Correctable Memory Error messages found in the event log report an incorrect DIMM slot as having an error - IBM eServer xSeries 445

Resolving The Problem

RETAIN tip: H182175

Correctable Memory Error messages found in the event log report an incorrect DIMM slot as having an error (e.g. reports Socket J07 instead of Socket J06).

1 ERR SERVPROC 06/10/04 17:43:21 PFA Alert, see preceding error in system error log.
 
2 ERR SMI Hdlr 06/10/04 17:43:21 00151500
Excessive Correctable Memory Errors Detected Chassis=01 SMP
Expansion Module=02 Socket=J07
 
3 INFO SMI Hdlr 06/10/04 14:39:29 00151803
Memory ProteXion(C) Event Detected Chassis=01 SMP Expansion Module=02 Socket=J07 Symbol=05

If the Socket in actual error is J05, J06, J07, or J08, then the socket number that is reported as defective is one above the correct socket number.

The system may be any of the following IBM eServer systems:

• an IBM eServer xSeries 445, type 8870, any model.

The system is configured with the following option(s):

• an IBM Remote Supervisor Adapter, Option part number (p/n) 09N7585, Field Replacement Unit (FRU) p/n 36L9912.
• an IBM Remote Supervisor Adapter II-EXA, Option p/n 13N0382, FRU p/n 73P9246.

The BIOS level(s) affected is:

• 42B or earlier

This behavior is corrected in the latest release of BIOS.

If Correctable Memory Error messages are found in event log, check the Socket in error. If the socket is J05, J06, J07, or J08, then subtract 1 from the Socket in error and replace that DIMM.

Correctable error messages incorrect for DIMMS J05, J06, J07, J08 on any SMP (Simple Muliti-Processor) board or any Chassis. BIOS code misidentified the DIMM socket in error. so this will be seen on J5-8 on any SMP Expansion Module in any position in any chassis.