Home » HP Manuals » Servers » HP Cluster Platform Interconnects v2010 » Manual Viewer

HP Cluster Platform Interconnects v2010 HP Cluster Platform InfiniBand Interco - Page 108

Debugging a Fabric Failure by Using Performance Management (PM)

Get HP Cluster Platform Interconnects v2010 PDF manuals and user guides

View all HP Cluster Platform Interconnects v2010 manuals

Add to My Manuals
Save this manual to your list of manuals

Page 108 highlights

3. Check the Link LEDs on the line boards front panel for the various ports. In the event of a system lock-up in the ISR 9XXX, use the reset button to reset and reboot the system. The system has two reset buttons: one is located on the front panel of the sFU-8 fan assembly module and one of the front panel of the sCTRL board. To reset the router push the button, using a thick wire or tip of a pen until the system reboots. Remove the wire immediately afterwards. 4. Verify that the PC terminal emulation program is set correctly and that the PC is connected properly to the Console management port if the boot information and CLI prompt are not displayed. 9.3 Debugging a Fabric Failure by Using Performance Management (PM) The interconnect's fabric manager enables you to debug problem with fabric connections by using the performance management (PM) features. The following two PM functions support fabric debugging: • Port counters monitoring and report. The PM generates a periodic port counters report file in text comma separated value format (CSV format). You can import this file format into spreadsheets such as Microsoft Excel for analysis. The PM monitors port counter errors and reports every port that passed its user-configured error threshold limit. Examine the report file to find problem ports. • Event logging. the PM creates an event log file for both IB traps and SM internal events. Examine the log to find problem ports. Refer to the Voltaire InfiniBand Fabric Management and Diagnostic Guide and the HP Cluster Platform InfiniBand Fabric Management and Diagnostic Guide for information on using the PM and interpreting the log files. 9.4 Identifying a Leaf or Spine Port Malfunction on an ISR 9XXX You can identify a failed leaf or spine port by using the fabric manager as follows: 1. Launch the fabric manager and select the relevant ISR 9XXX interconnect in the topology map of the main window. 2. Click the nodes information icon in the toolbar to launch a new browser window. 3. A separate browser window displays the following data: • ISR 9XXX node information. • The leaf/spine chip status. • The port connection matrix, including: - Each ISR 9XXX external port. - Each internal chip port. - The channel adapter connection. For example: L4:4, Anafa:1, LID:0680, Port:1 Attached to CA, LID:06A9, Port:2, GUID:0008f10403965020 9.5 Detecting a Failed Port By combining the node information and the port counter file information described in the Voltaire InfiniBand Fabric Management and Diagnostic Guide and theHP Cluster Platform InfiniBand Fabric Management and Diagnostic Guide, you can identify any fabric port connectivity failure. A port is considered failed if one of the following conditions occur: • It does not respond to a NodeInfo request. • It has a GUID that is a duplicate of another port's GUID. 108 Postinstallation Troubleshooting and Diagnostics

Section	Page
InfiniBand Interconnect Installation and User's Guide	1
Table of Contents	3
About This Manual	11
Organization	11
Audience	12
Related Documentation	12
Voltaire Documentation	12
HP Cluster Platform Documentation	12
Cluster Component Documentation	13
HP Encourages Your Comments	13
Text Annotations	13
Conventions	14
Safety Considerations	15
Information on Ergonomics	16
Recycling	17
1 InfiniBand Technology Overview	19
1.1 Identifying InfiniBand Fabric Components	20
1.2 ISR 9xxxx-series Interconnects	21
1.3 Identifying the ISR 9024 Interconnect (SDR)	22
1.4 Identifying the ISR 9024 S/D Interconnect (RoHS Compliant)	22
1.5 Identifying the ISR 9096 Interconnect	23
1.6 Identifying the ISR 9288 Interconnect	24
1.7 Understanding the Interconnect Management Capabilities	26
2 Installing and Maintaining the ISR 9024 Interconnect	29
2.1 Overview of the ISR 9024 Interconnect	29
2.1.1 Internally Managed ISR 9024 Front Panel	30
2.1.2 ISR 9024 Management Card	31
2.1.3 Externally Managed ISR 9024 Front Panel	32
2.1.4 ISR 9024 Rear Panel	32
2.1.5 Cooling and Airflow	33
2.2 Unpacking the ISR 9024 Interconnect	33
2.3 Mounting the ISR 9024 in the Rack	34
2.4 Installing the ISR 9024 Power Supply Unit	36
2.5 Operating the ISR 9024 Interconnect	36
2.6 Troubleshooting the ISR 9024	36
3 Installing and Maintaining the ISR 9024 S/D Interconnect (RoHS Compliant)	39
3.1 Overview of the ISR 9024 S/D Interconnect	39
3.1.1 Internally Managed ISR 9024 S/D–M Front Panel	40
3.1.1.1 ISR 9024S/D-M Management Card	41
3.1.2 Externally Managed ISR 9024 S/D Front Panel	41
3.1.3 ISR 9024 S/D Rear Panel	42
3.2 Unpacking the ISR 9024 S/D Interconnect	43
3.3 Mounting the ISR 9024 S/D in the Rack	44
3.4 The ISR 9024 S/D Fan Unit	45
3.4.1 Replacing the Fan Unit	46
3.5 Installing the ISR 9024 S/D Power Supply Units	46
3.6 Operating the ISR 9024 S/D Interconnect	47
3.7 Troubleshooting the ISR 9024 S/D	47
4 Installing and Maintaining the ISR 9096 and ISR 9288 Interconnects	49
4.1 Overview of the ISR 9XXX Interconnects	49
4.1.1 ISR ISR 9XXX Interconnect Features	49
4.1.2 ISR ISR 9XXX Interconnect Architecture	50
4.1.3 ISR ISR 9XXX Building Blocks	50
4.1.3.1 Field Replaceable Units	51
4.1.4 ISR ISR 9XXX System Firmware	51
4.1.5 ISR ISR 9XXX Control Panels and Reset Buttons	52
4.2 Unpacking the Chassis	52
4.2.1 Unpacking the ISR 9288	53
4.2.2 Unpacking the ISR 9096	55
4.2.3 Verifying the Shipment	56
4.3 Repacking the Chassis	58
4.4 Mounting the ISR 9XXX in the Rack	58
4.4.1 Resources Required to Perform an Installation	59
4.4.2 Mounting The Chassis in the Rack	59
4.4.3 Installing the Cabling Clamps	63
4.5 Standard Configurations	63
4.6 Post-installation Tasks	67
5 HP 4x DDR IB Switch Module for c-Class BladeSystems Overview	69
5.1 Installing the 4x DDR IB Switch Module	69
5.1.1 Subnet Manager	70
6 Installing Field-Replaceable Units in the ISR 9XXX Chassis	71
6.1 Guidelines for Inserting and Extracting Boards	71
6.2 Installing or Replacing the sFB-12 Fabric Board	72
6.3 Installing or Replacing sLB-24 Line Boards	73
6.4 Installing or Replacing the sMB (Management) Board	75
6.5 Installing and Replacing Power Supply Units	76
6.6 Populating the Router Blade Drawer (sRBD)	77
6.6.1 Overview of the IPR (TCP/IP) Blade	77
6.6.2 Overview of the FCR (Fibre Channel Router) Blade	78
6.6.3 Installing a Router Blade	79
6.7 Installing or Replacing the Router Blade Drawer (sRBD)	80
6.8 Installing and Replacing the sCTRL Board	81
6.9 Installing the sFU-4 Fan Module	82
6.10 Installing the sFU-8 Fan Module	83
7 Cabling the Interconnect	85
7.1 InfiniBand Cabling	85
7.1.1 General InfiniBand Cabling Guidelines	85
7.1.2 Cabling Tables	86
7.1.3 Connecting InfiniBand Cables	86
7.1.4 Cable Routing Procedure for the ISR 9024 or ISR 9024 S/D	86
7.1.5 Cable Routing Procedure for the ISR 9096	87
7.1.6 Connecting the ISR 9096 Router Blades	88
7.1.6.1 Connecting to IPR GbE Ports	88
7.1.6.2 Connecting to FCR FC Ports	88
7.1.7 Cable Routing Procedure for the ISR 9288	88
7.2 Connecting the ISR 9288 Router Blades	89
7.2.1 Connecting to IPR GbE Ports	89
7.2.2 Connecting to FCR FC Ports	90
7.3 Chassis Administrative Connections	90
7.3.1 Cabling the ISR 9024 Management Connections	90
7.3.2 Cabling the ISR 9096 and the ISR 9288 Management Connections	90
7.3.3 Connecting the Management Port Serial Interface	91
8 Installing and Maintaining HCA Cards and Mezzanines	93
8.1 Overview of the Voltaire HCA 400 PCI–X, and HCA 400 PCI-Express Cards	93
8.2 Installing the HCA	94
8.3 Topspin/Mellanox PCI-X HCA	97
8.4 Topspin/Mellanox PCI-Express HCA	98
8.5 Mellanox PCI-X HCA (SDR)	99
8.5.1 Mellanox PCI-X SDR HCA LEDs	99
8.6 Mellanox PCI-Express HCA (SDR)	100
8.6.1 Mellanox PCI-Express SDR HCA LEDs	100
8.7 Mellanox Memory Free PCI-Express HCA (SDR)	101
8.7.1 LEDs	102
8.8 Mellanox Memory Free PCI-Express HCA (DDR)	102
8.8.1 LEDs	103
8.9 Mellanox PCI-Express HCA (DDR)	103
8.9.1 Mellanox PCI-Express DDR HCA LEDs	104
8.9.2 HP HPC 4x DDR IB Mezzanine HCA	105
8.9.3 Operating System and Software Requirements	105
9 Postinstallation Troubleshooting and Diagnostics	107
9.1 Postinstallation Troubleshooting	107
9.2 Startup Checks for the ISR 9XXX	107
9.3 Debugging a Fabric Failure by Using Performance Management (PM)	108
9.4 Identifying a Leaf or Spine Port Malfunction on an ISR 9XXX	108
9.5 Detecting a Failed Port	108
A ISR 9024 Interconnect Specifications	111
B ISR 9024 S/D Interconnect Specifications (RoHS Compliant)	113
C ISR 9096 Interconnect Specifications	117
D ISR 9288 Interconnect Specifications	119
E 4x DDR IB Switch Module Specifications	123
F HCA PCI Card Specifications	125
F.1 Voltaire HCA–400 PCI Card Specifications	125
F.2 Topspin/Mellanox PCI-X HCA Specifications	126
F.3 Topspin/Mellanox PCI-Express HCA Specifications	127
F.4 Mellanox PCI-X HCA Specifications	128
F.5 Mellanox PCI-Express HCA Specifications	128
F.6 Mellanox Memory Free PCI-Express HCA (SDR) Specifications	129
F.7 Mellanox Memory Free PCI-Express HCA (DDR) Specifications	129
F.8 Mellanox PCI-Express DDR HCA Specifications	130
F.9 HPC 4x DDR IB Mezzanine HCA Specifications	130

Match case Limit results 1 per page

Check the Link LEDs on the line boards front panel for the various ports. In the event of a

system lock-up in the ISR 9XXX, use the reset button to reset and reboot the system.

The system has two reset buttons: one is located on the front panel of the sFU-8 fan assembly

module and one of the front panel of the sCTRL board. To reset the router push the button,

using a thick wire or tip of a pen until the system reboots. Remove the wire immediately

afterwards.

Verify that the PC terminal emulation program is set correctly and that the PC is connected

properly to the Console management port if the boot information and CLI prompt are not

displayed.

9.3 Debugging a Fabric Failure by Using Performance Management (PM)

The interconnect's fabric manager enables you to debug problem with fabric connections by

using the performance management (PM) features. The following two PM functions support

fabric debugging:

•

Port counters monitoring and report.

The PM generates a periodic port counters report file in text comma separated value format

(CSV format). You can import this file format into spreadsheets such as Microsoft Excel for

analysis. The PM monitors port counter errors and reports every port that passed its

user-configured error threshold limit. Examine the report file to find problem ports.

•

Event logging.

the PM creates an event log file for both IB traps and SM internal events. Examine the log

to find problem ports.

Refer to the

Voltaire InfiniBand Fabric Management and Diagnostic Guide

and the

HP Cluster Platform

InfiniBand Fabric Management and Diagnostic Guide

for information on using the PM and interpreting

the log files.

9.4 Identifying a Leaf or Spine Port Malfunction on an ISR 9XXX

You can identify a failed leaf or spine port by using the fabric manager as follows:

Launch the fabric manager and select the relevant ISR 9XXX interconnect in the topology

map of the main window.

Click the nodes information icon in the toolbar to launch a new browser window.

A separate browser window displays the following data:

•

ISR 9XXX node information.

•

The leaf/spine chip status.

•

The port connection matrix, including:

—

Each ISR 9XXX external port.

—

Each internal chip port.

—

The channel adapter connection.

For example:

L4:4, Anafa:1, LID:0680, Port:1 Attached to CA, LID:06A9, Port:2, GUID:0008f10403965020

9.5 Detecting a Failed Port

By combining the node information and the port counter file information described in the

Voltaire

InfiniBand Fabric Management and Diagnostic Guide

and the

HP Cluster Platform InfiniBand Fabric

Management and Diagnostic Guide

, you can identify any fabric port connectivity failure. A port is

considered failed if one of the following conditions occur:

•

It does not respond to a NodeInfo request.

•

It has a GUID that is a duplicate of another port's GUID.

108

Postinstallation Troubleshooting and Diagnostics