SPARC T4-1
 
SPARC T4-2
 

PLA-Support>Sun Systems>SPARC>CMT>SN-SPARC: T4
 

In this Document

Symptoms

Changes

Cause

Solution

References

Created from <SR 3-9955038141>

Applies to:

Symptoms

 System will shut down following a  ERROR:   Unexpected zero POK value, adjusting state machine event.

SP logs (show /SP/logs/event/list):

148    Tue Dec  2 06:07:54 2014  System    Log       minor          Host: Powered Off
147    Tue Dec  2 06:07:54 2014  System    Log       minor          ERROR:   Unexpected zero POK value, adjusting state machine

The vbsc log from the snapshot may show this error (/ilom/vbsc.log):

Dec  2 06:07:54 ERROR:   Unexpected zero POK value, adjusting state machine  
Dec  2 06:07:54 WARNING: sp_port_to_host_ldc: dropping data due to incomplete
                        write on fma channel, expected:actual (24:0)
Dec  2 06:07:54 DEBUG:   POK REG0: 0x80 NOT OK:PSU0|PSU1|CPU0|MEM|OIO|DC_POK|
Dec  2 06:07:54 DEBUG:   POK REG1: 0xa0 NOT
                        OK:CORE|CPU0_1V|CPU0_1V5|VDD_3V3|SAS_1V|5V|
Dec  2 06:07:54 DEBUG:   POK REG2: 0xf4 NOT OK:
Dec  2 06:07:54 DEBUG:   POK REG3: 0xff NOT OK: 

The above issue may be preceded or proceeded by a ereport.cpu.generic-sparc.c2c-prot-uc event.

Additionally, you may encounter a hostconfig problem:

1904 Wed May 17 17:51:10 2017 System Log major May 17 17:51:10 ERROR: Unexpected zero POK value, adjusting state machine

1901 Wed May 17 17:51:08 2017 System Log major May 17 17:51:08 ERROR: Fatal Hostconfig Error - Node: 0, Strand: 0

NOTE:

It's also possible the machine may report a SUN4V-8002-Q2 (fault.memory.memlink-uc) and/or SUN4V-8002-MJ (fault.cpu.generic-sparc.c2c-uc).

Changes

Cause

 Issue is possibly caused by an erroneous sensor reading or an actual issue on the board itself.

Solution

Update the firmware if it's not up to date to make sure we're not hitting a tuning issue on a sensor.

Please open a service request with Oracle Support to investigate this issue. An ILOM snapshot would be necessary for the investigation.

Though the MB is the most likely suspect, please use the data to determine whether other electrically dead components may cause this, such as a MR as there have been a few special cases of other components causing this.

References

This solution has no attachment