-28
IB0054607-01 A
2–FastFabric TUI Menu
Fabric Monitoring
For detail levels 2-4, the additional information is only gathered on the node running
the captureall command. The information is gathered for every fabric specified in
the /etc/sysconfig/iba/ports file.
Run a command on all hosts
(Linux) This will run the cmdall command. A Linux shell command (or sequence
of commands separated by semicolons) may be specified to be executed against
all selected hosts.
View iba_host_admin result files
(All) This permits viewing of the test.log and test.res files which reflect the
results from iba_host_admin runs (such as those for installing InfiniBand
software or rebooting all hosts per menu items above). The user is also given the
option to remove these files after viewing them.
If not removed, subsequent runs of iba_chassis_admin, iba_host_admin or
iba_switch_admin from within the current directory will continue to append to
these files.
Fabric Monitoring
This menu is focused on monitoring the performance of the fabric. Pressing the
key corresponding to menu item (0) will toggle the Skip/Perform selection for the
given item. More than one item may be selected. Once the desired set of items
have been selected, enter P. To unselect all items, enter N. Entering X or pressing
ESC will exit this menu and return to the Main Menu.
Selecting 5 from the Fast Fabric InfiniBand Tools menu (Figure 2-3), displays
the Fast Fabric IB Fabric Monitoring Menu (Figure 2-8)
Fast Fabric IB Fabric Monitoring Menu
0) Fabric Performance Monitoring
[Perform]
P) Perform the selected actions
N) Select None
X) Return to Previous Menu (or ESC)
Figure 2-8. Fast Fabric IB Fabric Monitoring Menu
Menu Items Description
Selecting item 0 will change the item from skip to perform. Selecting N will
unselect all items and X will exit the menu system. The item is described below.
IB0054607-01 A
2-29
2–FastFabric TUI Menu
Fabric Monitoring
Fabric Performance Monitoring
(All) This selection initiates iba_top. for full details about iba_top refer to
Section 3
2-30
IB0054607-01 A
3
iba_top Fabric Performance
Monitor
Introduction
iba_top is a command line tool which displays performance, congestion, and
error information about a fabric. Fabric information is divided into two areas performance and error statistics, which are the main starting points for analyzing
fabric traffic. Performance (bandwidth utilization) can identify over-utilized areas
(bottle necks) and under-utilized areas (potentially misconfigured); errors can
identify problems in fabric hardware or configuration, as well as congestion and
other performance situations.
iba_top TUI
The iba_top TUI screen layout and options consist of four areas in the layout.
Image Identification, Screen Specific Information, Common Input Commands, and
Screen Specific Input Commands. Figure 3-9shows the top level (summary
screen which is being used as an example to show the screen layout. The section
following the figure explains each of the areas and the common commands that
are available on each screen.
IB0054607-01 A
3-1
3–iba_top Fabric Performance Monitor
iba_top TUI
iba_top: Img:Tue Apr 13 14:11:46 2010, Live
Summary:
Link:21
SW:4
SM:1
NodeFail:0
AvgMBps
0 All
Int
1 HCAs
TCA-Port:0
PortSkip:3
MaxMBps
AvgKPps
MinKPps
0
0
0
0
SmaCong:min
Secure:min
Snd
0
0
0
0
0
Rcv
0
0
0
0
0
Congst:min
SmaCong:min
Secure:min
MaxKPps
0
0
0
Routing:min
0
0
0
0
0
0
Rcv
0
0
0
0
0
0
Congst:min
SmaCong:min
Secure:min
0
0
0
0
0
0
Snd
0
0
0
0
0
0
Rcv
0
0
0
0
0
Master-SM: LID:0x0001
Congst:min
Port:
1
SmaCong:min
Secure:min
Priority: 0
State:Master
Screen Specific
Information
Routing:min
Int
Integ:min
Image
Identification
Routing:min
Snd
Integ:min
3 SWs
PortFail:0
Congst:min
Integ:min
2 TCAs
HCA-Port:8
NodeSkip:0
MinMBps
0
Integ:min
SW-Port:38
0
Routing:min
Name:admin1 HCA-1
PortGUID:0x00066A00A000EC5B
Secondary-SM: none
Quit up Live/rRev/fFwd/bookmrked Bookmrk Unbookmrk ?help | Pmcfg Imginfo 0-3:
Common Input
Commands
Screen-Specific
Input commands
Figure 3-9. iba_top TUI Screen Layout
Image Identification
The first display line of iba_top shows the timestamp for the PM sweep (image)
being displayed and the type of image (Live, Hist, Bkmk). If a Live image is not
being displayed, the current time ('Now:') is also shown.
Screen Specific Information
The information and layout of this area of the screen will vary depending on which
screen is selected. The information and the layout of this section will be discussed
for each specific screen in the following sections.
Command Entry
The last display line of iba_top is a prompt showing available input commands.
The left section of commands are available in every screen and perform the same
action in each screen. The right section are screen-specific. Commands are case
insensitive except as noted by *; the ENTER key must be pressed after a
command. Note that a help command ? is available at every screen and provides
information about the screen contents and input commands.
3-2
IB0054607-01 A
3–iba_top Fabric Performance Monitor
iba_top TUI
Common Input Commands
The following input commands are available in every screen:
Q: Quit program;
u*: Up to previous screen;
L: Select Live image;
R: Navigate reverse 1 (r*) or 5 (R*) sweeps;
F: Navigate forward 1 (f*) or 5 (F*) sweeps;
b*: Select (previously) Bookmarked image;
B*: Bookmark currently selected image;
U*: Unbookmark Bookmarked image;
?: Help
Screen-Specific Input Commands
The screen-specific input commands will be discussed with each screen
description in the following sections.
Access to Live and Recent PM Historical Data
iba_top allows the user to access statistics from sequential PM sweeps (the PM
keeps a history of previous sweep images) and queries the PM at a user-specified
interval (10 seconds by default). When iba_top queries for statistics for the most
recent PM sweep it is in “Live” mode. In Live mode the data will change, at the
iba_top interval rate, as iba_top queries new PM sweeps. At each screen
(summary or detail) the data being displayed is refreshed for the current PM
sweep. A PM sweep can be in “frozen” mode. The data in a frozen sweep will not
change, allowing the statistics to be examined in summary and detail screens.
Two user actions result in a sweep being frozen.
The first is when the user “Bookmarks” a sweep. A Bookmarked sweep will remain
frozen until the user explicitly “Unbookmarks” it; iba_top allows one sweep at a
time to be bookmarked. The second action is when the user moves (navigates)
iba_top's focus to another sweep within the history of sweeps maintained by
the PM. For the duration of iba_top's focus on such a sweep it will remain
frozen. Navigation can occur from Live mode or when displaying a Bookmarked
image; during navigation iba_top is in “Historic” mode. Navigation can be
performed backward or forward, 1 or 5 sweeps at a time.
IB0054607-01 A
3-3
3–iba_top Fabric Performance Monitor
iba_top TUI Screens
iba_top TUI Screens
Additional screens, described in the following paragraphs, are available to display
detailed information about: PM configuration, PM sweep (image) configuration,
performance statistics, error statistics, port group configuration, and port statistics
(port counters). The screens can be navigated in a hierarchal manner to examine
the state of a fabric.
Summary Screen
The top level (summary) screen of iba_top shows basic fabric configuration
information as well as performance and error information. An example of the
iba_top summary screen is shown in Figure 3-10.
iba_top: Img:Tue Apr 13 14:11:46 2010, Live
Summary:
Link:21
SW:4
SM:1
NodeFail:0
SW-Port:38
HCA-Port:8
NodeSkip:0
PortFail:0
TCA-Port:0
PortSkip:3
AvgMBps
MinMBps
MaxMBps
AvgKPps
MinKPps
MaxKPps
0
0
0
0
0
0
0 All
Int
1 HCAs
Snd
0
0
0
0
0
Rcv
0
0
0
0
0
Integ:min
Congst:min
Integ:min
2 TCAs
Congst:min
SmaCong:min
Secure:min
Secure:min
Routing:min
0
0
Routing:min
Snd
0
0
0
0
0
0
Rcv
0
0
0
0
0
0
Integ:min
3 SWs
SmaCong:min
Congst:min
SmaCong:min
Secure:min
Routing:min
Int
0
0
0
0
0
0
Snd
0
0
0
0
0
0
Rcv
0
0
0
0
0
Integ:min
Master-SM: LID:0x0001
Congst:min
Port:
1
SmaCong:min
Secure:min
Priority: 0
State:Master
Fabric
Configuration
Information
Performance and
Error Status for
Each Port Group
0
Routing:min
Name:admin1 HCA-1
PortGUID:0x00066A00A000EC5B
Secondary-SM: none
SM Configuration
Information
Quit up Live/rRev/fFwd/bookmrked Bookmrk Unbookmrk ?help | Pmcfg Imginfo 0-3:
Figure 3-10. iba_top Summary Screen
Fabric configuration information
Fabric configuration information includes numbers of links, switches, SMs, and
ports, as well as details about the master and secondary (if present) SMs.
3-4
IB0054607-01 A
3–iba_top Fabric Performance Monitor
iba_top TUI Screens
Performance and Error Statistics for Each Port Group
Fabric performance and error statistics are presented based on four groupings of
ports: All (all ports in the fabric), HCAs, TCAs and SWs. These groups provide a
natural subdivision of the ports in a fabric for analysis. For more information about
Groups and the operation of the PM, refer to the QLogic Fabric Manager User
Guide.
For each port group, average, minimum and maximum MBps (megabytes per
second), and KPps (kilopackets per second) are shown, as well as a status
indicator for each of five error categories.
Performance Statistics
Performance statistics for each port group are further divided into up to three
subgroups - Internal, Send and Receive - based on whether a port's neighbor port
is in its group. If a port's neighbor port is in its group, all performance statistics are
contained in the Internal subgroup. If a port's neighbor is not in its group, statistics
for data leaving the port (group) are contained in the Send subgroup and statistics
for data entering the port are contained in the Receive subgroup.
All Group
In the All group all ports are Internal since, by definition, the neighbor port must be
in the All group.
HCAs and TCAs Groups
In the HCAs and TCAs groups all neighbor ports are outside the group so
statistics are contained in the Send and Receive subgroups.
SWs group
In the SWs group neighbor ports are either outside the group (HCA or TCA) or
inside the group (another switch) so statistics are contained in all three subgroups.
Error Categories
The five error categories are:

Integ — Integrity

Congst — Congestion

SmaCong — SMA Congestion

Secure — Security

Routing — Routing
These error categories are each based on one or more port error counters. Each
error category’s status indicator is shown at one of five values/colors:
minimum/green, Low/blue, Moderate/cyan, Warning/yellow or OVER/red based
on the error value as compared to a threshold value.
IB0054607-01 A
3-5
3–iba_top Fabric Performance Monitor
iba_top TUI Screens
Screen-Specific Input Commands
The summary screen accepts the following input commands:
P: PM Configuration screen;
I: Image Information screen
0-3: Select Port Group - All (0), HCAs (1), TCAs (2), SWs (3);
Additional Screens
After looking at the summary screen a user can decide which area of the fabric
(performance or error) and which group of ports most warrants investigation, and
can then drill down into that area.
PM Configuration Screen
The PM Configuration screen (Figure 3-11) displays information as provided by
the PM (refer to the QLogic Fabric Manager User Guide). The Sweep Interval
parameter is separate from the iba_top interval. Normally the iba_top interval
should be set to a value greater than or equal to Sweep Interval. The PM
configuration screen shows the results for image information (total images, freeze
images, freeze lease time), error thresholds, integrity weights, PM memory
footprint, PMA MADs retry/timeout, and sweep information. The PM Config screen
has no screen-specific input commands.
iba_top: Img:Mon Feb
1 16:26:02 2010, Bkmk
Now:Mon Feb
1 18:07:09 2010
PM Config:
Sweep Interval:10 sec
PM Flags(0x17):CaStat EnSwVnd EHcaStat En64Bit
Total Images:10
Freeze Images:5
Freeze Lease:60
seconds
Err Thresholds: Integrity:100
Integrity Wts:
Congestion:100000
SmaCongest:100
Security:10
Routing:100
Symbol Err:1
Link Recovery:30
Port Rcv Err:1
Link Downed:30
Loc Link Integ:30
Congest Wts: Tx Discard %:1
Excess Bfr Ovrn:30
Tx Congest %:1
Tx Ineffic %:1
PM Memory Size:9 MB (10443860 bytes)
PMA MADs: MaxAttempts: 0
Sweep: MaxParallelNodes: 0
MinRespTimeout: 0
RespTimeout: 0
PmaBatchSize: 0
Quit up Live/rRev/fFwd/bookmrked Bookmrk Unbookmrk ?help |
Figure 3-11. PM Configuration Screen
3-6
IB0054607-01 A
3–iba_top Fabric Performance Monitor
iba_top TUI Screens
Image Information Screen
The Image Information screen (Figure 3-12) displays image information as
provided by the PM. Sweep start and duration, numbers of ports in each group,
node and port information for the sweep, and SM information is shown. The Image
Information screen has no screen-specific input commands.
iba_top: Img:Tue Apr 13 15:01:53 2010, Live
Image Info:
Sweep Start:Tue Apr 13 15:01:53 2010
Sweep Duration:0.016 Seconds
Num SW-Ports:38
HCA-Ports:8
TCA-Ports:0
Num SWs:4
Num Links:21
Num SMs:1
Num Fail Nodes:0
Fail Ports:0
Num Skip Nodes:0
Skip Ports:3
Master-SM: LID:0x0001
Port:
1
RTR-Ports:0
Unexpected Clear Ports:0
Priority: 0
State:Master
Name:admin1 HCA-1
PortGUID:0x00066A00A000EC5B
Secondary-SM: none
Quit up Live/rRev/fFwd/bookmrked Bookmrk Unbookmrk ?help |
Figure 3-12. Image Information Screen
Group Information Select Screen
The Group Information Select screen (Figure 3-13) allows the user to select the
type of group information to display for the group selected in the summary screen.
The following input commands are accepted and lead to the corresponding
screen.
W: Performance (Bandwidth Utilization) statistics
E: Error statistics
C: Group configuration (port list)
IB0054607-01 A
3-7
3–iba_top Fabric Performance Monitor
iba_top TUI Screens
iba_top: Img:Tue Apr 13 15:05:25 2010, Live
Group Info Select:All
NumIntPorts:43
NumExtPorts:0
Group BW Summary (W)
Group Err Summary (E)
Group Config (C)
Quit up Live/rRev/fFwd/bookmrked Bookmrk Unbookmrk ?help | W E C:
Figure 3-13. Group Information Screen
Bandwidth Statistics Screen
The Bandwidth Statistics screen (Figure 3-14) displays, for each valid
performance data subgroup (Internal, Send, Receive), the total, average,
minimum and maximum MBps and KPps. For each subgroup ten performance
'buckets', from 0+% to 90+% in 10% increments, count the number of ports whose
'MBps compared to link speed' value corresponds to that bucket. This provides an
indication of how the data rate of the group compares to its potential.
iba_top: Img:Mon Feb 1 16:26:02 2010, Bkmk Now:Mon Feb
Group BW Stats:HCAs Criteria:Util-High Number:10
1 17:31:24 2010
Snd TotMBps AvgMBps MinMBps MaxMBps TotKPps AvgKPps MinKPps MaxKPps
2307
769
20
1864
2274
758
20
1841
Buckt 0+%
1
10+%
0
20+%
1
30+%
0
40+%
0
50+%
0
60+%
0
70+%
0
80+%
0
90+%
1
Rcv TotMBps AvgMBps MinMBps MaxMBps TotKPps AvgKPps MinKPps MaxKPps
2307
769
0
1884
2274
758
0
1861
Buckt 0+%
1
10+%
0
20+%
1
30+%
0
40+%
0
50+%
0
60+%
0
70+%
0
80+%
0
90+%
1
Quit up Live/rRev/fFwd/bookmrked Bookmrk Unbookmrk ?help | cC N0-n Detail:
Figure 3-14. Bandwidth Statistics Screen
3-8
IB0054607-01 A
3–iba_top Fabric Performance Monitor
iba_top TUI Screens
The Bandwidth Statistics screen accepts input commands which specify
parameters to be used in a group focus query, which will provide a list of ports (in
the port group) sorted according to a specified performance criterion. The second
line of the Bandwidth Statistics screen displays the group name, and the currently
selected focus criterion and number of ports for a group focus query. The D
command causes the group focus query to be performed and displayed in a
Group Focus screen. The following input commands are accepted in the
Bandwidth Statistics screen.
C: Select group focus criterion forward (c*) or reverse (C*):
Util-High: Bandwidth Utilization (highest first)
UtlPkt-Hi: Packet Utilization (highest first)
Util-Low: Bandwidth Utilization (lowest first)
Nn: Number of entries n in group focus list
D:
Display detail group focus list
Error Statistics Screen
The Error Statistics screen (Figure 3-15) displays error statistics for a port group
and are divided into up to two subgroups, Internal or External, based on whether a
port's neighbor port is in its group (Internal) or not (External). In the All group all
ports are Internal. In the HCAs and TCAs groups all ports are External. In the SWs
group ports are Internal and External.
IB0054607-01 A
3-9
3–iba_top Fabric Performance Monitor
iba_top TUI Screens
iba_top: Img:Tue Apr 13 15:11:47 2010, Live
Group Err Stats:All
Criteria:Integ
Int
Max
0+%
25+%
50+%
75+%
100+%
Integrity
0
43
0
0
0
0
Congestion
0
43
0
0
0
0
SmaCongest
0
43
0
0
0
0
Security
0
43
0
0
0
0
Routing
0
43
0
0
0
0
Congest %:
0.0
Ext
Discard:
Number:10
0
Ineffic %:
0.0
Adapt Rt:0
Max
0+%
25+%
50+%
75+%
100+%
Integrity
0
43
0
0
0
0
Congestion
0
43
0
0
0
0
SmaCongest
0
43
0
0
0
0
Security
0
43
0
0
0
0
Routing
0
43
0
0
0
0
Congest %:
0.0
Discard:
0
Ineffic %:
0.0
Adapt Rt:0
Quit up Live/rRev/fFwd/bookmrked Bookmrk Unbookmrk ?help | cC N0-n Detail:
Figure 3-15. Error Statistics Screen
The five error categories are each based on one or more port error counters. The
integrity and congestion error values are calculated by using a weighted sum. The
weights for each and the threshold value for each error category can be seen in
the PM Configuration screen (see Figure 3-11). For more details about how the
values for each error category is composed, refer to the QLogic Fabric Manager
User Guide.
Integrity:
Symbol Errors
Link Error Recovery Count
Link Downed Count
Port Receive Errors
Local Link Integrity Errors
Excessive Buffer Overrun Errors (neighbor port)
3-10
IB0054607-01 A
3–iba_top Fabric Performance Monitor
iba_top TUI Screens
Congestion:
Port Transmit Discards (neighbor port)
Port Transmit Congestion (neighbor port)
Port Transmit Wait (neighbor port)
SmaCongestion:
VL15 Dropped Errors
Security:
Port Receive Constraint Errors
Port Transmit Constraint Errors (neighbor port)
Routing:
Port Receive Switch Relay Errors
For each error subgroup five error 'buckets', from 0+% to 100+% in 25%
increments, count the number of ports whose 'error compared to error threshold'
value corresponds to that bucket. This provides an indication of how error rates
compare to their thresholds.
In addition, to aid analysis of congestion, the raw values for maximum
Congestion, Transmit Discards, Inefficiency, and number of Adaptive Routing
adjustments are shown.
The Error Statistics screen accepts input commands which specify parameters to
be used in a group focus query, which will provide a list of ports sorted according
to a specified error criterion. The second line of the Error Statistics screen displays
the group name, and the currently selected focus criterion and number of ports for
a group focus query. The D command causes the group focus query to be
performed and displayed in a Group Focus screen. The following input commands
are accepted in the Error Statistics screen.
C: Select group focus criterion forward (c*) or reverse (C*):
Integrity errors (highest first)
Congestion errors (highest first)
SmaCongestion errors (highest first)
Security errors (highest first)
Routing errors (highest first)
Adaptive routing (highest first)
Nn: Number of entries n in group focus list
D:
IB0054607-01 A
Display detail group focus list
3-11
3–iba_top Fabric Performance Monitor
iba_top TUI Screens
Group Configuration Screen
The Group Configuration screen (Figure 3-16) displays a list of the ports in a
group, including the LID, port number, port GUID and NodeDesc of each. The
second line of the screen displays the group name and the number of ports
returned in the group config query. If more ports exist than will fit on a screen, the
list can be scrolled forward and backward. An index value, shown with each port,
can be used to select the port and show the port's counters in a Port Stats screen.
The Group Configuration screen accepts the following input commands.
S: Scroll forward (s*) or backward (S*) through port list
Pn: Select port index value n
iba_top: Img:Wed Sep 21 12:51:58 2011, Live
Group Config:All
Ix LIDx Port
NumPorts:43
Node GUID 0x
NodeDesc
0 0001
1
00066A009800EC5B admin1 HCA-1
1 0002
1
00066A00070014DC i9k066 L02
2 0002
2
00066A00070014DC i9k066 L02
3 0002
7
00066A00070014DC i9k066 L02
4 0002
8
00066A00070014DC i9k066 L02
5 0002
9
00066A00070014DC i9k066 L02
6 0002
11
00066A00070014DC i9k066 L02
7 0002
13
00066A00070014DC i9k066 L02
8 0002
14
00066A00070014DC i9k066 L02
9 0002
15
00066A00070014DC i9k066 L02
10 0002
16
00066A00070014DC i9k066 L02
11 0002
17
00066A00070014DC i9k066 L02
12 0002
18
00066A00070014DC i9k066 L02
13 0002
19
00066A00070014DC i9k066 L02
14 0002
20
00066A00070014DC i9k066 L02
15 0002
21
00066A00070014DC i9k066 L02
16 0002
22
00066A00070014DC i9k066 L02
17 0002
23
00066A00070014DC i9k066 L02
18 0002
24
00066A00070014DC i9k066 L02
Quit up Live/rRev/fFwd/bookmrked Bookmrk Unbookmrk ?help | sS P0-n:
Figure 3-16. Group Configuration Screen
3-12
IB0054607-01 A
3–iba_top Fabric Performance Monitor
iba_top TUI Screens
Group Focus Screen
The Group Focus screen (Figure 3-17) displays a list of the ports the user has
selected to focus on within a group, including the LID, port number, focus criterion,
port GUID and NodeDesc of each. If the port has a neighbor port, the same
information is displayed for the neighbor. The second line of the screen displays
the group name, the number of ports selected by in the combination of group,
criteria and requested ports, and the number of ports requested in the group focus
query. If more ports exist than will fit on a screen, the list can be scrolled forward
and backward. Like the Bandwidth Statistics and Error Statistics screens which
precede this screen, the focus criterion and number of requested focus ports can
be changed to modify the focus port list. An index value, shown with each port,
can be used to select the port and show the port's counters in a Port Stats screen.
The Group Focus screen accepts the following input commands.
S: Scroll forward (s*) or backward (S*) through port list
C: Select group focus criteria forward (c*) or reverse (C*):
Bandwidth Utilization (highest first)
Packet Utilization (highest first)
Bandwidth Utilization (lowest first)
Integrity errors (highest first)
Congestion errors (highest first)
SmaCongestion errors (highest first)
Security errors (highest first)
Routing errors (highest first)
Adaptive Routing (highest first)
Nn: Number of entries n in group focus list
Pn: Select port index value n
IB0054607-01 A
3-13
3–iba_top Fabric Performance Monitor
iba_top TUI Screens
iba_top: Img:Wed Sep 21 13:01:21 2011, Live
Group Focus:All
GrpNumPorts:43
Ix Integrity LIDx Port
NumPorts:3
Node GUID 0x
Number:3
NodeDesc
0
0 0001
1
00066A009800EC5B admin1 HCA-1
<->
0 0002
1
00066A00070014DC i9k066 L02
1
0 0002
2
00066A00070014DC i9k066 L02
<->
0 0003
1
00066A009800EC51 compute0001 HCA-1
2
0 000C
1
00066A0098007B5E compute0004 HCA-1
<->
0 0007
24
00066A00D9000108 i9k108
Quit up Live/rRev/fFwd/bookmrked Bookmrk Unbookmrk ?help | sS cC N0-n P0-n:
Figure 3-17. Group Focus Screen
Port Statistics Screen
The Port Statistics screen (Figure 3-18) displays a port's counters (performance
and error). Error counters are grouped according to the error category to which
they belong. A trailing asterisk ('*') on the counter name indicates the count will be
used in computing Error Category information for the neighbor port. When the
Port Statistics screen is entered from the Group Focus screen, port neighbor and
link information is available. When the Port Statistics screen is entered from the
Group Configuration screen this information is not available. The second line of
the Port Statistics screen displays the group name, and LID and port number of
the port, as well as link rate and MTU (if available). The third line of the screen
displays the NodeDesc and Node GUID of the port. The fourth line of the screen
displays the NodeDesc, LID and port number of the neighbor port (if available).
The Port Statistics screen accepts the following input command (when neighbor
information is available).
N: Switch between statistics for port and port's neighbor
3-14
IB0054607-01 A
3–iba_top Fabric Performance Monitor
iba_top TUI Screens
iba_top: Img:Mon Apr
Port Stats:All
5 00:01:28 2010, Live
LID:0x28
NodeDesc:i12k71f
PortNum:10
Rate: 40g
MTU:4096
NodeGUID:0x00066A00E300271F
Neighbor:pyro HCA-1
LID:0x7
PortNum:2
Xmit: Data:0
MB (3168
Quads) Pkts:44
Recv: Data:0
MB (1152
Quads) Pkts:16
Integrity:
SmaCongest:
Symbol:0
VL15 Dropped:0
Link Recovery:0
Link Downed:0
Port Rcv:0
Security:
Loc Lnk Integrity:0
Port Rcv Constrain:0
Excess Bfr Overrun*:0
Port Xmt Constrain*:0
Congestion:
Routing:
Port Xmt Discards*:0
Port Rcv Sw Relay:0
Port Xmt Congest*:0
(
Congest Inefficiency*:
0.0%)
0.0%
Port Rcv Rmt Phy:0
Port Adapt Route:0
Port Check Rate:2000 per sec
Quit up Live/rRev/fFwd/bookmrked Bookmrk Unbookmrk ?help | Neighbor
Figure 3-18. Port Statistics Screen
Command Line Options
The following command line options are available for iba_top:
Usage:
iba_top [-v][-q] [-h hca] [-p port] [-i seconds]
or
iba_top --help
Options
--help — produce full help text
-v/--verbose level — verbose output level (additive):
1 — Screen
IB0054607-01 A
3-15
3–iba_top Fabric Performance Monitor
iba_top TUI Screens
4 — STDERR (iba_top)
16 — STDERR PaClient
-q/--quiet — disable progress reports
-h/--hca hca — hca to send by, default is 1st hca
-p/--port port — port to send by, default is 1st active port
-i/--interval seconds — obtain performance stats over interval
seconds
3-16
IB0054607-01 A
4
Real-Time Fabric Monitor
Real-time Fabric Monitor Overview
The Real-time Fabric Monitor (RFM) is a TUI based user interactive application
that provides real-time fabric monitoring support. It must be run on a host
connected to the InfiniBand fabric with FastFabric installed. To use the
Performance view of RFM the QLogic Fabric Manager must be installed.
The RFM obtains all its data in an InfiniBand Trade Association (IBTA)-compliant
manner. Therefore, it will interoperate with both QLogic and 3rd party InfiniBand
components, provided those components are IBTA compliant and implement the
IBTA optional features required by the RFM
The RFM requires that the subnet manager implement all the IBTA SA queries
defined in the standard (such as SM Info records, Link Records, Trace Routes,
Port Records, Node Records, etc). As such, it is recommended that the QLogic
Fabric Manager version 4.0 or later be used. The RFM requires all end nodes to
implement the PMA PortCounters (IBTA mandatory counters). Also any end
nodes which report support of a IBTA device management agent must implement
the IOU Info, IOC Profile and Service Entry queries as outlined in the IBTA 1.1
standard.
The RFM takes advantage of these interfaces to obtain extensive information
about the fabric from the subnet manager and the end nodes. Using this
information, the RFM is able to cross reference it and produce analysis greatly
beyond what any single subnet manager request could provide. As such, it
exceeds the capabilities previously available in tools such as iba_saquery and
fabric_info.
The RFM displays the information in a user-friendly TUI based user interface. The
RFM user interface enables users to analyze the information, using interactive
requests from the user.
The RFM obtains up to date information about the fabric by performing a
discovery of the fabric when it is first started. Each discovery will take a few
seconds (or less, depending on the size of the fabric) to obtain all the fabric data.
The RFM is started by entering the following command at a command prompt.
IB0054607-01 A
4-1
4–Real-Time Fabric Monitor
RFM Screen Layout and User Interaction
Syntax
iba_rfm
[-v] [-h<hca>] [-p<port>] [-i<seconds>] [-I<seconds>]
or
iba_rfm --help
Options
-v — verbose output.This will cause all message types (Errors, Warnings
and Debug) to be output to the screen
-h/--hca<hca> — hca to send by, default is 1st hca
-p/--port<port> — port to send by, default is 1st active port
-i/ --interval<seconds> — topology sweep interval, default is 10
seconds
-I/ --interval<seconds> — PM sweep interval, default is 10 seconds
--help — produce full help text and information on how to use iba_rfm
RFM Screen Layout and User Interaction
The RFM essentially provides a portal into the fabric environment, in order to
enable the administrator to focus on specific facets of the fabric that are of
interest. This is accomplished by a view in the RFM, which defines a specific
category of information pertaining to the fabric environment. Figure 4-19 is the
main menu of the RFM.
4-2
IB0054607-01 A
4–Real-Time Fabric Monitor
RFM Screen Layout and User Interaction
>
MENU:
(A)dmin, (Q)uit
FABRIC: Img: [Tue Mar 16 09:16:38 2010, Live],
SWs: 8,
HCAs: 5,
Links: 28,
GROUP
1 All
TCAs: 0,
SWsPorts: 62,
TCAsPorts: 0
MinMBps
MaxMBps
AvgKPps
MinKPps
MaxKPps
0
0
0
0
0
Snd
0
0
0
0
Integ:Min SmaCong:Min
Congst:Min
Secure:Min
0
0
0
0
Congst:Min
0
0
Secure:Min
Rtng:Min
0
0
Rtng:Min
Snd
0
0
0
0
0
0
Rcv
0
0
0
0
0
0
Integ:Min
4 SWs
HCAsPorts: 2,
0
Rcv
3 TCAs
SMs: 2
AvgMBps
Int
Integ:Min SmaCong:Min
2 HCAs
Nodes: 9,
SmaCong:Min
Congst:Min
Secure:Min
Rtng:Min
Snd
0
0
0
0
0
0
Rcv
0
0
0
0
0
0
Integ:Min SmaCong:Min
VID
VIEWS
1
Infrastructure View
2
Fabric View
Congst:Min
Secure:Min
Rtng:Min
SUBMENU: (L)v, (rR)v, F(wW)d, (b)kmrkd, (B)kmrk, (U)nbkmrk, (G)rp<1-4><u|e>
Figure 4-19. RFM Main Screen
The Views are as follows:

Infrastructure View — enables a user to focus on the infrastructure
specific elements and characteristics of the fabric, such as Switches and
CAs.

Fabric View — enables a user to focus on the networking specific
elements and characteristics of the fabric, such as links and SMs.

Performance View — enables a user to focus on the customer defined
performance specific elements and characteristics of the fabric. A user gains
access to this view using the SUBMENU command: (G)rp<1-4><u|e>.
After selecting a view, the user may drill down into increasing levels of detail
regarding items in that category of view.
IB0054607-01 A
4-3
4–Real-Time Fabric Monitor
Main Screen
The screens of RFM are segmented into fixed sections that provide specific
information about the fabric environment, and the actions taken by the user. The
following subsections describe the different screens and the segments of each
screen starting with the main menu shown in Figure 4-19.
Main Screen
The RFM main screen shown in Figure 4-20 shows a brake down of the segments
in the TUI screen. The following subsections describe each segment of the
screen.
Command-line Segment
>
MENU:
Menu Segment
Fabric Segment
(A)dmin, (Q)uit
Selection Segment
FABRIC: Img: [Tue Mar 16 09:16:38 2010, Live],
SWs: 8,
HCAs: 5,
Links: 28,
GROUP
1 All
Int
TCAs: 0,
SWsPorts: 62,
MaxMBps
AvgKPps
MinKPps
MaxKPps
0
0
0
0
0
0
Congst:Min
Secure:Min
Rtng:Min
Snd
0
0
0
0
0
0
Rcv
0
0
0
0
0
0
Snd
0
Rcv
0
Integ:Min
4 SWs
TCAsPorts: 0
MinMBps
Integ:Min SmaCong:Min
3 TCAs
HCAsPorts: 2,
SMs: 2
AvgMBps
Integ:Min SmaCong:Min
2 HCAs
Nodes: 9,
0
0
SmaCong:Min
Congst:Min
Secure:Min
0
0
0
0
Congst:Min
0
0
Secure:Min
Rtng:Min
0
0
Rtng:Min
Snd
0
0
0
0
0
0
Rcv
0
0
0
0
0
0
Integ:Min SmaCong:Min
VID
VIEWS
1
Infrastructure View
2
Fabric View
Congst:Min
Secure:Min
Rtng:Min
SUBMENU: (L)v, (rR)v, F(wW)d, (b)kmrkd, (B)kmrk, (U)nbkmrk, (G)rp<1-4><u|e>
Submenu Segment
Figure 4-20. Main Menu Segments
4-4
IB0054607-01 A
4–Real-Time Fabric Monitor
Main Screen
Command-line Segment
The Command-line segment of the screen (Figure 4-20) enables the user to
submit commands to control RFM. This is the only section of the screen where
user typing will appear. All other sections of the screen are output only.
Menu Segment
The Menu segment of the screen (Figure 4-20) displays any menu selections
available for the user to utilize for the screen being displayed. The available
selections have a bolded first letter. The other selections are not available.
The menu selections for the main screen are as follows:

(A)dmin — Selects the Administration screen. This controls operation of the
RFM, such as viewing and/or changing the RFM configuration options.

(Q)uit — Exits the RFM.
Fabric Segment
The Fabric segment of the screen (Figure 4-20) displays summary information
about the fabric, specific to the View that the user is currently accessing.
The following are the possible types of information that could be shown:

Image — PV performance image time-stamp and view mode: Live, Hist, or
Bkmk,

SWs — number of switches within the fabric

HCAs — number of Host Channel adapters within the fabric

TCAs — number of Target Channel adapters within the fabric

SMs — number of Subnet Managers within the fabric

Links — number of links within the fabric

SWsPorts — number of ports specific to switches within the fabric

HCAsPorts — number of ports specific to HCAs within the fabric

TCAsPorts — number of ports specific to TCAs within the fabric

Nodes — number of nodes within the fabric

Montrd — number of nodes the RFM is configured to monitor.

Oprtnl — number of nodes that are currently operational.

HlthErrs — number of health related errors within the fabric.

CommErrs — number of communication related errors within the fabric.

Alarms — number of alarms generated within the fabric
IB0054607-01 A
4-5
4–Real-Time Fabric Monitor
Main Screen
.
NOTE:
HlthErrs and Alarms are not presently implemented and will report “Not
Avail”
Selection Segment
The Selection segment of the screen (Figure 4-20) displays the various Views and
Contexts selections available. Each View selection has a numeric View Identifier
(VID) associated with it and each Context selection has a numeric Context
Identifier (CID) associated with it. The user may select an available View/Context
by entering the numeric VID/CID at the Command-line section.
The area of this segment listing groups and their respective bandwidth utilization
and error statistics, is summary information for the Performance View. Unlike the
other View selections, a user gains access to this View using the SUBMENU
command: (G)rp<1-4><u|e>
Submenu Segment
The Submenu segment of the screen (Figure 4-20) displays the following
submenus that can be selected for the main screen.

(L)v — this option selects the live image

(rR)v — this option reverse step/skips thru historical images

F(Ww)d — this option forward step/skips thru historical images

(b)kmrkd — this option selects the current bookmarked image

(B)kmrk — this option bookmarks the current image

(U)nbkmrk — this option unbookmarks the current bookmarked image

(G)rp<1-4><u|e>

4-6

1 — this option selects the All group

2 — this option selects the HCAs group

3 — this option selects the TCAs group

4 — this option selects the SWs group

u — this option selects the utilization statistics category

e — this option selects the error statistics category
(e)rror<1-6>

1 — this option selects the Integrity error category

2 — this option selects the Congestion error category
IB0054607-01 A
4–Real-Time Fabric Monitor
View Screen


3 — this option selects the Sma Congestion error category

4 — this option selects the Security error category

5 — this option selects the Routing error category

6 — this option selects the Adaptive Route error category
(u)til<h|p|l>

h — this option selects the High BW utilization category

p — this option selects the Packet Utilization High category

l — this option selects the Low BW utilization category

(p)rvPg — this sub-menu selection returns the user to the previous screen

Scrl(d)wn — this sub-menu selection lists the next page of information

Scrl(u)p — this sub-menu selection lists the previous page of information

Ls(t)<n> — this sub-menu selection adjusts the size of lists displaying a list
of information

(cC)rit — this sub-menu selection forward/reverse switch between utilization
and error categories

Im(g)Info — this sub-menu selection displays the Image Info

Fr(Z) — this sub-menu selection freezes the topology

Unfr(Z) — this sub-menu selection unfreezes the topology

(+/-) — this sub-menu selection toggles between primary/secondary
sub-menu option selections
View Screen
Each View has a unique Top-level View Summary screen, which enables a user to
focus on elements and characteristics of the fabric specific to the selected View.
The Top-level View Summary screen displays when the Infrastructure View,
Fabric View, or Performance View is selected from the Main screen. An example
of a Top-level Infrastructure View Summary screen is shown in Figure 4-21. The
following subsections describe the segments of the screen.
IB0054607-01 A
4-7
4–Real-Time Fabric Monitor
View Screen
>
MENU:
(H)ome, (F)abricView, (P)erformanceView, (A)dmin, (Q)uit
VIEW:
[I]:H
CONTEXT: Name: Infrastructure View
FABRIC:
Top: [Wed Sep 21 13:10:13 2011, Live],
SWs:
4,
HCAs:
Links:
21,
8,
SWsPorts:
HlthErrs: Not Avail,
TCAs:
38,
CommErrs:
0,
Nodes:
HCAsPorts:
65631,
12,
8,
SMs: 2
TCAsPorts:
0
Alrms: Not Avail
INFRASTRUCTURE VIEW INFORMATION
CID
CATEGORY
TOTAL
MONTRD
OPRTNL HLTH-ERRS COMM-ERRS
ALRMS
1
Switches
4
4
4
0
65600
Not Avail
2
CAs
8
8
8
0
31
Not Avail
3
Chassis
0
0
0
0
0
Not Avail
4
Servers
0
0
0
0
0
Not Avail
5
Cables
0
0
0
0
0
Not Avail
6
Routers
0
0
0
0
0
Not Avail
7
Applications
0
0
0
0
0
Not Avail
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-21. View Screen
Menu Section
This section of the screen displays any menu selections available for the user to
utilize for the screen being displayed (Figure 4-22). The available selections have
a bolded first letter. The other selection are not available.
>
MENU:
(H)ome, (F)abricView, (P)erformanceView, (A)dmin, (Q)uit
VIEW:
[I]:H
Figure 4-22. Menu Section
The menu selections for the main screen are as follows:

4-8
(H)ome - this menu selection returns the user to the Main screen
IB0054607-01 A
4–Real-Time Fabric Monitor
View Screen

(F)abricView - this menu selection takes the user to the Fabric View main
screen, which enables a user to focus on networking specific elements and
characteristics of the fabric. This selection is available when in Infrastructure
and Performance View.

(I)nfrastructureView - this menu selection takes the user to the
Infrastructure View main screen, which enables a user to monitor and query
components in the fabric. This selection is available when in Fabric and
Performance View

(P)erformanceView - this menu selection takes the user to the active
Performance View main screen, which enables a user to focus on
performance specific elements and characteristics of the fabric.

(A)dmin — this menu selection takes the user to the Administration screen,
which enables a user to focus on administrative related operations of the
RFM, such as changing/viewing configuration options.

(Q)uit — Exits the RFM.
View Section
This provides a historical description of the screens traversed by the user to arrive
at the present screen (Figure 4-23). The first letter is the view type followed by the
hierarchy of screens as the user drives deeper into the fabric. Each screen
abbreviation (first letter) is shown in order.
>
MENU:
(H)ome, (F)abricView, (P)erformanceView, (A)dmin, (Q)uit
VIEW:
[I]:H
CONTEXT: Name: Infrastructure View
FABRIC:
Top: [Wed Sep 21 13:10:13 2011, Live],
Figure 4-23. View Section
The following format is used:
[view-abbreviation]:<screen-abbreviation>:<screen-abbreviatio
n>
The presently supported view-abbreviations are Infrastructure View [I], Fabric
View [F], and Performance View [P].
Context Section
This displays the Context of the current screen (Figure 4-24).
IB0054607-01 A
4-9
4–Real-Time Fabric Monitor
View Screen
MENU:
(H)ome, (F)abricView, (P)erformanceView, (A)dmin, (Q)uit
VIEW:
[I]:H
CONTEXT: Name: Infrastructure View
FABRIC:
Top: [Wed Sep 21 13:10:13 2011, Live],
Figure 4-24. Context Section
The Context Section will always contain the name of the context and could include
other information at various levels. The full list of information included in the
Context section are:

Name — name of the current context

Total <element-name> — number of this specific element that exist in
the current context

Alarms — number of alarms generated relevant to the current Context.
Fabric Section
This section includes the same type of information in all screens. Refer to “Fabric
Segment” on page 4-5
Information Section
This section displays general information about a specific context within the fabric.
This section of the screen contains different information based on the view,
category, and element type. The Infrastructure, Fabric, and Performance view
information is described below. The other layouts of this section will be discussed
in their appropriate subsections.
4-10
IB0054607-01 A
4–Real-Time Fabric Monitor
View Screen
HlthErrs: Not Avail,
CommErrs:
65631,
Alrms: Not Avail
INFRASTRUCTURE VIEW INFORMATION
CID
CATEGORY
TOTAL
MONTRD
OPRTNL HLTH-ERRS COMM-ERRS
ALRMS
1
Switches
4
4
4
0
65600
Not Avail
2
CAs
8
8
8
0
31
Not Avail
3
Chassis
0
0
0
0
0
Not Avail
4
Servers
0
0
0
0
0
Not Avail
5
Cables
0
0
0
0
0
Not Avail
6
Routers
0
0
0
0
0
Not Avail
7
Applications
0
0
0
0
0
Not Avail
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-25. Infrastructure and Fabric View Information Section
The following are the definitions of each column in the Information Section for the
Infrastructure and Fabric Views:

CID — Numeric Context Identifier (CID)

CATEGORY — Element type

TOTAL — Total number of the element type in the fabric

MONTRD — Indicates number of the element type that RFM is configured to
monitor. Applies to Switches, CAs, Links, and SMs only

OPRTNL — Indicates number of the element type that is currently
operational. Applies to Switches, CAs, Links, and SMs only

HLTHERRS — Number of health related errors associated with the element
type

COMM-ERRS — Number of communication related errors associated with the
element type.

ALRMS — number of alarms generated that are associated with the element
type.
IB0054607-01 A
4-11
4–Real-Time Fabric Monitor
View Screen
PERF:
GrpName: All, GrpNumPrts: 107, LstNumPrts: 10,
LstMaxNumPrts: 10, NodesFailed: 0, NodesSkipped: 0,
PortsFailed: 0, PortsSkipped: 9
All GROUP ERROR STATS INFORMATION
Int
Max
0+%
25+%
50+%
75+%
100+%
Integrity
0
98
0
0
0
0
Congestion
0
98
0
0
0
0
SmaCongest
0
98
0
0
0
0
Security
0
98
0
0
0
0
Routing
0
98
0
0
0
0
Congest %:
0.0
Discard:
0
Ineffic %:
0.0
Adapt Rt:0
SUBMENU: (+), (p)rvPg, (e)rr<1-6>, Ls(t)<n>, Im(g)Info
Figure 4-26. Performance View Error Conditions Information Section
The following are the definitions of each column in the Information Section for the
Error Conditions category of the Performance View:
4-12

Buckt % — number of ports within given percentage of error conditions

Integrity — link integrity related errors

Congestion — link congestion related errors

SmaCongest — link SMA congestion related errors

Security — link security related errors

Routing — link routing related errors

Adaptive routing — Adaptive routing related errors
IB0054607-01 A
4–Real-Time Fabric Monitor
View Screen
PERF:
GrpName: All, GrpNumPrts: 107, LstNumPrts: 10,
LstMaxNumPrts: 10, NodesFailed: 0, NodesSkipped: 0,
PortsFailed: 0, PortsSkipped: 9
All GROUP BW STATS INFORMATION
Int TotMBps AvgMBps MinMBps MaxMBps TotKPps AvgKPps MinKPps MaxKPps
0
0
0
0
0
0
0
0
Buckt 0+%
10+%
20+%
30+%
40+%
50+%
60+%
70+%
80+%
90+%
98
0
0
0
0
0
0
0
0
0
SUBMENU: (+), (p)rvPg, (u)til<h|l>, Ls(t)<n>, Im(g)Info
Figure 4-27. Performance View Bandwidth Utilization Information Section
The following are the definitions of each column in the Information Section for the
Bandwidth Utilization category of the Performance View:

Int — ports sub-group, which indicates that both ports of a link are in a single
group

Snd — ports sub-group, which indicates that a port of a link sends from
group to outside

Rcv — ports sub-group, which indicates that a port of a link receives by
group from outside

TotMBps — Total of MBps of all selected ports

AvgMBps — average MB per second of all selected ports

MinMBps — minimum MB per second of all selected ports

MaxMBps — maximum MB per second of all selected ports

TotKPps — Total of KPps of all selected ports

AvgKPps — average kilo packets/sec of all selected ports

MinKPps — minimum kilo packets/sec of all selected ports

MaxKPps — maximum kilo packets/sec of all selected ports

Buckt % — number of ports within given percentage of bandwidth utilization
IB0054607-01 A
4-13
4–Real-Time Fabric Monitor
View Screen
Sub-Menu Section
Displays an additional set of minor menu selections that are available. These
menu selections exist to assist the user with miscellaneous operations (i.e.,
maneuvering through the screens).
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-28. Infrastructure and Fabric View Sub-Menu Section
The menu selections supported are the following:

(p)rvPg — return to the previous screen

Scroll (d)own — scroll down to the next screen of information being
listed.

Scroll (u)p — scroll up to the previous screen of information being listed.

Fr(Z) — freezes the current topology sweep

Unfr(z) — unfreezes the current topology sweep
SUBMENU: (+), (p)rvPg, (u)til<h|p|l>, Ls(t)<n>, Im(g)Info
Figure 4-29. Performance View Bandwidth Utilization Sub-Menu Section
The primary menu selections supported for the Performance View are the
following:

(+/-) — toggles between the primary/secondary sub-menu selections

(p)rvPg — return to the previous screen.

Scrl(d)wn — scroll down to the next screen of information being listed.

Scrl(u)p — scroll up to the previous screen of information being listed.

(u)til<h|p|l> — selects the high/low bandwidth or high packet utilization
category
4-14

(e)rr<1-6> — selects an error category: 1) Integrity, 2) Congestion, 3) Sma
Congestion, 4) Security, 5) Routing, 6) Adaptive Route

Ls(t)<n> — adjusts the size of lists displaying a list of information

Im(g)Info — displays the Image Information of the current image
IB0054607-01 A
4–Real-Time Fabric Monitor
Infrastructure View Sub-Screens

(cC)rit — forward/reverse switch between utilization and error categories

(n)eighbor — displays information about the neighbor port of a link
The secondary menu selections supported for the Performance View are the following:

(L)v — selects the live image

(rR)v — reverse step/skips thru historical images

F(wW)d —forward step/skips thru historical images

(b)kmrkd —selects the current bookmarked image

(B)kmrk —bookmarks the current image

(U)nbkmrk —unbookmarks the current bookmarked image
Infrastructure View Sub-Screens
Switch Node Screens
When the user selects the Switches category from the main screen of the
Infrastructure Section View Screen discussed in “Submenu Segment” on page 4-6
above, screens are displayed to enable the user to drill-down to a specific switch
of interest. The order in which these screens are displayed will depend upon the
number of switches found within the fabric.
NOTE:
A Switch Node represents a single Switch Chip within the fabric. Large
switches, such as the QLogic12000 or SilverStorm 9240 are composed of
many Switch Chips and will be displayed as multiple Switch Nodes within
the screens which follow.
Switch Node Selection Screen (500+ Switch Nodes)
If there are 500 or more switch nodes found within the fabric, Figure 4-30 is an
example of the screen that will be displayed first.
IB0054607-01 A
4-15
4–Real-Time Fabric Monitor
Infrastructure View Sub-Screens
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[I]:H:I
CONTEXT: Name: Switches, Total Switches:
500,
Alrms: Not Avail
FABRIC: Top: [Mon Feb 1 16:50:13 2010, Hist], Now: Mon Feb 1 16:54:39 2010
SWs: 500, HCAs: 50, TCAs: 0, Nodes: 527, SMs: 1
Links: 48, SWsPorts: 12000, HCAsPorts: 54, TCAsPorts: 0
HlthErrs: Not Avail, CommErrs: 297, Alrms: Not Avail
CID
SWITCHES
ERRORS
|
CID
SWITCHES
ERRORS
1
1
-
499
0
|
11
5000
-
5499
0
2
500
-
999
0
|
12
5500
-
5999
0
3
1000
-
1499
0
|
13
6000
-
6499
0
4
1500
-
1999
0
|
14
6500
-
6999
0
5
2000
-
2499
0
|
15
7000
-
7499
0
6
2500
-
2999
0
|
16
7500
-
7999
0
7
3000
-
3499
0
|
17
8000
-
8499
0
8
3500
-
3999
0
|
18
8500
-
8999
0
9
4000
-
4499
0
|
19
9000
-
9499
0
10
4500
-
4999
0
|
20
9500
- 10000
0
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-30. Switch Node Selection Screen (500+ Switch Nodes)
4-16
IB0054607-01 A
4–Real-Time Fabric Monitor
Infrastructure View Sub-Screens
Switch Node Selection Screen (26-500 Switch Nodes)
After selecting a block of 500 switch nodes or if there are between 26 to 500
switch nodes found within the fabric Figure 4-31 is an example of the screen that
will be displayed.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[I]:H:I:Sw1-499
CONTEXT: Name: Switches 1 - 499, Total Switches: 500, Alrms: Not Avail
FABRIC: Top: [Mon Feb 1 16:50:13 2010, Hist], Now: Mon Feb 1 16:54:39 2010
SWs: 500, HCAs: 50, TCAs: 0, Nodes: 527, SMs: 1
Links: 48, SWsPorts: 12000, HCAsPorts: 54, TCAsPorts: 0
HlthErrs: Not Avail, CommErrs: 297, Alrms: Not Avail
CID
SWITCHES
ERRORS
|
CID
SWITCHES
ERRORS
1
1
-
25
0
|
11
251
-
275
0
2
26
-
50
0
|
12
276
-
300
0
3
51
-
75
0
|
13
301
-
325
0
4
76
-
100
0
|
14
326
-
350
0
5
101
-
125
0
|
15
351
-
375
0
6
126
-
150
0
|
16
376
-
400
0
7
151
-
175
0
|
17
401
-
425
0
8
176
-
200
0
|
18
426
-
450
0
9
201
-
225
0
|
19
451
-
475
0
10
226
-
250
0
|
20
476
-
500
0
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-31. Switch Node Selection Screen (26-500 Switch Nodes)
IB0054607-01 A
4-17
4–Real-Time Fabric Monitor
Infrastructure View Sub-Screens
Switch Node Selection Screen (1-25 Switch Nodes)
After selecting a block of 25 switch nodes or if there are between 1 to 25 switch
nodes found within the fabric, Figure 4-32 is an example of the screen that will be
displayed.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[I]:H:I:Sw1-499:Sw1-25
CONTEXT: Name: Switches 1 - 25, Total Switches: 14, Alrms: Not Avail
FABRIC: Top: [Mon Feb 1 16:50:13 2010, Hist], Now: Mon Feb 1 16:54:39 2010
SWs: 500, HCAs: 50, TCAs: 0, Nodes: 527, SMs: 1
Links: 48, SWsPorts: 12000, HCAsPorts: 54, TCAsPorts: 0
HlthErrs: Not Avail, CommErrs: 297, Alrms: Not Avail
CID
SWITCH
TYPE PORTS ERRORS
NODE DESC
1
0x00066a000600025a
DDR
24
0
SilverStorm 9080 GUID=0x00066a
2
0x00066a00060002f7
DDR
24
0
SilverStorm 9080 GUID=0x00066a
3
0x00066a0007000311
DDR
24
0
SilverStorm 9080 GUID=0x00066a
4
0x00066a00070003bb
DDR
24
0
SilverStorm 9080 GUID=0x00066a
5
0x00066a0007000df3
DDR
24
0
SilverStorm 9080 GUID=0x00066a
6
0x00066a0007000df6
DDR
24
0
SilverStorm 9080 GUID=0x00066a
7
0x00066a0007000e03
DDR
24
0
SilverStorm 9080 GUID=0x00066a
8
0x00066a0007000e6d
DDR
24
0
SilverStorm 9080 GUID=0x00066a
9
0x00066a00d900045f
DDR
24
0
i9k45f
10
0x00066a00d9000479
DDR
24
0
i9k479
11
0x00066a00dd000075
DDR
24
0
SilverStorm 9080 GUID=0x00066a
12
0x00066a00de00004d
DDR
24
0
SilverStorm 9080 GUID=0x00066a
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-32. Switch Node Selection Screen (1-25 Switch Nodes)
4-18
IB0054607-01 A
4–Real-Time Fabric Monitor
Infrastructure View Sub-Screens
Switch Node Information Selection Screen
Figure 4-33 is an example of the screen for selecting specific information for a
switch node within the fabric.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[I]:H:I:Sw13
CONTEXT: Name: 0x00066a00d90000479, Total Switches:
11
FABRIC: Top: [Tue Feb 2 07:31:25 2010, Hist], Now: Tue Feb 2 07:51:50 2010
SWs: 11, HCAs: 3, TCAs: 0, Nodes: 14, SMs: 1
Links: 48, SWsPorts: 103, HCAsPorts: 4,
HlthErrs: Not Avail,
SWITCH:
CommErrs:
TCAsPorts: 0
297,
Alrms: Not Avail
Montrd: TRUE, Oprtnl: TRUE, HlthErrs: Not Avail,
CommErrs: 285, Alrms: Not Avail, Type: n/a, Ports: 24,
NodeGUID: 0x00066a00d9000479, NodeDesc: i9k479
CID
NODE INFORMATION
1
Device Info
2
Ports Info
3
Chassis Info
4
Module Info
5
Partition Info
6
Health Info
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-33. Switch Node Information Selection Screen
NOTE:
In the present release only the Device Info and Ports Info Categories are
implemented.
IB0054607-01 A
4-19
4–Real-Time Fabric Monitor
Infrastructure View Sub-Screens
Switch Node Device Information Screen
Figure 4-34 is an example of the screen for viewing device specific information for
a switch node within the fabric.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[I]:H:I:Sw1:Sw1Info
CONTEXT: Name: Device Info, Total Switches: 11
FABRIC:
Top: [Tue Feb 2 07:31:25 2010, Hist], Now: Tue Feb 2 07:51:50 2010
SWs: 11, HCAs: 3, TCAs: 0, Nodes: 14, SMs: 1
Links: 48, SWsPorts: 103, HCAsPorts: 4,
HlthErrs: Not Avail,
SWITCH:
CommErrs:
TCAsPorts: 0
297,
Alrms: Not Avail
Montrd: TRUE, Oprtnl: TRUE, HlthErrs: Not Avail,
CommErrs: 285, Alrms: Not Avail, Type: n/a, Ports: 24,
NodeGUID: 0x00066a00d9000479, NodeDesc: i9k479
State:
Active
Number of Ports:
Node GUID:
0x00066a00de00004d
SysImageGuid:
Node Type:
SW
PartitionCap:
Device ID:
0xb924
Revision Number:
0x01a1
Vendor ID:
0x066a
Class Version:
0x0001
Base Version:
0x0001
LID:
0x0004
LinearFDBCap:
49152
LinearFDBTop:
1024
RandomFDBCap:
MCFDBCap:
LIDsPerPort:
0
LifeTime:
DefPort:
0
DefMCPrimPort:
DefMCNotPrimPort:
PortStateChange:
Capability:
255
0
24
0x00066a00da000159
8
51
0
33 ms
255
PartEnfCap:
32
Connected Ports:
16
0xf0: IE OE FI FO
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-34. Switch Node Device Information Screen
4-20
IB0054607-01 A
4–Real-Time Fabric Monitor
Infrastructure View Sub-Screens
Switch Node Port Selection Screen
Figure 4-35 is an example of the screen for selecting a port within a specific
switch node within the fabric.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[I]:H:I:Sw1:Sw1Info
CONTEXT: Name: Ports Info, Total Switches: 11
FABRIC:
Top: [Tue Feb 2 07:31:25 2010, Hist], Now: Tue Feb 2 07:51:50 2010
SWITCH:
NodeGUID: 0x00066a00d9000479, NodeDesc: i9k479
CID
0
CID
PORT
ERRORS
Port-00 (Active)
PORT
0
ERRORS
|
CID
PORT
ERRORS
1
Port-01
0
|
13
Port-13 (Active)
0
2
Port-02
0
|
14
Port-14 (Active)
0
3
Port-03
0
|
15
Port-15 (Active)
0
4
Port-04
0
|
16
Port-16 (Active)
0
5
Port-05
0
|
17
Port-17 (Active)
0
6
Port-06
0
|
18
Port-18 (Active)
0
7
Port-07
0
|
19
Port-19 (Active)
0
8
Port-08
0
|
20
Port-20 (Active)
0
9
Port-09 (Active)
0
|
21
Port-21 (Active)
0
10
Port-10
0
|
22
Port-22 (Active)
0
11
Port-11 (Active)
0
|
23
Port-23 (Active)
0
12
Port-12 (Active)
0
|
24
Port-24 (Active)
0
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-35. Switch Node Port Selection Screen
IB0054607-01 A
4-21
4–Real-Time Fabric Monitor
Infrastructure View Sub-Screens
Switch Node Port Information Selection Screen
Figure 4-36 is an example of the screen for selecting the information to display for
a port within a switch node within the fabric.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[I]:H:I:Sw1:Sw1Info:P0
CONTEXT: Name: Port-00, Total Switches: 11
FABRIC:
Top: [Tue Feb 2 07:31:25 2010, Hist], Now: Tue Feb 2 07:51:50 2010
SWITCH:
Montrd: TRUE,
Oprtnl: TRUE,
HlthErrs: Not Avail,
CommErrs: 285, Alrms: Not Avail, Type: n/a,
Ports: 24,
NodeGUID: 0x00066a00d9000479, NodeDesc: i9k479,
PORT:
CID
PortNo: 0, PortErrs: 0
PORT INFORMATION
1
General Port Info
2
Port Statistics Info
3
Port Attributes Info
4
Port SL2VL Mapping Info
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-36. Switch Node Port Information Selection Screen
NOTE:
In the present release only the General Port Info and Port Statistics Info are
implemented.
4-22
IB0054607-01 A
4–Real-Time Fabric Monitor
Infrastructure View Sub-Screens
Switch Node General Port Information Screen
Figure 4-37 is an example of the screen for viewing general port information,
about a specific port within a switch node within the fabric.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[I]:H:I:Sw1:Sw1Info:P0:P0Info
CONTEXT: Name: General Port Info, Total Switches: 11
FABRIC:
Top: [Tue Feb 2 07:31:25 2010, Hist], Now: Tue Feb 2 07:51:50 2010
SWITCH:
NodeGUID: 0x00066a00d9000479, NodeDesc: i9k479
PortNum:
0 LID: 0x0004 GUID: 0x00066a00de00004d
PortState: Active
PhysState: LinkUp
DownDefault: Polling
LID:
0x0004
LMC: 0
Subnet: 0xfe80000000000000
SMLID:
0x000e
M_KEY:
0x0000000000000000 Lease:
MTU:
SMSL:
0
RespTimeout: 536 ms SubnetTimeout:
0 s
536 ms
Protect: Readonly
Active:
2048 Supported:
2048
VL Stall: 0
LinkWidth: Active:
4x Supported:
1-4x
Enabled:
4x
LinkSpeed: Active:
2.5Gb Supported:
2.5Gb
Enabled:
2.5Gb
8+1 Supported:
8+1
VLs:
Active:
HOQLife: 4096 ns
Capability 0x00100848: VDR SIG SL Trap
Violations: M_Key:
0 P_Key:
ErrorLimits: Overrun:
0 LocalPhys:
P_Key Enforcement: In: Off Out: Off
0 Q_Key:
0
0
DiagCode: 0x0000
FilterRaw: In: Off Out: Off
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-37. Switch Node General Port Information Screen
IB0054607-01 A
4-23
4–Real-Time Fabric Monitor
Infrastructure View Sub-Screens
Switch Node Port Statistics Selection Screen
Figure 4-38 is the screen for viewing port statistic information about a specific port
within a switch node within the fabric.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[I]:H:I:Sw13:Sw13Info:P0:P0Info
CONTEXT: Name: Port Statistics Info, Total Switches: 11
FABRIC:
Top: [Tue Feb 2 07:31:25 2010, Hist], Now: Tue Feb 2 07:51:50 2010
SWITCH:
Montrd: TRUE,
Oprtnl: TRUE,
HlthErrs: Not Avail,
CommErrs: 285, Alrms: Not Avail, Type: DDR, Ports: 24,
NodeGUID: 0x00066a00d9000479, NodeDesc: i9k479,
PORT:
CID
PortNo:
0,
PortErrs:
0
PORT STATISTICS INFORMATION
1
Port Performance Info
2
Port Async Events/Traps Info
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-38. Switch Node Port Statistics Selection Screen
NOTE:
In the present release only the Port Performance Info is implemented.
4-24
IB0054607-01 A
4–Real-Time Fabric Monitor
Infrastructure View Sub-Screens
Switch Node Port Performance Screen
Figure 4-39 is the screen for viewing performance information about a specific
port within a switch node within the fabric.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[I]:H:I:Sw1:Sw1Info:P19:P19Info:P19StatInfo
CONTEXT: Name: Port Performance Info, Total Switches:
4
FABRIC:
Top: [Sat Apr 3 14:37:06 2010, Hist], Now: Sat Apr 3 14:39:56 2010
SWITCH:
Montrd: TRUE,
Oprtnl: TRUE,
CommErrs:
0,
HlthErrs: Not Avail,
Alrms: Not Avail,
Type: n/a,
Ports: 24,
NodeGUID: 0x00066a00060008bc, NodeDesc: SilverStorm 9040 GUID=0x
PMA PORT PERFORMANCE TRANSMIT [PA]
Xmit Data:
Xmit Pkts:
4889 MB (1281863304 Quads)
17803657
PMA PORT PERFORMANCE RECEIVE [PA]
Rcv Data:
9230 MB (2419802026 Quads)
Rcv Pkts:
33608385
PMA PORT ERRORS [PA]
Symbol Errors:
0
Link Error Recovery:
0
Link Downed:
0
Port Rcv Errors:
0
Port Rcv Rmt Phys Err:
0
Port Rcv Sw Relay Err:
Port Xmit Discards:
0
Port Xmit Constraint:
0
Port Rcv Constraint:
0
Local Link Integrity:
0
Exc. Buffer Overrun:
0
VL15 Dropped:
0
48
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-39. Switch Node Port Performance Screen
IB0054607-01 A
4-25
4–Real-Time Fabric Monitor
Infrastructure View Sub-Screens
Channel Adapter (CA) Screens
When the user selects the CAs category from the main screen of the
Infrastructure View (Refer to “Submenu Segment” on page 4-6), screens are
displayed to enable the user to drill-down to a specific channel adapter of interest.
The order in which these screens are displayed will depend upon the number of
channel adapters found within the fabric.
NOTE:
If a given Server or IO Device contains multiple Channel Adapters (CAs),
each will be show separately and can be accessed independently in the
screens which follow.
CA Selection Screen (26+ CAs)
If there are 26 or more CAs found within the fabric, Figure 4-40 is an example of
the screen that will be displayed first.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[I]:H:I
CONTEXT: Name: CAs, Total CAs: 3,
FABRIC:
Alrms: Not Avail
Top: [Tue Feb 2 07:31:25 2010, Hist], Now: Tue Feb 2 07:51:50 2010
SWs: 11, HCAs: 3, TCAs: 0, Nodes: 14, SMs: 1
Links: 48, SWsPorts: 103, HCAsPorts: 4,
TCAsPorts: 0
HlthErrs: Not Avail, CommErrs: 297, Alrms: Not Avail
CID
CAs
ERRORS
|
CID
CAs
ERRORS
1
1
-
25
0
|
11
251
-
275
0
2
26
-
50
0
|
12
276
-
300
0
3
51
-
75
0
|
13
301
-
325
0
4
76
-
100
0
|
14
326
-
350
0
5
101
-
125
0
|
15
351
-
375
0
6
126
-
150
0
|
16
376
-
400
0
7
151
-
175
0
|
17
401
-
425
0
8
176
-
200
0
|
18
426
-
450
0
9
201
-
225
0
|
19
451
-
475
0
10
226
-
250
0
|
20
476
-
500
0
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-40. CA Selection Screen (26+ CAs)
4-26
IB0054607-01 A
4–Real-Time Fabric Monitor
Infrastructure View Sub-Screens
CA Selection Screen (1-25 CAs)
After selecting a block of 25 CAs or if there are between 1 to 25 CAs found within
the fabric, Figure 4-41 is an example of the screen that will be displayed.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[I]:H:I
CONTEXT: Name: CAs 1 - 25, Total CAs: 13, Alrms: Not Avail
FABRIC:
Top: [Tue Feb 2 07:31:25 2010, Hist], Now: Tue Feb 2 07:51:50 2010
SWs: 11, HCAs: 3, TCAs: 0, Nodes: 14, SMs: 1
Links: 48, SWsPorts: 103, HCAsPorts: 4,
TCAsPorts: 0
HlthErrs: Not Avail, CommErrs: 297, Alrms: Not Avail
CID
CA
TYPE PORTS ERRORS
NODE DESC
1
0x0002c9020025a678
DDR
2
0
mindy2 HCA-1
2
0x00066a0098000405
DDR
2
0
endrin
3
0x00066a009800413e
DDR
2
0
cuda
4
0x00066a009800447b
DDR
2
0
duster
5
0x00066a0098006cad
DDR
2
0
InfiniServ
6
0x00066a0098006f74
DDR
2
0
QuickSilver
7
0x00066a0098007b70
DDR
2
0
stewie HCA-1
8
0x00066a10dd000075
DDR
1
0
FVIC in Chassis 0x00066a00da00
9
0x00066a10de00004d
DDR
1
0
EVIC in Chassis 0x00066a00da00
10
0x00066a20dd000075
DDR
1
0
FVIC in Chassis 0x00066a00da00
11
0x00066a20de00004d
DDR
1
0
EVIC in Chassis 0x00066a00da00
12
0x0011750000ff8f4d
DDR
1
0
scooby HCA-1
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-41. CA Selection Screen (1-25 CAs)
IB0054607-01 A
4-27
4–Real-Time Fabric Monitor
Infrastructure View Sub-Screens
CA Information Selection Screen
Figure 4-42 is an example of the screen for selecting the information to display for
a CA within the fabric.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[I]:H:I:Ca1
CONTEXT: Name: 0x00066a0098007b70, Total CAs: 3
FABRIC:
Top: [Tue Feb 2 07:31:25 2010, Hist], Now: Tue Feb 2 07:51:50 2010
SWs: 11, HCAs: 3, TCAs: 0, Nodes: 14, SMs: 1
Links: 48, SWsPorts: 103, HCAsPorts: 4,
TCAsPorts: 0
HlthErrs: Not Avail, CommErrs: 297, Alrms: Not Avail
CA:
Montrd: TRUE,
CommErrs:
87,
Oprtnl: TRUE,
HlthErrs: Not Avail,
Alrms: Not Avail,
Type: n/a,
Ports:
2,
NodeGUID: 0x00066a0098007b70, NodeDesc: stewie HCA-1
CID
NODE INFORMATION
1
Device Info
2
Ports Info
3
Chassis Info
4
Module Info
5
Partition Info
6
Health Info
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-42. CA Information Selection Screen
NOTE:
In the present release only the Device Info and Ports Info Categories are
implemented.
4-28
IB0054607-01 A
4–Real-Time Fabric Monitor
Infrastructure View Sub-Screens
CA Device Information Screen
Figure 4-43 is an example of the screen for viewing device specific information for
a CA within the fabric.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[I]:H:I:Ca1:Ca1Info
CONTEXT: Name: 0x00066a0098007b70, Total CAs: 3
FABRIC:
Top: [Tue Feb 2 07:31:25 2010, Hist], Now: Tue Feb 2 07:51:50 2010
SWs: 11, HCAs: 3, TCAs: 0, Nodes: 14, SMs: 1
Links: 48, SWsPorts: 103, HCAsPorts: 4,
TCAsPorts: 0
HlthErrs: Not Avail, CommErrs: 297, Alrms: Not Avail
CA:
NodeGUID: 0x00066a0098007b70, NodeDesc: stewie HCA-1
State:
Active
Number of Ports:
2
Node GUID:
0x0002c9020025a678
SysImageGuid:
Node Type:
CA
PartitionCap:
Device ID:
0x634a
Revision Number:
0x00a0
Vendor ID:
0x02c9
Class Version:
0x0001
Base Version:
0x0001
Connected Ports:
0x0001
0x0002c9020025a67b
128
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-43. CA Device Information Screen
IB0054607-01 A
4-29
4–Real-Time Fabric Monitor
Infrastructure View Sub-Screens
CA Port Selection Screen
Figure 4-44 is an example of the screen for selecting a port within a specific CA
within the fabric.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[I]:H:I:Ca1:Ca1Info
CONTEXT: Name: Ports Info, Total CAs: 3
FABRIC:
Top: [Tue Feb 2 07:31:25 2010, Hist], Now: Tue Feb 2 07:51:50 2010
CA:
NodeGUID: 0x00066a0098007b70, NodeDesc: stewie HCA-1
CID
0
CID
PORT
ERRORS
Port-00
0
PORT
ERRORS
|
CID
PORT
ERRORS
1
Port-01 (Active)
0
|
13
Port-13
0
2
Port-02
0
|
14
Port-14
0
3
Port-03
0
|
15
Port-15
0
4
Port-04
0
|
16
Port-16
0
5
Port-05
0
|
17
Port-17
0
6
Port-06
0
|
18
Port-18
0
7
Port-07
0
|
19
Port-19
0
8
Port-08
0
|
20
Port-20
0
9
Port-09
0
|
21
Port-21
0
10
Port-10
0
|
22
Port-22
0
11
Port-11
0
|
23
Port-23
0
12
Port-12
0
|
24
Port-24
0
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-44. CA Port Selection Screen
4-30
IB0054607-01 A
4–Real-Time Fabric Monitor
Infrastructure View Sub-Screens
CA Port Information Selection Screen
Figure 4-45 is an example of the screen for selecting the information to be
displayed for a port within a CA within the fabric
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[I]:H:I:Ca1:Ca1Info:P1
CONTEXT: Name:
FABRIC:
Ports Info, Total CAs: 3
Top: [Tue Feb 2 07:31:25 2010, Hist], Now: Tue Feb 2 07:51:50 2010
SWs: 11, HCAs: 3, TCAs: 0, Nodes: 14, SMs: 1
Links: 48, SWsPorts: 103, HCAsPorts: 4,
TCAsPorts: 0
HlthErrs: Not Avail, CommErrs: 297, Alrms: Not Avail
CA:
Montrd: TRUE,
CommErrs:
87,
Oprtnl: TRUE,
HlthErrs: Not Avail,
Alrms: Not Avail,
Type: n/a,
Ports:
2,
NodeGUID: 0x00066a0098007b70, NodeDesc: stewie HCA-1
PORT:
CID
PortNo: 1, PortErrs: 0
PORT INFORMATION
1
General Port Info
2
Port Statistics Info
3
Port Attributes Info
4
Port SL2VL Mapping Info
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-45. CA Port Information Selection Screen
NOTE:
In the present release only the General Port Info and Port Statistics Info are
implemented.
Port Element
The following are shown for a port, within a node:

PortNo — port number of the port.

PortErrs — number of errors observed on the port.
IB0054607-01 A
4-31
4–Real-Time Fabric Monitor
Infrastructure View Sub-Screens
CA Port General Information Screen
Figure 4-46 is an example of the screen for viewing general port information for a
specific port within a CA within the fabric.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[I]:H:I:Ca1:Ca1Info:P1:P1Info
CONTEXT: Name: General Port Info, Total CAs: 3
FABRIC:
Top: [Tue Feb 2 07:31:25 2010, Hist], Now: Tue Feb 2 07:51:50 2010
SWs: 11, HCAs: 3, TCAs: 0, Nodes: 14, SMs: 1
Links: 48, SWsPorts: 103, HCAsPorts: 4,
TCAsPorts: 0
HlthErrs: Not Avail, CommErrs: 297, Alrms: Not Avail
CA:
NodeGUID: 0x00066a0098007b70, NodeDesc: stewie HCA-1
PortNum:
1 LID: 0x0028 GUID: 0x0002c9020025a679
PortState: Active
PhysState: LinkUp
DownDefault: Polling
LID:
0x0028
LMC: 1
Subnet: 0xfe80000000000000
SMLID:
0x000e
M_KEY:
0x0000000000000000 Lease:
MTU:
SMSL:
0
RespTimeout: 268 ms SubnetTimeout:
0 s
Protect: Readonly
Active:
2048 Supported:
2048
VL Stall: 0
LinkWidth: Active:
4x Supported:
1-4x
Enabled:
LinkSpeed: Active:
VLs:
Active:
5.0Gb Supported: 2.5-5Gb
8+1 Supported:
536 ms
8+1
1-4x
Enabled: 2.5-5Gb
HOQLife: 4096 ns
Capability 0x02510868: CR CN VDR CM SIG SL APM Trap
Violations: M_Key:
0 P_Key:
0 Q_Key:
ErrorLimits: Overrun: 15 LocalPhys: 15
P_Key Enforcement: In: Off Out: Off
Neighbor:
0x0002c9020025a678
0
DiagCode: 0x0000
FilterRaw: In: Off Out: Off
1 CA mindy2 HCA-1
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-46. CA Port General Information Screen
4-32
IB0054607-01 A
4–Real-Time Fabric Monitor
Infrastructure View Sub-Screens
CA Port Statistics Selection Screen
Figure 4-47 is an example of the screen for viewing port statistic information for a
specific port within a CA within the fabric
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[I]:H:I:Ca1:Ca1Info:P1:P1Info
CONTEXT: Name: Port Statistics Info, Total CAs: 3
FABRIC:
Top: [Tue Feb 2 07:31:25 2010, Hist], Now: Tue Feb 2 07:51:50 2010
SWs: 11, HCAs: 3, TCAs: 0, Nodes: 14, SMs: 1
Links: 48, SWsPorts: 103, HCAsPorts: 4,
TCAsPorts: 0
HlthErrs: Not Avail, CommErrs: 297, Alrms: Not Avail
CA:
Montrd: TRUE,
CommErrs:
87,
Oprtnl: TRUE,
HlthErrs: Not Avail,
Alrms: Not Avail,
Type: n/a,
Ports:
2,
NodeGUID: 0x00066a0098007b70, NodeDesc: stewie HCA-1
PORT:
CID
PortNo: 1, PortErrs: 0
PORT STATISTICS INFORMATION
1
Port Performance Info
2
Port Async Events/Traps Info
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-47. CA Port Statistics Selection Screen
NOTE:
In the present release only the Port Performance Info is implemented.
IB0054607-01 A
4-33
4–Real-Time Fabric Monitor
Infrastructure View Sub-Screens
CA Port Performance Screen
Figure 4-48 is an example of the screen for viewing performance information for a
specific port within a CA within the fabric.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[I]:H:I:Ca1:Ca1Info:P1:P1Info:P1StatInfo
CONTEXT: Name: Port Performance Info, Total CAs:
8
FABRIC:
Top: [Sat Apr 3 14:44:50 2010, Hist], Now: Sat Apr 3 14:52:50 2010
CA:
Montrd: TRUE,
Oprtnl: TRUE,
CommErrs:
0,
HlthErrs: Not Avail,
Alrms: Not Avail,
Type: n/a,
Ports:
2,
NodeGUID: 0x00066a0098007b5e, NodeDesc: compute0004 HCA-1
PMA PORT PERFORMANCE TRANSMIT [PA]
Xmit Data:
86 MB (22798588 Quads)
Xmit Pkts:
69962
PMA PORT PERFORMANCE RECEIVE [PA]
Rcv Data:
167 MB (44018464 Quads)
Rcv Pkts:
110902
PMA PORT ERRORS [PA]
Symbol Errors:
0
Link Error Recovery:
0
Link Downed:
0
Port Rcv Errors:
0
Port Rcv Rmt Phys Err:
0
Port Rcv Sw Relay Err:
0
Port Xmit Discards:
2
Port Xmit Constraint:
0
Port Rcv Constraint:
0
Local Link Integrity:
0
Exc. Buffer Overrun:
0
VL15 Dropped:
2
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-48. CA Port Performance Screen
4-34
IB0054607-01 A
4–Real-Time Fabric Monitor
Fabric View Sub-Screens
Fabric View Sub-Screens
Link Screens
When a user selects the Links category from the Fabric View main screen
(“Submenu Segment” on page 4-6), the user is able to drill-down to a specific
communication link of interest. The order in which these screens are displayed will
depend upon the number of links found within the fabric.
Link Category Selection Screen
Figure 4-49 is an example of the screen for selecting the category of links to view.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[F]:H:F
CONTEXT: Name: Links
FABRIC:
Top: [Wed Sep 21 14:11:42 2011, Hist], Now: Wed Sep 21 14:13:02 2011
SWs:
4,
Links:
21,
HCAs:
8,
SWsPorts:
HlthErrs: Not Avail,
TCAs:
38,
0,
Nodes:
HCAsPorts:
CommErrs:
65631,
8,
12,
SMs: 2
TCAsPorts:
0
Alrms: Not Avail
FABRIC LINK INFORMATION
CID
CATEGORY
TOTAL
MONTRD
OPRTNL
HLTH-ERRS
COMM-ERRS
ALRMS
21
21
21
0
65631
Not Avail
1
All Links
2
External Links
9
9
9
0
65595
Not Avail
3
Internal Links
12
12
12
0
36
Not Avail
4
Error Links
0
0
0
0
0
Not Avail
5
Slow Links
0
0
0
0
0
Not Avail
6
TopTalker Links
9
9
9
0
65595
Not Avail
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-49. Link Category Selection Screen
IB0054607-01 A
4-35
4–Real-Time Fabric Monitor
Fabric View Sub-Screens
Link Selection Screen (24000+ Links)
If there are 24000 or more links found within the fabric, Figure 4-50 is an example
of the screen that will be displayed first.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[F]:H:F:All
CONTEXT: Name: All Links, Total links: 25007,
Alrms: Not Avail
FABRIC:
Top: [Tue Feb 2 07:31:25 2010, Hist], Now: Tue Feb 2 07:51:50 2010
SWs: 11, HCAs: 3, TCAs: 0, Nodes: 14, SMs: 1
Links: 48, SWsPorts: 103, HCAsPorts: 4, TCAsPorts: 0
HlthErrs: Not Avail, CommErrs: 0, Alrms: Not Avail
CID
LINKS
ERRORS
|
CID
LINKS
ERRORS
1
1
-
23999
0
|
11
240000 - 263999
0
2
24000
-
47999
0
|
12
264000 - 287999
0
3
48000
-
71999
0
|
13
288000 - 311999
0
4
72000
-
95999
0
|
14
312000 - 335999
0
5
96000
- 119999
0
|
15
336000 - 359999
0
6
120000 - 143999
0
|
16
360000 - 383999
0
7
144000 - 167999
0
|
17
384000 - 407999
0
8
168000 - 191999
0
|
18
408000 - 431999
0
9
192000 - 215999
0
|
19
432000 - 455999
0
10
216000 - 239999
0
|
20
456000 - 480000
0
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-50. Link Selection Screen (24000+ Links)
4-36
IB0054607-01 A
4–Real-Time Fabric Monitor
Fabric View Sub-Screens
Link Selection Screen (1200-24000 Links)
After selecting a block of 24000 Links or if there are between 1200 to 24000 Links
found within the fabric, Figure 4-51 is an example of the screen that will be
displayed.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[F]:H:F:All:Link1-23999
CONTEXT: Name: Links 1 - 23999, Total links: 25007,
Alrms: Not Avail
FABRIC:
Top: [Tue Feb 2 07:31:25 2010, Hist], Now: Tue Feb 2 07:51:50 2010
SWs: 11, HCAs: 3, TCAs: 0, Nodes: 14, SMs: 1
Links: 48, SWsPorts: 103, HCAsPorts: 4, TCAsPorts: 0
HlthErrs: Not Avail, CommErrs: 0, Alrms: Not Avail
CID
LINKS
ERRORS
|
CID
LINKS
ERRORS
1
1
-
1200
0
|
11
12001
-
13200
0
2
1201
-
2400
0
|
12
13201
-
14400
0
3
2401
-
3600
0
|
13
14401
-
15600
0
4
3601
-
4800
0
|
14
15601
-
16800
0
5
4801
-
6000
0
|
15
16801
-
18000
0
6
6001
-
7200
0
|
16
18001
-
19200
0
7
7201
-
8400
0
|
17
19201
-
20400
0
8
8401
-
9600
0
|
18
20401
-
21600
0
9
9601
-
10800
0
|
19
21601
-
22800
0
10
10801
-
12000
0
|
20
22801
-
24000
0
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-51. Link Selection Screen (1200-24000 Links)
IB0054607-01 A
4-37
4–Real-Time Fabric Monitor
Fabric View Sub-Screens
Link Selection Screen (60-1200 Links)
After selecting a block of 1200 Links or if there are between 60 to 1200 Links
found within the fabric, Figure 4-52 is an example of the screen that will be
displayed.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[F]:H:F:All:Link1-23999:Link1-1200
CONTEXT: Name: Links 1 - 1200, Total links: 25007,
Alrms: Not Avail
FABRIC:
Top: [Tue Feb 2 07:31:25 2010, Hist], Now: Tue Feb 2 07:51:50 2010
SWs: 11, HCAs: 3, TCAs: 0, Nodes: 14, SMs: 1
Links: 48, SWsPorts: 103, HCAsPorts: 4, TCAsPorts: 0
HlthErrs: Not Avail, CommErrs: 0, Alrms: Not Avail
CID
LINKS
ERRORS
|
CID
LINKS
ERRORS
1
1
-
60
0
|
11
601
-
660
0
2
61
-
120
0
|
12
661
-
720
0
3
121
-
180
0
|
13
721
-
780
0
4
181
-
240
0
|
14
781
-
840
0
5
241
-
300
0
|
15
841
-
900
0
6
301
-
360
0
|
16
901
-
960
0
7
361
-
420
0
|
17
961
-
1020
0
8
421
-
480
0
|
18
1021
-
1080
0
9
481
-
540
0
|
19
1081
-
1140
0
10
541
-
600
0
|
20
1141
-
1200
0
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-52. Link Selection Screen (60-1200 Links)
4-38
IB0054607-01 A
4–Real-Time Fabric Monitor
Fabric View Sub-Screens
Link Selection Screen (1-60 Links)
After selecting a block of 60 Links or if there are between 1 to 60 Links found
within the fabric, Figure 4-53 is an example of the screen that will be displayed.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[F]:H:F:All:Link1-23999:Link1-1200:Link1-60
CONTEXT: Name: Links 1 - 60, Total links: 25007,
Alrms: Not Avail
FABRIC:
Top: [Tue Feb 2 07:31:25 2010, Hist], Now: Tue Feb 2 07:51:50 2010
SWs: 11, HCAs: 3, TCAs: 0, Nodes: 14, SMs: 1
Links: 48, SWsPorts: 103, HCAsPorts: 4, TCAsPorts: 0
HlthErrs: Not Avail, CommErrs: 0, Alrms: Not Avail
CID
1
2
3
4
5
6
RATE
20g
20g
20g
20g
20g
20g
LINK NODE
TYPE PORT ERRORS
NODE DESC
0x0002c9020025a678
CA
1
0
mindy2 HCA-1
0x00066a0007000e6d
SW
4
0
SilverStorm 9080 GUID=
0x00066a000600025a
SW
1
0
SilverStorm 9080 GUID=
0x00066a0007000e6d
SW
19
0
SilverStorm 9080 GUID=
0x00066a000600025a
SW
2
0
SilverStorm 9080 GUID=
0x00066a0007000e6d
SW
20
0
SilverStorm 9080 GUID=
0x00066a000600025a
SW
3
0
SilverStorm 9080 GUID=
0x00066a0007000e6d
SW
21
0
SilverStorm 9080 GUID=
0x00066a000600025a
SW
4
0
SilverStorm 9080 GUID=
0x00066a00de00004d
SW
22
0
SilverStorm 9080 GUID=
0x00066a000600025a
SW
5
0
SilverStorm 9080 GUID=
0x00066a00de00004d
SW
23
0
SilverStorm 9080 GUID=
SUBMENU: (p)rvPg, Scrl(d)wn, Fr(Z), Unfr(z)
Figure 4-53. Link Selection Screen (1-60 Links)
IB0054607-01 A
4-39
4–Real-Time Fabric Monitor
Fabric View Sub-Screens
Link Information Selection Screen
Figure 4-54 is an example of the screen for selecting the information to display or
a link within the fabric.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[F]:H:F:All:Link1-23999:Link1-1200:Link1-60:Link1
CONTEXT: Name: Link1, Total links:
FABRIC:
25007
Top: [Tue Feb 2 07:31:25 2010, Hist], Now: Tue Feb 2 07:51:50 2010
SWs: 11, HCAs: 3, TCAs: 0, Nodes: 14, SMs: 1
Links: 48, SWsPorts: 103, HCAsPorts: 4, TCAsPorts: 0
HlthErrs: Not Avail, CommErrs: 0, Alrms: Not Avail
LINK:
LnkNo:
1, LnkTyp: External,
EndNode1: 0x00066a0098007b70, PortNo: 1, NodeDesc: stewie HCA-1
EndNode2: 0x00066a00d9000479, PortNo: 18, NodeDesc: i9k479
LinkWidth: Active:
LnkSpd: Active:
CID
LINK INFORMATION
4x, Supported:
1-4x, Enabled:
1-4x,
5.0Gb, Supported: 2.5-5Gb, Enabled: 2.5-5Gb
ERRORS
1
End-Node 1 Info
0
2
End-Node 2 Info
0
3
End-Port 1 Info
0
4
End-Port 2 Info
0
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-54. Link Information Selection Screen (60-1200 Links)
NOTE:
A link consists of two connected ports between two end nodes (Switch, CA,
or router). In this screen the user may select to drill down into either port or
either end node.
4-40
IB0054607-01 A
4–Real-Time Fabric Monitor
Fabric View Sub-Screens
Link End-Node Information Selection Screen
Figure 4-55 is an example of the screen for selecting the information to display for
an end-node of a specific link within the fabric
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[F]:H:F:All:Link1:Node1Info
CONTEXT: Name: End-Node 1 Info, Total links:
FABRIC:
111
Top: [Tue Feb 2 07:31:25 2010, Hist], Now: Tue Feb 2 07:51:50 2010
SWs: 11, HCAs: 3, TCAs: 0, Nodes: 14, SMs: 1
Links: 48, SWsPorts: 103, HCAsPorts: 4, TCAsPorts: 0
HlthErrs: Not Avail, CommErrs: 0, Alrms: Not Avail
LINK:
LnkNo: 1, LnkTyp: External,
EndNode1: 0x00066a0098007b70, PortNo: 1, NodeDesc: stewie HCA-1
EndNode2: 0x00066a00d9000479, PortNo: 18, NodeDesc: i9k479
LinkWidth: Active:
LnkSpd: Active:
CID
4x, Supported:
1-4x, Enabled:
1-4x,
5.0Gb, Supported: 2.5-5Gb, Enabled: 2.5-5Gb
NODE INFORMATION
1
Device Info
2
Ports Info
3
Chassis Info
4
Module Info
5
Partition Info
6
Health Info
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-55. Link End-Node Information Selection Screen
NOTE:
In the present release only the Device Info and Ports Info Categories are
implemented.
Selections on this screen will display Device Information screen described in “CA
Device Information Screen” on page 4-29 or Port Information Selection screen
described in “CA Port Information Selection Screen” on page 4-31.
IB0054607-01 A
4-41
4–Real-Time Fabric Monitor
Fabric View Sub-Screens
Link Element
A link is an established connection between two ports (cable or
backplane/internal). The following are shown in Figure 4-55:
4-42

LnkNo — numeric identifier associated with the link.

LnkTyp — type of link: Internal or External.

EndNode1/2 — These two fields lists the Node GUIDs of the two nodes
connected through the link.

NodeDesc — lists the Node Description of the node.

PortNo — lists the port number on the node which is used by the link.

LinkWidth — lists the Active, Supported, and Enabled widths of the link.

LnkSpd — lists the Active, Supported, and Enabled speeds of the link.
IB0054607-01 A
4–Real-Time Fabric Monitor
Fabric View Sub-Screens
Link End-Port Information Selection Screen
Figure 4-56 is an example of the screen for selecting the information to display for
an end-port of a specific link within the fabric
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[F]:H:F:All:Link1:Node1Info:Ca1Info:P1
CONTEXT: Name: End-Port 1 Info, Total links:
FABRIC:
111
Top: [Tue Feb 2 07:31:25 2010, Hist], Now: Tue Feb 2 07:51:50 2010
SWs: 11, HCAs: 3, TCAs: 0, Nodes: 14, SMs: 1
Links: 48, SWsPorts: 103, HCAsPorts: 4, TCAsPorts: 0
HlthErrs: Not Avail, CommErrs: 0, Alrms: Not Avail
CA:
Montrd: TRUE,
CommErrs: 8,
Oprtnl: TRUE,
HlthErrs: Not Avail,
Alrms: Not Avail,
Type: n/a,
Ports:
2,
NodeGUID: 0x00066a0098007b70, NodeDesc: stewie HCA-1
LINK:
LnkNo: 1, LnkTyp: External,
EndNode1: 0x00066a0098007b70, PortNo: 1, NodeDesc: stewie HCA-1
EndNode2: 0x00066a00d9000479, PortNo: 18, NodeDesc: i9k479
LinkWidth: Active:
LnkSpd: Active:
PORT:
CID
PortNo:
1,
4x, Supported:
1-4x, Enabled:
1-4x,
5.0Gb, Supported: 2.5-5Gb, Enabled: 2.5-5Gb
PortErrs:
0
PORT INFORMATION
1
General Port Info
2
Port Statistics Info
3
Port Attributes Info
4
Port SL2VL Mapping Info
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-56. Link End-Port Information Selection Screen
NOTE:
In the present release only the General Port Info and Port Statistics Info are
implemented.
Selections on this screen will display General Port Information described in “CA
Port General Information Screen” on page 4-32 or Port Statistics Selection screen
described in “CA Port Statistics Selection Screen” on page 4-33
IB0054607-01 A
4-43
4–Real-Time Fabric Monitor
Fabric View Sub-Screens
Slow Link Selection Screen
Figure 4-57 is an example of the screen for selecting which types of slow link
performance to analyze.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[F]:H:F:Slow
CONTEXT: Name: Slow Links, Total links:
FABRIC:
25007
Top: [Tue Feb 2 07:31:25 2010, Hist], Now: Tue Feb 2 07:51:50 2010
SWs: 11, HCAs: 3, TCAs: 0, Nodes: 14, SMs: 1
Links: 48, SWsPorts: 103, HCAsPorts: 4, TCAsPorts: 0
HlthErrs: Not Avail, CommErrs: 0, Alrms: Not Avail
SLOW LINK INFORMATION
CID
TOTAL
ERRS
1
Links running slower than expected speed
CATEGORY
0
0
2
Links running slower than supported speed
0
0
3
Links running slower than faster port
5
0
4
Links configured slower than supported speed
0
0
5
Links connected with mismatched supported speeds
5
0
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-57. Slow Link Selection Screen
Selections on this screen will display Link screens as shown in “Link Selection
Screen (24000+ Links)” on page 4-36, “Link Selection Screen (1200-24000
Links)” on page 4-37, “Link Selection Screen (60-1200 Links)” on page 4-38, or
“Link Selection Screen (1-60 Links)” on page 4-39
4-44
IB0054607-01 A
4–Real-Time Fabric Monitor
Fabric View Sub-Screens
SM Screens
SM Selection Screen
Figure 4-58 is an example of the initial screen for selecting a SM within the fabric.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[F]:H:F
CONTEXT: Name: SMs
FABRIC:
Top: [Tue Feb 2 07:31:25 2010, Hist], Now: Tue Feb 2 07:51:50 2010
SWs: 11, HCAs: 3, TCAs: 0, Nodes: 14, SMs: 1
Links: 48, SWsPorts: 103, HCAsPorts: 4, TCAsPorts: 0
HlthErrs: Not Avail, CommErrs: 0, Alrms: Not Avail
CID
TYPE
NODE GUID
NAME
1
Master
0x00066a000600025a
SilverStorm 9080 GUID=
2
Standby
0x00066a00060002f7
SilverStorm 9080 GUID=
3
Standby
0x0011750000ffd66c
strawberry HCA-1
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-58. SM Selection Screen
IB0054607-01 A
4-45
4–Real-Time Fabric Monitor
Fabric View Sub-Screens
SM Detailed Information Screen
Figure 4-59 is an example of the screen showing detailed information about an
SM within the fabric.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[F]:H:F:Sm1
CONTEXT: Name: Sm1
FABRIC:
Top: [Tue Feb 2 07:31:25 2010, Hist], Now: Tue Feb 2 07:51:50 2010
SWs: 11, HCAs: 3, TCAs: 0, Nodes: 14, SMs: 1
Links: 48, SWsPorts: 103, HCAsPorts: 4, TCAsPorts: 0
HlthErrs: Not Avail, CommErrs: 0, Alrms: Not Avail
State: Master
Name: SilverStorm 9080 GUID=0x00066a
NodeGUID: 0x00066a000600025a
Type: SW
PortNum:
0
LID: 0x000e
PortGUID: 0x00066a000600025a
SM_Key: 0x0000000000000000 Priority:
0 ActCount: 0x0025e41d
SUBMENU: (p)rvPg, Fr(Z), Unfr(z)
Figure 4-59. SM Detailed Information Screen
4-46
IB0054607-01 A
4–Real-Time Fabric Monitor
Performance View Sub-Screens
Performance View Sub-Screens
Bandwidth Utilization Screens
When a user selects the (G)rp<1-4><u> sub-menu selection from the Main
screen (“Submenu Segment” on page 4-6), the user is able to drill-down to the
utilization statistics for a specific group of ports of interest.
Switch Group Bandwidth Utilization Selection Screen
Figure 4-60 is an example of the screen for selecting the utilization category of the
Switch group to view.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (A)dmin, (Q)uit
VIEW:
[P]:H
CONTEXT: Name: Low BW Stats
FABRIC:
PERF:
Top: [Tue Feb 2 07:31:25 2010, Hist], Now: Tue Feb 2 07:51:50 2010
GrpName: SWs, GrpNumPrts: 94, LstNumPrts: 10,
LstMaxNumPrts: 10, NodesFailed: 0, NodesSkipped: 0,
PortsFailed: 0, PortsSkipped: 9
LOW BW STATS PORTS LISTING
CID
1
2
3
4
5
PORT NO
LID
VAL
Node GUID
NODE
13
0x0003
0.0
0x00066a00d9000479
i9k479
9
0x0400
0.0
0x00066a0007000e6d
SilverStorm 9080 GUID=
18
0x0003
0.0
0x00066a00d9000479
i9k479
1
0x0100
0.0
0x00066a0098007b70
stewie HCA-1
20
0x0003
0.0
0x00066a00d9000479
i9k479
1
0x0f00
0.0
0x0002c9020025a678
mindy2 HCA-1
7
0x0004
0.0
0x00066a0007000e6d
SilverStorm 9080 GUID=
2
0x0200
0.0
0x00066a0098007b70
stewie HCA-1
13
0x0004
0.0
0x00066a0007000e6d
SilverStorm 9080 GUID=
10
0x0500
0.0
0x00066a100600025a
SilverStorm 9080 GUID=
SUBMENU: (+), (p)rvPg, (u)til<h|l>, Ls(t)<n>, Im(g)Cfg
Figure 4-60. Switch Group Bandwidth Utilization Selection Screen
IB0054607-01 A
4-47
4–Real-Time Fabric Monitor
Performance View Sub-Screens
Low Bandwidth Utilization Selection Screen
After selecting the (u)til<l> sub-menu selection, for ports associated with links
with a low bandwidth utilization, a list of links is displayed to the user. Figure 4-61
is an example of the screen that will be displayed.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[P]:H:LowStats
CONTEXT: Name: Performance Group BW View
FABRIC:
Top: [Tue Feb 2 07:31:25 2010, Hist], Now: Tue Feb 2 07:51:50 2010
PERF:
GrpName: SWs, GrpNumPrts: 94, LstNumPrts: 10, LstMaxNumPrts: 10
SWs GROUP BW STATS INFORMATION
Int TotMBps AvgMBps MinMBps MaxMBps TotKPps AvgKPps MinKPps MaxKPps
0
Buckt
0
0+% 10+%
90
0
0
20+%
0
0
30+%
0
40+%
0
0
50+%
0
0
60+%
0
0
0
70+%
80+%
90+%
0
0
0
Snd TotMBps AvgMBps MinMBps MaxMBps TotKPps AvgKPps MinKPps MaxKPps
0
Buckt
0
0+% 10+%
4
0
0
20+%
0
0
30+%
0
40+%
0
0
50+%
0
0
60+%
0
0
70+%
0
80+%
0
0
90+%
0
Rcv TotMBps AvgMBps MinMBps MaxMBps TotKPps AvgKPps MinKPps MaxKPps
0
Buckt
0
0+% 10+%
4
0
0
20+%
0
0
30+%
0
0
40+%
0
50+%
0
0
60+%
0
0
70+%
0
80+%
0
0
90+%
0
SUBMENU: (+), (p)rvPg, (u)til<h|l>, Ls(t)<n>, Im(g)Cfg
Figure 4-61. Low Bandwidth Utilization Selection Screen
4-48
IB0054607-01 A
4–Real-Time Fabric Monitor
Performance View Sub-Screens
Error Condition Screens
When a user selects the (G)rp<1-4><e> sub-menu selection from the Main
screen (“Submenu Segment” on page 4-6), the user is able to drill-down to the
error condition statistics for a specific group of ports of interest.
Switch Group Error Condition Selection Screen
Figure 4-62 is an example of the screen for selecting the error category of the
Switch group to view.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (A)dmin, (Q)uit
VIEW:
[P]:H
CONTEXT: Name: Performance Group Errors View
FABRIC:
Img: [Sat Apr 3 14:29:23 2010, Hist], Now: Sat Apr 3 15:00:25 2010
PERF:
GrpName:
SWs, GrpNumPrts:
LstMaxNumPrts:
35, LstNumPrts:
10,
10
SWs GROUP ERROR STATS INFORMATION
Int
Max
0+%
25+%
50+%
75+%
100+%
Integrity
0
27
0
0
0
0
Congestion
0
27
0
0
0
0
SmaCongest
0
27
0
0
0
0
Security
0
27
0
0
0
0
Routing
0
27
0
0
0
0
Congest %:
0.0
Ext
Discard:
0
Ineffic %:
0.0
Adapt Rt:0
Max
0+%
25+%
50+%
75+%
100+%
Integrity
0
8
0
0
0
0
Congestion
0
8
0
0
0
0
SmaCongest
0
8
0
0
0
0
Security
0
8
0
0
0
0
Routing
0
8
0
0
0
0
Congest %:
0.0
Discard:
0
Ineffic %:
0.0
Adapt Rt:0
SUBMENU: (+), (p)rvPg, (e)rr<1-6>, Ls(t)<n>, Im(g)Info
Figure 4-62. Switch Group Error Condition Selection Screen
IB0054607-01 A
4-49
4–Real-Time Fabric Monitor
Performance View Sub-Screens
Integrity Error Selection Screen
After selecting the (e)rr<1> sub-menu selection, for ports associated with links
that have integrity related errors, a list of links is displayed to the user. Figure 4-63
is an example of the screen that will be displayed.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (A)dmin, (Q)uit
VIEW:
[P]:H:IntegErr
CONTEXT: Name: Integrity Error
FABRIC:
Top: [Tue Feb 2 07:31:25 2010, Hist], Now: Tue Feb 2 07:51:50 2010
PERF:
GrpName: SWs, GrpNumPrts: 94, LstNumPrts: 10,
LstMaxNumPrts: 10, NodesFailed: 0, NodesSkipped: 0,
PortsFailed: 0, PortsSkipped: 9
INTEGRITY ERROR PORTS LISTING
CID
1
2
3
4
5
PORT NO
LID
VAL
Node GUID
NODE DECS
13
0x0003
0.0
0x00066a00d9000479
i9k479
9
0x0400
0.0
0x00066a0007000e6d
SilverStorm 9080 GUID=
18
0x0003
0.0
0x00066a00d9000479
i9k479
1
0x0100
0.0
0x00066a0098007b70
stewie HCA-1
20
0x0003
0.0
0x00066a00d9000479
i9k479
1
0x0f00
0.0
0x0002c9020025a678
mindy2 HCA-1
7
0x0004
0.0
0x00066a0007000e6d
SilverStorm 9080 GUID=
2
0x0200
0.0
0x00066a0098007b70
stewie HCA-1
13
0x0004
0.0
0x00066a0007000e6d
SilverStorm 9080 GUID=
10
0x0500
0.0
0x00066a100600025a
SilverStorm 9080 GUID=
SUBMENU: (+), (p)rvPg, (u)til<h|l>, Ls(t)<n>, Im(g)Cfg
Figure 4-63. Integrity Error Selection Screen
4-50
IB0054607-01 A
4–Real-Time Fabric Monitor
Performance View Sub-Screens
Link Selection Screen
Figure 4-64 is an example of the screen for selecting the information to display for
a specific link within the fabric.
MENU:
(H)ome, (I)nfrastructureView, (F)abricView, (P)erformanceView,
(T)opContext, (A)dmin, (Q)uit
VIEW:
[P]:H:IntegErr
CONTEXT: Name: Integrity Error
FABRIC:
Img: [Sat Apr 3 14:29:23 2010, Hist], Now: Sat Apr 3 15:06:05 2010
PERF:
GrpName:
SWs, GrpNumPrts:
LstMaxNumPrts:
PortsFailed:
35, LstNumPrts:
10, NodesFailed:
0, PortsSkipped:
10,
0, NodesSkipped:
0,
3
INTEGRITY ERROR PORTS LISTING
CID
PORT NO
LID
VAL
1
0x0002
0
0x00066a00070014dc
SilverStorm 9040 GUID=
1
0x0001
0
0x00066a009800ec5b
admin1 HCA-1
2
2
0x0002
0
0x00066a00070014dc
SilverStorm 9040 GUID=
1
0x0003
0
0x00066a009800ec51
compute0001 HCA-1
3
7
0x0002
0
0x00066a00070014dc
SilverStorm 9040 GUID=
1
0x0004
0
0x00066a009800ec5a
compute0006 HCA-1
4
8
0x0002
0
0x00066a00070014dc
SilverStorm 9040 GUID=
1
0x0006
0
0x00066a009800ec47
compute0007 HCA-1
5
9
0x0002
0
0x00066a00070014dc
SilverStorm 9040 GUID=
1
0x0008
0
0x00066a009800ec73
compute0002 HCA-1
1
Node GUID
NODE
SUBMENU: (+), (p)rvPg, Scrl(d)wn, (cC)rit, Im(g)Info
Figure 4-64. Link Selection Screen
IB0054607-01 A
4-51
4–Real-Time Fabric Monitor
Admin Menu Screens
Admin Menu Screens
Main Screen
Figure 4-65 is the main screen of the Admin menu selection, which enables a user
to perform administration related operations with RFM
MENU:
(H)ome, (Q)uit
VIEW:
[A]:H
CONTEXT: Name: Admin
CID
ADMIN OPERATIONS
1
Performance View Configuration
2
Infrastructure View Configuration
3
Fabric View Configuration
4
General Configuration
SUBMENU: (p)rvPg
Figure 4-65. Admin Menu Main Screen
NOTE:
In the present release only the Performance View Configuration category is
implemented.
4-52
IB0054607-01 A
4–Real-Time Fabric Monitor
Admin Menu Screens
Fabric Discovery Screen
Figure 4-66 is the screen for viewing results from the execution of a fabric
discovery operation.
# iba_rfm
Please wait, while the fabric is being discovered...
Beginning Fabric Discovery...
Getting general information about all PM groups
Getting All Node Records...
Done Getting All Node Records
Done Getting All Link Records
Done Getting All SM Info Records
Getting All PA Port Counters...
Done Getting All PA Port Counters
Completed Topology related discovery: OK
Completed enhanced PM related discovery: OK
FABRIC DISCOVERY SUMMARY
Total Nodes:
48
Total Switch Nodes:
20
Total CA Nodes:
28
Total Router Nodes:
0
Total SMs:
1
Completed Topology related discovery: OK
Completed enhanced PM related discovery: OK
[Press the Enter key to continue]
Figure 4-66. Fabric Discovery Screen
IB0054607-01 A
4-53
4–Real-Time Fabric Monitor
Admin Menu Screens
4-54
IB0054607-01 A
5
Configuration of IPoIB
Name Mapping
The FastFabric tools support the concept of a management network and an IPoIB
network. For some clusters the management network will be a low speed network
such as 10/100 Ethernet. For other clusters IPoIB may serve double duty as the
host management network.
NOTE:
When using IPoIB as the management network, the initial installation of
InfiniBand software cannot be done using FastFabric.
The various FastFabric tools will translate from host names provided to and from
IPoIB names as needed. This permits the host names given to be either
management network or IPoIB network names. The default configuration file
assumes that IPoIB host names are formed by adding a -ib suffix to the
management network name. If a different suffix is desired, FF_IPOIB_SUFFIX
can be changed. If IPoIB is also being used as the management network,
FF_IPOIB_SUFFIX can be set to an empty string "".
The translation is driven by the following functions within fastfabric.conf:
ff_host_basename - given a management network or IPoIB hostname,
translate to management network name, should match hostname -s
ff_host_basename_to_ipoib - given a management network name, translate
to IPoIB hostname
More complex mappings can be specified by implementing alternate algorithms
for these functions.
NOTE:
When managing a cluster where the IPoIB settings on the compute nodes
are incompatible with the InfiniBand Management node (for example when a
4K MTU is used on the compute nodes and a 2K MTU is used on the
InfiniBand Management Nodes), it is recommended not to run IPoIB on the
InfiniBand management nodes.
IB0054607-01 A
5-1
5–Configuration of IPoIB Name Mapping
5-2
IB0054607-01 A
A
FastFabric Configuration
Files
Table A-2 list the configuration files that are used by FastFabric. The description in
the table also list the following sections that have detailed descriptions of each file.
For a given release consult the files with -sample at the end of the file name for a
sample file with the defaults of the given release.
Table A-2. FastFabric Configuration Files
Configuration File
IB0054607-01 A
Description
/etc/sysconfig/fastfabric.conf
Overall configuration file. Refer to FastFabric Configuration File.
/etc/sysconfig/iba/iba_mon.conf
Error thresholds. Refer to Port Statistics Thresholds Configuration File.
/etc/sysconfig/iba/iba_mon.si.conf
Error thresholds related to Signal Integrity. Refer to
Signal Integrity Thresholds Configuration File.
/etc/sysconfig/iba/allhosts
List of all hosts managed by FastFabric including
the localhost. Refer to Host List Files.
/etc/sysconfig/iba/hosts
List of all hosts managed by FastFabric except the
localhost. Refer to Host List Files.
/etc/sysconfig/iba/chassis
List of all chassis managed by FastFabric. Refer to
Chassis List Files.
/etc/sysconfig/iba/esm_chassis
List of all chassis running an embedded SM which
are to be monitored using esm_analysis. Refer to
Chassis List Files.
/etc/sysconfig/iba/ibnodes
List of all externally managed switches managed
by FastFabric. Refer to Externally Managed Switch
List File.
/etc/sysconfig/iba/ports
List of local HCA ports (for example subnets) to be
used for fabric health analysis. Refer to Port List
File.
A-1
A–FastFabric Configuration Files
FastFabric Configuration File
Table A-2. FastFabric Configuration Files
Configuration File
/etc/sysconfig/iba/topology.0:0.xml
Description
Fabric topology input file used by iba_reports and
fabric health tools. Refer to Fabric Topology Input
File.
FastFabric Configuration File
The FastFabric tools support a configuration file
/etc/sysconfig/fastfabric.conf. This file can be used to provide default
settings for most of the FastFabric command line options. The configuration file is
a bash shell script which will be included by each tool. As such the file should be
implemented such that environment variables defined before the con fig file is
executed will not be altered. The sample displayed below makes use of the bash
syntax:
var= "${var:-value}"
such that only uninitialized variables are overwritten by the configuration file.
An example of a sample file is provided, and matches the internal defaults of the
FastFabric tools. For a given release consult
/etc/sysconfig/fastfabric.conf-sample for a sample file with the defaults
of the given release. If fastfabric.conf does not assign a value to a given
configuration variable, the default value will be used.
NOTE:
Do not edit /etc/sysconfig/fastfabric.conf-sample.
The use of various configuration variables are discussed in the Environment
Variables section for each command.
#!/bin/bash
# [ICS VERSION STRING: @(#) ./fastfabric/ib_tools/defaults x_x_x_x_x [MM/DD/YY hh:mm]
# This is a bash sourced config file which defines variables used in
# fast fabric tools. Command line arguments will override these settings.
# Assignments should be scripted such that this file does not override
# exported environment settings, as shown in the defaults below
if [ "$CONFIG_DIR" = "" ]
then
if [ -d /etc/sysconfig ]
then
CONFIG_DIR=/etc/sysconfig
else
CONFIG_DIR=/etc
fi
A-2
IB0054607-01 A
A–FastFabric Configuration Files
FastFabric Configuration File
export CONFIG_DIR
fi
# Override default location for HOSTS_FILE
export HOSTS_FILE=${HOSTS_FILE:-$CONFIG_DIR/iba/hosts}
# Override default location for CHASSIS_FILE
export CHASSIS_FILE=${CHASSIS_FILE:-$CONFIG_DIR/iba/chassis}
# Override default location for ESM_CHASSIS_FILE
export ESM_CHASSIS_FILE=${ESM_CHASSIS_FILE:-$CONFIG_DIR/iba/esm_chassis}
# Override default location for IBNODES_FILE
export IBNODES_FILE=${IBNODES_FILE:-$CONFIG_DIR/iba/ibnodes}
# Override default location for PORTS_FILE
export PORTS_FILE=${PORTS_FILE:-$CONFIG_DIR/iba/ports}
# Default suffix for IPoIB host names
# the special value "NONE" is used to indicate a value of ""
# this value can be used in ff_host_basename_to_ipoib and ff_host_basename
# below to convert between IPoIB and Ethernet hostnames
export FF_IPOIB_SUFFIX=${FF_IPOIB_SUFFIX:--ib}
# Alternatively a IPoIB prefix can be provided, commented out
# examples in ff_host_basename and ff_host_basename can be enabled if desired
export FF_IPOIB_PREFIX=${FF_IPOIB_PREFIX:-ic-}
# Default management host
export MGMT_HOST=${MGMT_HOST:-localhost}
# The shell functions below are only defined if no existing function/command
# with given name, hence allowing use of shell functions or creation of a
# command for this operation
# shell Function to convert a basic hostname into an IPoIB hostname
# if FF_IPOIB_SUFFIX is "", this should return $1 unmodified
# such that commands can be used with -i "" to skip IPoIB operations
# (special value of NONE will be converted to "" before this is called)
if ! type ff_host_basename_to_ipoib >/dev/null 2>/dev/null
then
function ff_host_basename_to_ipoib()
{
# $1 = hostname provided
echo "$1$FF_IPOIB_SUFFIX"
# comment out line above and uncomment line below if using prefixes
#echo "$FF_IPOIB_PREFIX$1"
}
fi
# shell Function to convert a hostname into a basic hostname
# (eg. remove IPoIB suffix, etc) should match result of "hostname -s" on host
# if FF_IPOIB_SUFFIX is "", this should return $1 unmodified
# such that commands can be used with -i "" to skip IPoIB operations
# (special value of NONE will be converted to "" before this is called)
if ! type ff_host_basename >/dev/null 2>/dev/null
then
function ff_host_basename()
{
IB0054607-01 A
A-3
A–FastFabric Configuration Files
FastFabric Configuration File
# $1 = hostname provided (could be ethernet or IPoIB name)
echo "$1"|sed -e "s/$FF_IPOIB_SUFFIX\$//"
# comment out line above and uncomment line below if using prefixes
#echo "$1"|sed -e "s/^$FF_IPOIB_PREFIX//"
}
fi
# IP netmask for IPoIB subnet [-m option]
# if "" default will be determined based on class of IP address [A, B, C]
export FF_IPOIB_NETMASK=${FF_IPOIB_NETMASK:-}
# Maximum parallel processes for ibtest and -p option on other commands
# set to 1 to avoid parallel execution
export FF_MAX_PARALLEL=${FF_MAX_PARALLEL:-20}
# If the systems are slow for some reason, this can be used to provide a
# multiplier for all timeouts in ibtest, iba_host_admin, iba_chassis_admin,
# iba_switch_admin
export FF_TIMEOUT_MULT=${FF_TIMEOUT_MULT:-2}
# where to put result files from FastFabric runs of
# ibtest, iba_host_admin, iba_chassis_admin, iba_switch_admin
#export FF_RESULT_DIR=${FF_RESULT_DIR:-.}
export FF_RESULT_DIR=${FF_RESULT_DIR:-$HOME}
# QLogic IB product to install during iba_host_admin load and upgrade
export FF_PRODUCT=${FF_PRODUCT:-QLogicIB-Basic.`cat /etc/sysconfig/iba/osid_wrapper`}
# QLogic IB product version to install during iba_host_admin load and upgrade
export FF_PRODUCT_VERSION=${FF_PRODUCT_VERSION:-`cat /etc/sysconfig/iba/version_wrapper
2>/dev/null`}
# QLogic IB packages to install during iba_host_admin load [-P option]
#export FF_PACKAGES=${FF_PACKAGES:-iba ipoib verbs_mpi}
#export FF_PACKAGES=${FF_PACKAGES:-iba ipoib psm_mpi}
export FF_PACKAGES=${FF_PACKAGES:-iba ipoib mpi}
# Additional INSTALL options during iba_host_admin load [-I option]
#export FF_INSTALL_OPTIONS=${FF_INSTALL_OPTIONS:- -D iba_mon -D s20tune -E qlogic_sa}
export FF_INSTALL_OPTIONS=${FF_INSTALL_OPTIONS:- -D iba_mon -D s20tune}
# Additional INSTALL options during iba_host_admin upgrade [-U option]
#export FF_UPGRADE_OPTIONS=${FF_UPGRADE_OPTIONS:- -D iba_mon -D s20tune -E qlogic_sa}
export FF_UPGRADE_OPTIONS=${FF_UPGRADE_OPTIONS:- -D iba_mon -D s20tune}
# where to upload server specific files to during uploadall, captureall
# [-d option]
export UPLOADS_DIR=${UPLOADS_DIR:-./uploads}
# where to download server specific files from during downloadall
# [-d option]
export DOWNLOADS_DIR=${DOWNLOADS_DIR:-./downloads}
# where to save history and baselines for cluster analysis and health checks
# [-d option]
export FF_ANALYSIS_DIR=${FF_ANALYSIS_DIR:-/var/opt/iba/analysis}
# topology input file(s) to use for fabric_analysis, all_analysis and
# iba_reports operations. The %P marker will be replaced with the
A-4
IB0054607-01 A
A–FastFabric Configuration Files
FastFabric Configuration File
# local HCA port/fabric selection string (for example 0:0 or 1:2) for
# the fabric being selected (see PORTS_FILE for more information)
# if this file is not found, or the value of this parameter is "NONE"
# no topology input file will be used
export FF_TOPOLOGY_FILE=${FF_TOPOLOGY_FILE:-$CONFIG_DIR/iba/topology.%P.xml}
# How to login to host systems during iba_host_admin
# can be ssh, rsh or telnet
# Note: at present cmdall and scpall always assume password-less ssh/scp
# Note: check_rsh always assumes password-less rsh/rcp
# Note: arguments (-s) to setup_ssh select ssh setup via rsh/rcp or ssh/scp
export FF_LOGIN_METHOD=${FF_LOGIN_METHOD:-ssh}
# initial user name (cannot be root if CFG_LOGIN_METHOD=telnet)
export FF_USERNAME=${FF_USERNAME:-root}
# password for FF_USERNAME (not needed if password-less ssh/rsh configured)
export FF_PASSWORD=${FF_PASSWORD:-}
# if FF_USERNAME is not root, what is the root password needed when
# suing to root
export FF_ROOTPASS=${FF_ROOTPASS:=}
# How to login to chassis
# can be ssh or telnet
export FF_CHASSIS_LOGIN_METHOD=${FF_CHASSIS_LOGIN_METHOD:-ssh}
# password for admin (not needed if password-less ssh configured)
export FF_CHASSIS_ADMIN_PASSWORD=${FF_CHASSIS_ADMIN_PASSWORD:-adminpass}
#base device number for IPoIB devices - IPoIB devices in the systems will
# be named "ib($FF_IPOIB_BASE_DEV_NUM) ib($FF_IPOIB_BASE_DEV_NUM+1) ..."
# On Silvertstorm stack it will be "ib1 ib2 ...."
# On OFED stack it will be "ib0 ib1 ..."
export FF_IPOIB_BASE_DEV_NUM=${FF_IPOIB_BASE_DEV_NUM:-0}
# shell Function to return the base IPoIB device number for this stack type.
# For Silverstorm stack installations this function should return 1, for
# OFED stack installations this should return 0.
if ! type ff_ipoib_base_dev_num >/dev/null 2>/dev/null
then
function ff_ipoib_base_dev_num()
{
# No arguments to this function
# $host is supplied as an arguemnt when this function is called
# but it is ignored here. This is a hack to limit the number of
# changes required.
echo "$FF_IPOIB_BASE_DEV_NUM"
}
fi
# options for iba_report during fabric_analysis health check
# alternative would be to use "-i seconds" instead of -C
# depending on fabric design, more strict link speed checks, such as
# -o slowconfiglinks or -o slowconnlinks may be appropriate
export FF_FABRIC_HEALTH=${FF_FABRIC_HEALTH:- -s -C -o errors -o slowlinks}
# list of CLI commands to issue during chassis_analysis
IB0054607-01 A
A-5
A–FastFabric Configuration Files
FastFabric Configuration File
export FF_CHASSIS_CMDS=${FF_CHASSIS_CMDS:-showInventory fwVersion showIBNodeDesc
ismShowPStatThresh ismChassisSet12x timeZoneConf timeDSTConf snmpCommunityConf
snmpTargetAddr showChassisIpAddr showDefaultRoute}
# other possible additions (if running newer chassis FW which supports these)
# ismIslSet12x, ismIslSetSpeed
# single CLI command to issue to check overall health during chassis_analysis
# hwCheck is prefered, but is not supported on older chassis FW
export FF_CHASSIS_HEALTH=${FF_CHASSIS_HEALTH:-hwCheck}
#export FF_CHASSIS_HEALTH=${FF_CHASSIS_HEALTH:-fruInfo}
# list of CLI commands to issue during esm_analysis
export FF_ESM_CMDS=${FF_ESM_CMDS:-smShowSMParms smShowDefBcGroup}
# list of analysis to perform during all_analysis
# pick appropriate type of SM to analyze
#export FF_ALL_ANALYSIS=${FF_ALL_ANALYSIS:-fabric chassis hostsm esm}
export FF_ALL_ANALYSIS=${FF_ALL_ANALYSIS:-fabric chassis}
# command to perform text files for baseline vs latest snapshot during
# fabric health analysis commands
#export FF_DIFF_CMD=${FF_DIFF_CMD:-diff}
export FF_DIFF_CMD=${FF_DIFF_CMD:-diff -C 1}
# env variables to set within each rank of MPI runs done by fastfabric directly
# alternatively the /opt/iba/src/mpi_apps/ofed.*.param files can be edited
# with appropriate values.
# example of PSM selection of vFabric by PKEY/SL
#export FF_MPI_ENV=${FF_MPI_ENV:-PSM_PKEY=0xffff IPATH_SL=0}
# example of PSM selection of path and vFabric by query to qlogic_sa
#export FF_MPI_ENV=${FF_MPI_ENV:-PSM_PATH_REC=opp PSM_IB_SERVICE_ID=0x1000117500000000}
# example of PSM selection of path and vFabric by default query to qlogic_sa
#export FF_MPI_ENV=${FF_MPI_ENV:-PSM_PATH_REC=opp}
export FF_MPI_ENV=${FF_MPI_ENV:-}
# configuration of arguments for MPI performance deviation testing
# via /opt/iba/src/mpi_apps/deviation/deviation
# The defaults should work for most fabrics. This can be made stricter
# and/or specific to the fabric by using arguments such as -h hostname
# -bwthres and -latthres. See run_deviation or the user's guide for more
# information.
#export FF_DEVIATION_ARGS=${FF_DEVIATION_ARGS:--bwtol 20 -lattol 50 -c -v}
#export FF_DEVIATION_ARGS=${FF_DEVIATION_ARGS:--bwtol 20 -lattol 50 -c -h hostname}
export FF_DEVIATION_ARGS=${FF_DEVIATION_ARGS:--bwtol 20 -lattol 50 -c}
# input to iba_host_admin configipoib indicating which mode to
# use for IP configuration of ipoib interface, valid values are one of
# static, dhcp , auto
export FF_IPOIB_CONFIG=${FF_IPOIB_CONFIG:-static}
A-6
IB0054607-01 A
A–FastFabric Configuration Files
Port Statistics Thresholds Configuration File
Port Statistics Thresholds Configuration File
The /etc/sysconfig/iba/iba_mon.conf configuration file defines port statistics
thresholds for use by iba_report, fabric_analysis, all_analysis and
iba_mon.
This file lists a threshold for each port statistic. If the threshold for a given statistic
is not defined or is set to 0, the given statistic will not be checked.
An example of a sample file is provided, and matches the internal defaults of the
FastFabric tools. For a given release consult
/etc/sysconfig/iba/iba_mon.conf-sample for a sample file with the defaults
of the given release
NOTE:
Do not edit /etc/sysconfig/iba/iba_mon.conf-sample.
#
#
#
#
#
#
#
#
#
#
#
#
#
#
This file controls the iba_mon Port Counter monitoring Thresholds.
[ICS VERSION STRING: @(#) ./config/iba_mon.conf x_x_x_x_x [MM/DD/YY hh:mm]
Error Counters are specified in absolute number of errors over Interval.
All Data Movement thresholds are specified in terms of average data/second
over the monitoring interval.
Setting a threshold to 0 disables monitoring of the given counter
Output is generated when a threshold is exceeded.
Counters for which a non-zero threshold is specified will be cleared by
iba_mon and may impact any remote Performance Managers which are monitoring
the given Counter
Interval10# monitoring interval in seconds
SyslogFacility
local6# syslog facility code, or disable
ThresholdGreater# how compare counter to threshold
# alternative value is Equal
# Normal Data Movement
PortXmitData
PortRcvData
PortXmitPkts
PortRcvPkts
# Error Counters
SymbolErrorCounter
LinkErrorRecoveryCounter
LinkDownedCounter
PortRcvErrors
PortRcvRemotePhysicalErrors
#PortRcvSwitchRelayErrors
PortXmitDiscards
PortXmitConstraintErrors
IB0054607-01 A
0#
0#
0#
0#
as
as
as
as
MB/second
MB/second
packets/second
packets/second
100
3
3
100
0# side effect of errors elsewhere, ignore
100# known Anafa2 issue, incorrectly increments
100
10
A-7
A–FastFabric Configuration Files
Port Statistics Thresholds Configuration File
PortRcvConstraintErrors
LocalLinkIntegrityErrors
ExcessiveBufferOverrunErrors
#VL15Dropped
10
3
3
100 # expected to optimize SM sweep time
NOTE:
When this file is used by iba_mon, the thresholds represent counts per
“Interval”. However, when used by iba_report or fabric health tools, the
counts are absolute values and are applied against the counters as found in
the system
A-8
IB0054607-01 A
A–FastFabric Configuration Files
Signal Integrity Thresholds Configuration File
Signal Integrity Thresholds Configuration File
The /etc/sysconfig/iba/iba_mon.si.conf configuration file defines port
counter signal integrity thresholds.
This file allows analysis for any non-zero error counters related to Signal Integrity
(bad cables, etc) and can be used by adding the -c option to iba_report,
iba_extract_error and other related fastfabric tools.
An example of a sample file is provided, and matches the internal defaults of the
FastFabric tools.
NOTE:
Do not edit /etc/sysconfig/iba/iba_mon.conf-sample.
# This file controls the iba_mon Port Counter monitoring Thresholds.
# [ICS VERSION STRING: @(#) ./config/iba_mon.si.conf x_x_x_x_x [MM/DD/YY hh:mm]
# This is a variation of the default iba_mon.conf file.
# checks error counters related to Signal Integrity.
This file only
Thresholds are set
# such that any and all non-zero counters will be visible.
This can be
# useful when using iba_report -o errors, iba_extract_error, and other
# related tools.
For many FastFabric tools this filename can be specified by
# the -c option.
#
# Error Counters are specified in absolute number of errors over Interval.
# All Data Movement thresholds are specified in terms of average data/second
# over the monitoring interval.
#
# Setting a threshold to 0 disables monitoring of the given counter
#
# Output is generated when a threshold is exceeded.
#
# Counters for which a non-zero threshold is specified will be cleared by
# iba_mon and may impact any remote Performance Managers which are monitoring
# the given Counter
Interval10# monitoring interval in seconds
SyslogFacility
local6# syslog facility code, or disable
ThresholdEqual# how compare counter to threshold
IB0054607-01 A
A-9
A–FastFabric Configuration Files
Signal Integrity Thresholds Configuration File
# alternative value is Greater
# Normal Data Movement
PortXmitData
0# as MB/second
PortRcvData
0# as MB/second
PortXmitPkts
0# as packets/second
PortRcvPkts
0# as packets/second
# Error Counters
A-10
SymbolErrorCounter
1
LinkErrorRecoveryCounter
1
LinkDownedCounter
1
PortRcvErrors
1
PortRcvRemotePhysicalErrors
0# side effect of errors elsewhere, ignore
PortRcvSwitchRelayErrors
0# not related to SI
PortXmitDiscards
0# not related to SI
PortXmitConstraintErrors
0# not related to SI
PortRcvConstraintErrors
0# not related to SI
LocalLinkIntegrityErrors
1
ExcessiveBufferOverrunErrors
1# can be side effect of SI
VL15Dropped
0 # expected to optimize SM sweep time
IB0054607-01 A
A–FastFabric Configuration Files
Host List Files
Host List Files
The /etc/sysconfig/iba/hosts and /etc/sysconfig/iba/allhosts files are
used to specify the hosts which FastFabric will operate against for many
operations.
If desired alternate filenames may be specified in fastfabric.conf, using
environment variables or on the command line. Refer to FastFabric Command
Line Interface Reference Guide for more information.
Below is a sample host list file:
# [ICS VERSION STRING: @(#) ./fastfabric/ib_tools/allhosts x_x_x_x_x [MM/DD/YY hh:mm]
# This file lists the TCP/IP names of ALL the hosts in the cluster.
# THIS SHOULD INCLUDE THE NODE RUNNING FASTFABRIC
#
# If Ethernet is being used for the management network, specify
# the hostname corresponding to the ethernet IP address.
# This file will be used by FastFabric to indicate which hosts should be
# operated on by various fastfabric menus and CLI commands.
include /etc/sysconfig/iba/hosts
# add line below with TCP/IP name of FastFabric host (eg. this host)
Each line of the host list file may specify a single host, a comment or another host
list file to include.
Hosts may be specified by IP address or a resolvable TCP/IP hostname. Typically
hostnames are used for readability. Also, some FastFabric tools will translate the
supplied host names to IPoIB hostnames, in which case names are generally
easier to translate than numeric IP addresses. Typically management network
host names are specified. However, if desired, IPoIB hostnames or IP addresses
may be used. This can accelerate large file transfers and other operations.
Files to be included may be specified using an include directive followed by a file
name. File names specified should generally be absolute path names. If relative
path names are used, they will be searched for within the current directory then
/etc/sysconfig/iba.
Comments may be placed on any line by using a # to precede the comment. On
lines with hosts or include directives, the # must be white-space separated from
any preceding host name, IP address or included file name.
IB0054607-01 A
A-11
A–FastFabric Configuration Files
Chassis List Files
Chassis List Files
The /etc/sysconfig/iba/chassis and /etc/sysconfig/iba/esm_chassis
files are used to specify the QLogic InfiniBand chassis that FastFabric will operate
against for many operations.
If desired alternate filenames may be specified in fastfabric.conf, using
environment variables or on the command line. Refer to FastFabric Command
Line Interface Reference Guide for more information.
The following is a sample chassis file:
#
#
#
#
#
#
#
#
[ICS VERSION STRING: @(#) ./fastfabric/ib_tools/chassis x_x_x_x_x [MM/DD/YY hh:mm]
This file lists the TCP/IP names of the QLogic 9000 and 12000
Internally Managed Switches in the cluster.
If Ethernet is being used for the management network, specify
the name corresponding to the ethernet IP address of the chassis.
This file will be used by FastFabric to indicate which chassis should be
operated on by various fastfabric menus and CLI commands.
Each line of the chassis list file may specify a single chassis, a comment or
another chassis list file to include.
Chassis may be specified by chassis management network IP address or a
resolvable TCP/IP name. Typically, names are used for readability.
Files to be included may be specified using an include directive followed by a file
name. File names specified should generally be absolute path names. If relative
path names are used, they will be searched for within the current directory then
/etc/sysconfig/iba.
Comments may be placed on any line by using a # to precede the comment. On
lines with chassis or include directives, the # must be white-space separated from
any preceding name, IP address or included filename.
The iba_gen_chassis command can be used to help locate chassis in the
fabric and generate a chassis file.
Selection of slots within a chassis
Normally, operations are performed against the management card in the chassis.
For operations such as cmdall, the command is executed against the
management interface for the given chassis. For more sophisticated operations,
such as firmware update, a directory with firmware for each chassis card type can
be supplied and all cards in the chassis will be updated with the appropriate
firmware from that directory.
A-12
IB0054607-01 A
A–FastFabric Configuration Files
Externally Managed Switch List File
However, in some cases it may be desirable to perform operations against a
specific subset of cards within the chassis. In this case the chassis IP address,
name within a chassis list or a chassis file can be augmented with a list of slot
numbers to operate on. This is done in the form:
chassis:slot1,slot2,…
For example:
i9k229:0
i9k229:0,1,5
192.168.0.5:0,1,5
NOTE:
There must be no spaces within the chassis name and/or slot list.
This format is used by cmdall and chassis firmware update. This format may be
used anywhere a chassis name or IP address is valid, such as the -H option, the
CHASSIS environment variable or chassis list files. The slot number specified is
ignored on some operations (such as pingall). Only slots containing
management cards, EVICs and FVICs may be specified with this format. For all
9000 series chassis, slot 0 is always an alias for the presently active management
card for the chassis. For the remainder of slot usages in the chassis, the
chassisQuery command can be executed against a given chassis to identify
which slots have management, EVIC or FVIC cards.
NOTE:
For any operation, care should be taken that a given chassis is listed only
once with all relevant slots as part of that single specification. This is
important so that parallel operations do not cause conflicting concurrent
operations against a given chassis.
Externally Managed Switch List File
The /etc/sysconfig/iba/ibnodes file is used to specify the
externally-managed QLogic InfiniBand switches that FastFabric will operate
against for many operations.
Alternate file names may be specified in fastfabric.conf, using environment
variables or on the command line. Refer to FastFabric Command Line Interface
Reference Guide for more information.
IB0054607-01 A
A-13
A–FastFabric Configuration Files
Externally Managed Switch List File
The following is a sample ibanodes file:
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
#
[ICS VERSION STRING: @(#) ./fastfabric/ib_tools/ibnodes x_x_x_x_x [MM/DD/YY hh:mm]
This file lists all the QLogic 9000 and 12000 Externally Managed Switches
specify one line per switch of the form guid,nodeDesc,distance
guid - node guid of the switch
nodeDesc - optional node description which should be programmed into
the switch by FastFabric. It is recommended to supply a unique
nodeDesc for each switch to simplify management of the cluster.
distance - optional relative distance of the switch from the FastFabric node
this is used by reboot operations to first operate on switches furthest
from the FastFabric node.
Nodes without a distance specified will be treated as furthest.
For fabrics with multiple IB subnets, the local hca and port to use may be
specified as: guid:hca:port,nodeDesc,distance.
See the FastFabric Manual for more info
The iba_gen_ibnodes tool can be used to query the SM and generate a list
Externally Managed switches in the proper form for this file.
for example:
0x00066a00e300299f,SwitchA1,2
Each line of the switch list file may specify a single switch, a comment, or another
switch list file to include.
Switches can be specified by node GUID optionally followed by a colon and the
hca:port, optionally followed by a comma and the IB Node Description
(nodename) to be assigned to the switch, and optionally followed by the distance
value indicating the relative distance from the FastFabric node for each switch.
The iba_gen_ibnodes can be used to help locate externally managed switches in
the fabric and generate an ibnodes file. The iba_gen_ibnodes tool will by default
provide the proper distance value relative to the FastFabric node from which it
was run. This capability requires use of IBTA standard TraceRecord queries which
are not supported by openSM, but can be supplied by the QLogic Fabric Manager.
Alternatively the iba_gen_ibnodes -R option can suppress generation of this
field. See the QLogic FastFabric Command Line Interface Reference Guide for
more information.
In a typical pure fat tree topology with externally managed switches as edge
switches and internally managed switches as core switches, the user can also
easily manually specify proper distance by simply specifying 1 for the distance
value of the switch next to the FastFabric node. Note that in such a topology all
other switches are an equal length from the FastFabric node and a missing
distance value will cause them to be treated as having a distance value which is
larger than any other found in the file. Therefore the other switches would be
rebooted first and the FastFabric node's switch would be rebooted last.
A-14
IB0054607-01 A
A–FastFabric Configuration Files
Externally Managed Switch List File
The GUID will be used to select the switch and on firmware update operations, the
node description will be written to the switch such that other FastFabric tools
(such as iba_aquery and iba_report) can provide a more easily readable name
for the switch. The node description can also be updated as part of switch basic
configuration.
The hca:port may be used to specify which local port (subnet) to use to access the
switch. If this is omitted, all local ports specified will be checked for the switch and
the first port found to be able to access the switch will be used to access it. See
the FastFabric Command Line Interface Reference Guide for more information
about how to specify and hca:port value.
Files to be included may be specified using an include directive followed by a file
name. File names specified should generally be absolute path names. If relative
path names are used, they will be searched for within the current directory then
/etc/sysconfig/iba.
Comments may be placed on any line. By using a # to precede the comment. On
lines with chassis or include directives, the # must be white-space separated from
any preceding GUID, name or included file name.
QLogic recommends that a unique node description be specified for each switch.
This name should follow typical naming rules and use the characters a-z, A-Z, 0-9,
and underscore. No spaces are allowed in the node description. Additionally,
names should not start with a digit.
For externally-managed switches, the node GUID can be found on a label on the
bottom of the switch. Alternately the node GUIDs for switches in the fabric can be
found using a command such as:
iba_saquery -t sw -o nodeguid
NOTE:
The preceding command will report all switch node GUIDs, including those
of internally-managed chassis such as the QLogic 12000 or SilverStorm
9120. GUIDs for internally-managed chassis cannot be specified for use in
the ibnodes file.
FastFabric is topology aware when updating externally managed switch firmware
or resetting the switches. Switches furthest from the FastFabric node are updated
or reset first and then each switch, working toward the FastFabric node. This way
switches that are rebooted are not in the path between the FastFabric node and
others that are being rebooted.
IB0054607-01 A
A-15
A–FastFabric Configuration Files
Externally Managed Switch List File
The ordering is controlled by an optional distance field in the ibnodes file or the
ibnodes provided on the command line. The distance field indicates the relative
distance from the FastFabric node for each switch. Any ibnodes file entries
which do not specify a distance value are treated as having a value larger than
any others in the file. The ibnodes file contains any of these formats per line:
nodeguid
nodeguid,,distance
nodeguid:hca:port
nodeguid:hca:port,,distance
nodeguid,nodename
nodeguid,nodename,distance
nodeguid:hca:port,nodename
nodeguid:hca:port,nodename,distance
The iba_gen_ibnodes tool by default, provides the proper distance value
relative to the FastFabric node on which it ran. This capability requires the use of
IBTA standard TraceRecord queries which are not supported by openSM, but can
be supplied by QLogic FM. Alternatively, the iba_gen_ibnodes -R option can
suppress generation of this field.
In a typical pure fat tree topology with externally managed switches as edge
switches and internally managed switches as core switches, the user can also
manually specify proper distance by simply specifying 1 for the distance value of
the switch next to the FastFabric node. Note that in such a topology all other
switches are an equal length from the FastFabric node and a missing hops value
will cause them to be treated as having a distance value which is larger than any
other found in the file. Therefore, the other switches would be rebooted first and
the FastFabric node's switch would be rebooted last.
A-16
IB0054607-01 A
A–FastFabric Configuration Files
Port List File
Port List File
The /etc/sysconfig/iba/ports file is used to specify the local HCA ports (i.e.,
subnets) that FastFabric will use for assorted commands (such as iba_reports,
fabric_info, iba_switch_admin, fabric_analysis, all_analysis) for fabric
access.
Alternate filenames may be specified in fastfabric.conf, using environment
variables or on the command line. Refer to FastFabric Command Line Interface
Reference Guide for more information.
The following is a sample port list file:
# [ICS VERSION STRING: @(#) ./fastfabric/ib_tools/ports x_x_x_x_x [MM/DD/YY hh:mm]
# This file defines the local HCA ports to use to access the fabric(s)
#
# specify one line per HCA port of the form hca:port such as:
#
0:0 = 1st active port in system
#
0:y = port y within system
#
x:0 = 1st active port on HCA x
#
x:y = HCA x, port y
# The first HCA in the system is 1.
The first port on an HCA is 1.
0:0
Each line of the port list file may specify a single port, a comment or another port
list file to include.
Ports are specified as hca:port. No spaces are permitted. The first Host Channel
Adapter is 1 and the first port is 1. The special value 0 for Host Channel Adapter
or port has special meaning. The allowed formats are shown in the previous
sample.
Files to be included may be specified using an include directive followed by a file
name. File names specified should generally be absolute path names. If relative
path names are used, they will be searched for within the current directory then
/etc/sysconfig/iba.
Comments may be placed on any line. By using a # to precede the comment. On
lines with a port or include directive, the # must be white-space separated from
any preceding port or included file name.
IB0054607-01 A
A-17
A–FastFabric Configuration Files
Fabric Topology Input File
Fabric Topology Input File
The /etc/sysconfig/iba/topology.0:0.xml file is used to specify the
expected fabric topology and augmented fabric information (such as cable labels,
types, lengths, SM details, node details, link details, etc). If present this file will be
used by assorted FastFabric commands (such as iba_reports,
fabric_analysis, all_analysis). Refer to the FastFabric Command Line
Interface Reference Guide for more information on how to create a topology file
describing the fabric.
If desired, alternate filenames may be specified in fastfabric.conf, using
environment variables or on the command line. Refer to FastFabric Command
Line Interface Reference Guide for more information.
The XML format of topology input can appear as follows (the example below is
purposely brief and omits many links, nodes, and SMs):
<?xml version="1.0" encoding="utf-8" ?>
<Report>
<LinkSummary>
<Link>
<Rate>20g</Rate>
<MTU>2048</MTU>
<Internal>0</Internal>
<LinkDetails>Bender to Switch</LinkDetails>
<Cable>
<CableLength>11m</CableLength>
<CableLabel>S4567</CableLabel>
<CableDetails>gore cable model 456</CableDetails>
</Cable>
<Port>
<NodeGUID>0x0002c9020020e004</NodeGUID>
<PortGUID>0x0002c9020020e005</PortGUID>
<PortNum>1</PortNum>
<NodeType>CA</NodeType>
<NodeDesc>bender HCA-1</NodeDesc>
<PortDetails>bender primary port</PortDetails>
</Port>
<Port>
<NodeGUID>0x00066a0007000df6</NodeGUID>
<PortNum>1</PortNum>
<NodeType>SW</NodeType>
<NodeDesc>i9k159 Leaf 4, Chip A</NodeDesc>
</Port>
</Link>
<Link>
<Rate>20g</Rate>
<MTU>2048</MTU>
<Internal>0</Internal>
<Port>
<NodeGUID>0x0002c9020025a678</NodeGUID>
<PortGUID>0x0002c9020025a679</PortGUID>
A-18
IB0054607-01 A
A–FastFabric Configuration Files
Fabric Topology Input File
<PortNum>1</PortNum>
<NodeType>CA</NodeType>
<NodeDesc>mindy2 HCA-1</NodeDesc>
</Port>
<Port>
<NodeGUID>0x00066a0007000e6d</NodeGUID>
<PortNum>4</PortNum>
<NodeType>SW</NodeType>
<NodeDesc>i9k159 Leaf 5, Chip A</NodeDesc>
</Port>
</Link>
</LinkSummary>
<Nodes>
<CAs>
<Node id="0x0002c9020025a678">
<NodeGUID>0x0002c9020025a678</NodeGUID>
<NodeDesc>mindy2 HCA-1</NodeDesc>
<NodeDetails>mindy2 only HCA</NodeDetails>
</Node>
</CAs>
<Switches>
<Node id="0x00066a000600025a">
<NodeGUID>0x00066a000600025a</NodeGUID>
<NodeDesc>i9k159 Spine 1, Chip A</NodeDesc>
<NodeDetails>core switch</NodeDetails>
</Node>
</Switches>
<SMs>
<SM id="0x0002c9020025a678:1">
<NodeGUID>0x0002c9020025a678</NodeGUID>
<NodeDesc>mindy2 HCA-1</NodeDesc>
<PortNum>1</PortNum>
<PortGUID>0x0002c9020025a679</PortGUID>
<NodeType>CA</NodeType>
<NodeType_Int>1</NodeType_Int>
<SMDetails>mindy2 SM</SMDetails>
</SM>
</SMs>
</Nodes>
</Report>
IB0054607-01 A
A-19
A–FastFabric Configuration Files
Fabric Topology Input File
A-20
IB0054607-01 A
Corporate Headquarters QLogic Corporation 26650 Aliso Viejo Parkway
Aliso Viejo, CA 92656 949.389.6000
www.qlogic.com
International Offices UK | Ireland | Germany | France | India | Japan | China | Hong Kong | Singapore | Taiwan
© 2011 QLogic Corporation. Specifications are subject to change without notice. All rights reserved worldwide. QLogic, the QLogic logo, and FastFabric
are registered trademarks of QLogic Corporation. All other brand and product names are trademarks or registered trademarks of their respective
owners. Information supplied by QLogic Corporation is believed to be accurate and reliable. QLogic Corporation assumes no responsibility for any
errors in this brochure. QLogic Corporation reserves the right, without notice, to make changes in product design or specifications.
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement