MB/CMP0/MCUx disabled after upgrading with 8GB DIMMs

Asset ID:	1-72-1566088.1
Update Date:	2018-03-06
Keywords:

Solution Type Problem Resolution Sure

Solution 1566088.1 : MB/CMP0/MCUx disabled after upgrading with 8GB DIMMs

Applies to:

Sun SPARC Enterprise T5120 Server - Version All Versions to All Versions [Release All Releases]
Information in this document applies to any platform.

Symptoms

After upgrading/replacing the existing RAM configuration by 8GB Memory Modules, the following error may occur during server Power On Self Test:

2013-06-20 03:10:40.002 0:0:0>INFO:
2013-06-20 03:10:40.055 0:0:0> POST Passed all devices.
2013-06-20 03:10:40.109 0:0:0>POST: Return to VBSC.
2013-06-20 03:10:40.162 0:0:0>Master set ACK for vbsc runpost command and spin...

ERROR: The following devices are disabled:
MB/CMP0/MCU2

- in ALOM compatibility shell, the following will be shown:

sc> showfaults -v
Last POST Run: Thu Jun 20 03:10:40 2013

Post Status: Passed all devices
ID Time                           FRU       Class     Fault
   1 Jun 20 03:06:26                /SYS/MB             SP detected fault: /SYS/MB/CMP0/MCU2 Forced fail (IBIST)
   2 Jun 20 03:06:28                /SYS                SP detected fault: Jun 20 03:06:28 ERROR: Operating with a
                                                        degraded memory configuration.
sc>

- the same error, shown in ILOM shell:

Changes

Memory Modules of the following type have been used:
511-1262 [C] 8GB DDR2-667/PC2-5300 1.8V FBDIMM

Cause

IBIST fails channel initialization on a single Memory Control Unit (MCU) on first initialization only,
this is not a real hardware fault.

Solution

1) Re-enable the MCU (MCU2 in this case) by executing:
-> set /SYS/MB/CMP0/MCU2 component_state=enabled

2) Clear the faults

a) in ILOM shell:

-> set /SYS/MB/CMP0/MCU2 clear_fault_action=true

   b) if in ALOM compatibility shell, you should still see faults in "sc> showfaults -v"
   execute "sc> clearfault" on each component. In this example:

   sc> clearfault /SYS/MB
   sc> clearfault /SYS

3) power down the server and remove power cables, Re-seat the memory modules.

4) Power on. The server should pass POST without further errors.

5) if POST fails any of the memory modules again, please open a Service Request at Oracle.

References

<BUG:15778946> - SUNBT7153775 IBIST MCU CHANNEL INITIALISATION FAILURES AFTER INSTALLING 8GB DIMM

Attachments

This solution has no attachment