SPARC T5-2 Power Supply SPT-8000-MJ faults triggered by V_OUT_OK (V_IN

Asset ID:	1-72-2092329.1
Update Date:	2018-01-19
Keywords:

Solution Type Problem Resolution Sure

Solution 2092329.1 : SPARC T5-2 Power Supply SPT-8000-MJ faults triggered by V_OUT_OK (V_IN_INFO) de-asserted events

Applies to:

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

Symptoms

Power One (or Bell Power) A239E and A239A PSUs installed in SPARC T5-2 systems may report a temporary interruption to supply fed to the chassis triggering a SPT-8000-MJ fault. In this specific scenario there is no indication of loss of inlet power to the supply itself, and sensor readings report normal operation is resumed within 20 seconds.

Please note there is no evidence of a genuine interruption to power supplied by the PSU, this is a sensor reporting issue.

The first indication of a problem occurs when the SP reports a fault on one of the power supplies, as indicated in the ILOM event logs

-> show /SP/logs/event/list
295 Sun Dec 13 14:34:30 2015 Fault Fault critical
Fault detected at time = Sun Dec 13 14:34:30 2015. The suspect
component: /SYS/PS1 has fault.chassis.power.fail with probability=100.
Refer to http://support.oracle.com/msg/SPT-8000-MJ for details.
->

Use the Service Process Fault Management shell to confirm the fault was triggered by V_OUT_OK being de-asserted with recovery occurring within 20 seconds;

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

faultmgmtsp> fmdump -v
2015-12-13/14:34:30 89be4ec1-63a1-cf8e-ef8a-98fbbfbbc333 SPT-8000-MJ

fault = fault.chassis.power.fail@/SYS/PS1
certainty = 100.0 %
FRU = /SYS/PS1
ASRU = /SYS/PS1
resource = /SYS/PS1

faultmgmtsp> fmdump -Vu 89be4ec1-63a1-cf8e-ef8a-98fbbfbbc333
2015-12-13/14:34:30 89be4ec1-63a1-cf8e-ef8a-98fbbfbbc333 SPT-8000-MJ

timestamp ereports
2015-12-13/14:34:20 ereport.psu.ok-deasserted@/sys/ps1

fault = fault.chassis.power.fail@/SYS/PS1
certainty = 100.0 %
FRU = /SYS/PS1
ASRU = /SYS/PS1
resource = /SYS/PS1
_list_sz = 1
_list_idx = 0
_diagnosis_engine_version = 1.0
_diagnosis_engine_name = fdd
system_serial_number = AKXXXXXXX
system_part_number = 33153483+1+1
system_name = SPARC T5-2
system_manufacturer = Oracle Corporation
chassis_serial_number = AKXXXXXXX
chassis_part_number = 33153483+1+1
chassis_name = SPARC T5-2
chassis_manufacturer = Oracle Corporation
system_component_serial_number = AKXXXXXXX
system_component_part_number = 33153483+1+1
system_component_name = SPARC T5-2
system_component_manufacturer = Oracle Corporation
fru_name = PSU
fru_manufacturer = Power-One China,Gongming Town,Guangming District,518132 Shenzhen
fru_serial_number = 465776G+1412B30JM7
fru_rev_level = 01
fru_part_number = 7060596
detector = /SYS/PS1/V_OUT_OK <<<
[skipped fruid update]

faultmgmtsp> fmdump -e
2015-12-13/14:34:20 ereport.psu.ok-deasserted@/SYS/PS1
2015-12-13/14:34:31 ereport.psu.ok-asserted@/SYS/PS1 <<<
faultmgmtsp>

OR as in the case of V_IN_INFO,

58 | 07/03/2016 | 20:29:12 | Power Supply PS1/V_IN_INFO | State Asserted
59 | 07/03/2016 | 20:29:17 | Power Supply PS1/V_IN_INFO | State Deasserted
5a | 07/04/2016 | 17:26:23 | Power Supply PS0/V_IN_INFO | State Asserted
5b | 07/04/2016 | 17:26:28 | Power Supply PS0/V_IN_INFO | State Deasserted

Please note that SPT-8000-MJ is a generic fault classification that can also occur as a result of a genuine fault on the impacted power supply, such as loss inlet to the system or a component failure.

V_OUT remaining de-asserted without recovery within 20 seconds, or where there is evidence of V_IN errors, may be an indicator of a failing PSU or environmental issue. In this situation please contact Oracle Support for further assistance.

Impacted PSUs;

7060596 Power One/Bel Power A239E
7081064 Power One/Bel Power A239A

Note: ASTEC PS is NOT impacted!

Cause

A transient sensor read issue causes an erroneous fault condition.

Solution

A new firmware image for "Power One/Bel Power" A239E and A239A power supplies is available which improves error detection and handling to prevent erroneous fault generation as a result of V_OUT_OK de-assert events.

Use the following procedure to update the firmware on all appropriate power supplies (CLI Only, not supported using BUI);

NOTE: prior to exercise the steps, clear all FMA faults related to PSUs in order to avoid issue such as "embedded script execution failed".

1. For Power One/Bel Power PSU:

Download Hardware_Programmables-1.0.8-SPARC_T5-2.pkg (patchID 24668560) from My Oracle Support, unpack and make available on a system that is accessible by the Service Processor via the network.

2. Verify the target system has Power One or Bel Power A239A, A239E Power Supplies;

-> show /SYS/PS0 fru_description fru_manufacturer fru_part_number
-> show /SYS/PS1 fru_description fru_manufacturer fru_part_number

For example on Power One PSU;

-> show /SYS/PS1 fru_description fru_manufacturer fru_part_number

/SYS/PS1
Properties:
fru_description = A239E_Power_Supply
fru_manufacturer = Power-One China,Gongming Town,Guangming District,518132 Shenzhen
fru_part_number = 7060596

3. Firmware installation can be performed with the host powered on and does not require a SP reset or AC power cycle once complete;

(follow example is from Hardware_Programmables-1.0.5-SPARC_T5-2.pkg. basically, no change with version 1.0.8)

-> load -source sftp://<user>@<remote_host_ip>/<path_to_Hardware_Programmables-1.0.5-SPARC_T5-2.pkg> /SP/firmware/backupimage

The firmware upgrade procedure will take approximately 5 minutes per PSU to complete.

In the following example both PS0 and PS1 are upgraded;

-> load -source sftp://<user>@<remote_host_ip>/var/tmp/22386031/Firmware/Hardware_Programmables/Hardware_Programmables-1.0.5-SPARC_T5-2.pkg /SP/firmware/backupimage
Enter remote user password: **********

Are you sure you want to load the specified file (y/n)? y
This may take up to 20 minutes. Do you wish to continue (y/n)? y
Updating /SYS/PS0...........done
Updating /SYS/PS1...........done

Firmware update is complete.

Comment:

You can look at the /SP/logs/event/list within the SP/ILOM via 'show /SP/logs/event/list'. Look for messages regarding the PS firmware update.
Examples:
End install of Hardware_Programmables-1.0.5-SPARC_T5-2
Install of Hardware_Programmables-1.0.5-SPARC_T5-2 successful
Hardware_Programmables-1.0.5-SPARC_T5-2 applied to /SYS/PS1
Hardware_Programmables-1.0.5-SPARC_T5-2 applied to /SYS/PS0

References

<BUG:21975569> - A239 BEL-POWER PSU FIRMWARE UPDATE PACKAGE
<BUG:21129665> - T5-X EREPORT.PSU.OK-DEASSERTED / ASSERTED

Attachments

This solution has no attachment