CM1 resources are reported missing on a single socket T5-2 (T5-2 1S)

Asset ID:	1-72-2268011.1
Update Date:	2018-05-01
Keywords:

Solution Type Problem Resolution Sure

Solution 2268011.1 : CM1 resources are reported missing on a single socket T5-2 (T5-2 1S)

Applies to:

SPARC T5-2 - Version All Versions to All Versions [Release All Releases]
Information in this document applies to any platform.

Symptoms

During initialisation the platform may erroneously report CM1 and all of it's associated Memory Risers are missing on a single socket T5-2 system;

2017-05-13/19:42:11 ereport.component.absent@/SYS/MB/CM1/CMP/MR3
2017-05-13/19:42:12 ereport.component.absent@/SYS/MB/CM1/CMP/MR2
2017-05-13/19:42:12 ereport.component.absent@/SYS/MB/CM1
2017-05-13/19:42:13 ereport.component.absent@/SYS/MB/CM1/CMP/MR1
2017-05-13/19:42:14 ereport.component.absent@/SYS/MB/CM1/CMP/MR0

This will lead to a fault being triggered against each of the Memory Riser locations (MR0-MR3), for example;

------------------- ------------------------------------ -------------- --------
Time UUID msgid Severity
------------------- ------------------------------------ -------------- --------
2017-05-13/19:46:44 749d3d19-cc8b-efbf-d72a-e14e98c19bb5 SPT-8000-J4 Major

Problem Status : open
Diag Engine : fdd 1.0
System
Manufacturer : Oracle Corporation
Name : SPARC T5-2
Part_Number : 33043257+1+1
Serial_Number : AKXXXXXXXX

----------------------------------------
Suspect 1 of 1
Fault class : fault.component.missing
Certainty : 100%
Affects : /SYS/MB/CM1/CMP/MR0
Status : not present

FRU
Status : faulty
Location : /SYS
Manufacturer : Oracle Corporation
Name : SPARC T5-2
Part_Number : 33043257+1+1
Serial_Number : AKXXXXXXXX
Chassis
Manufacturer : Oracle Corporation
Name : SPARC T5-2
Part_Number : 33043257+1+1
Serial_Number : AKXXXXXXXX

Description : A required device is missing from the chassis.

Response : The service required LED on the chassis may be illuminated.

Impact : The services provided by the components will not be
available. The Host may not be permitted to power on.

Action : Please refer to the associated reference document at
http://support.oracle.com/msg/SPT-8000-J4 for the latest
service procedures and policies regarding this diagnosis.

This issue will only occur if one of the following Motherboards are installed;

7088094: Single 16-Core 3.6GHz System Board
7302921: Single 16-Core 3.6GHz System Board
7340484: Single 16-Core 3.6GHz System Board
7340486: Single 16-Core 3.6GHz System Board
7341348: Single 16-Core 3.6GHz System Board

To confirm the currently installed Motherboards;

-> show /SYS/MB fru_part_number

/SYS/MB
Properties:
fru_part_number = 7088094

Cause

During initialisation the Service Processor may load an incorrect sensor list which results in various services probing non-existent devices, this in turn leads to the fault management service triggering faults.

Solution

Restore the factory default configuration on the Service Processor, followed by a full platform power cycle;

1. Shutdown /SYS if not done so already, halt services as appropriate;

-> stop /SYS

2. Clear any pre-existing SPT-8000-J4 FMA faults via the SP fault management shell;

-> start /SP/faultmgmt/shell/
Are you sure you want to start /SP/faultmgmt/shell (y/n)? y

faultmgmtsp> fmadm faulty -s
No faults found
faultmgmtsp>

3. Backup the Service Processor configuration in order to manually restore settings following the reset to factory defaults;

-> set /SP/config dump_uri=scp://<user>@<remote_host>/<location>/<filename.xml>

4. Set factory defaults, please note this must be 'factory' and not 'all';

-> set /SP/reset_to_defaults=factory

5. Remove AC from the platform for 60s, reapply and restore SP configuration settings using the xml backup taken in step#3.

For example;

/SP/network
/SP/clock
/SP/powermgmt
/SP/alertmgmt
/SP/services

6. Reinitialise the host by restarting /SYS;

-> start /SYS

7. Confirm that no further faults or events indicating CM1 resources are missing are reported;

-> start /SP/faultmgmt/shell/
Are you sure you want to start /SP/faultmgmt/shell (y/n)? y

faultmgmtsp> fmdump -ev
<..snip..>
faultmgmtsp> fmadm faulty -s
No faults found
faultmgmtsp>

8. If further events are reported please escalate to Oracle Support Services.

Attachments

This solution has no attachment