IBM BladeCenter GPU expansion Blade with NVIDIA Tesla 'FAULT' LED illuminated after power on - IBM BladeCenter

Resolving The Problem

Source

RETAIN tip: H207765

Symptom

The Fault Light Emitting Diode (LED) may turn on or flash intermittently after removing and restoring power to the Graphical Processor Unit (GPU) Expansion Blade configured with a NVIDIA Tesla GPU.

There will be no associated errors logged in IBM BladeCenter Advanced Management Module (AMM) or IBM Dynamic System Analysis (DSA).

Affected configurations

The system may be any of the following IBM servers:

• BladeCenter HS23, type 7875, any model
• BladeCenter HS23E, type 8038, any model
• BladeCenter HS23E, type 8039, any model

The system is configured with one or more of the following IBM Options:

• IBM BladeCenter GPU Expansion Blade II with NVIDIA Tesla M2070Q, Option part number 68Y7479, any replacement part number (CRU)
• IBM BladeCenter GPU Expansion Blade II with NVIDIA Tesla M2075, Option part number 68Y7478, any replacement part number (CRU)
• IBM BladeCenter GPU Expansion Blade II with NVIDIA Tesla M2090, Option part number 00D6881, any replacement part number (CRU)
• IBM BladeCenter PCI Express Gen 2 Expansion Blade II, Option part number 68Y7484, any replacement part number (CRU)

This tip is not software specific.

The Integrated Management Module (IMM) firmware for the Blade Server is affected.

The system has the symptom described above.

Solution

This issue was corrected in the Integrated Management Module II (IMM2) firmware update version 2.50 Build ID: 1AOO40z.

The file is available by selecting the appropriate Product Group, type of System, Product name, Product machine type, and operating system on IBM Support's Fix Central web page, at the following URL:

http://www.ibm.com/support/fixcentral/

Workaround

In order to work around the issue, reset the IMM by following the steps:

1. Log in to the AMM by pointing the web browser to the Internet Protocol (IP) of the AMM and entering the login credentials.
2. Once logged in to the AMM, expand the Management Module (MM) Control menu.
3. Select the Restart MM option.
4. Click the Restart radio button.

Note: Allow up to 10 minutes for AMM to restart completely and discover all the devices populated inside the BladeCenter chassis.

Additional information

There is no functional impact when the LED is illuminated erroneously under the condition described.