Sun SPARC(R) Enterprise M3000/M4000/M5000/M8000/M9000 Server: XSCF Email Reporting does not cover power and temperature events

Asset ID:	1-72-1330510.1
Update Date:	2018-03-01
Keywords:

Solution Type Problem Resolution Sure

Solution 1330510.1 : Sun SPARC(R) Enterprise M3000/M4000/M5000/M8000/M9000 Server: XSCF Email Reporting does not cover power and temperature events

Applies to:

Sun SPARC Enterprise M9000-32 Server - Version Not Applicable and later
Sun SPARC Enterprise M9000-64 Server - Version Not Applicable and later
Sun SPARC Enterprise M8000 Server - Version Not Applicable and later
Sun SPARC Enterprise M5000 Server - Version Not Applicable and later
Sun SPARC Enterprise M4000 Server - Version Not Applicable and later
Information in this document applies to any platform.

Symptoms

Email reporting is configured and was tested successfully.
The system is experiencing environmental warnings (temperature messages are self explaining, power messages show AC input power disappearing from PSU and AC input power being restored):

XSCF> showlogs event
May 09 12:26:47 CEST 2011 High temperature at air inlet. FANs are changed to high speed mode
May 09 17:48:27 CEST 2011 Air inlet temperature recovered from warning state
May 13 08:27:24 CEST 2011 High temperature at air inlet. FANs are changed to high speed mode
May 15 08:36:46 CEST 2011 Air inlet temperature recovered from warning state
May 15 08:54:26 CEST 2011 High temperature at air inlet. FANs are changed to high speed mode
May 15 09:17:44 CEST 2011 Air inlet temperature recovered from warning state
May 24 16:06:45 CEST 2011 High temperature at air inlet. FANs are changed to high speed mode
May 24 20:28:44 CEST 2011 Air inlet temperature recovered from warning state
Jun 05 02:01:40 CEST 2011 Power failure (/PSU#2)
Jun 05 02:01:40 CEST 2011 Power failure (/PSU#0)
Jun 05 02:21:36 CEST 2011 Power recovery start(/PSU#2)
Jun 05 02:21:37 CEST 2011 Power recovery start(/PSU#0)
Jun 05 02:22:00 CEST 2011 Power recovery complete(/PSU#2)
Jun 05 02:22:01 CEST 2011 Power recovery complete(/PSU#0)
Jun 05 05:48:48 CEST 2011 Power failure (/PSU#2)
Jun 05 05:48:48 CEST 2011 Power failure (/PSU#0)
Jun 05 06:20:41 CEST 2011 Power recovery start(/PSU#2)
Jun 05 06:20:42 CEST 2011 Power recovery start(/PSU#0)
Jun 05 06:21:05 CEST 2011 Power recovery complete(/PSU#2)
Jun 05 06:21:06 CEST 2011 Power recovery complete(/PSU#0)

But no email report is generated.

Changes

Cause

There were no FMA events recorded for that period of time:

$ cat @opt@sun@fm@sbin@fmdump_-e.out
TIME CLASS
...
Oct 10 17:47:09.9197 ereport.chassis.domain.panic
Oct 14 19:30:08.1165 ereport.chassis.domain.panic
Apr 02 21:15:28.1289 ereport.chassis.domain.panic

Because no FRU had a problem, no FMA ereport was generated, and therefore no email was sent.

Some background information can be found in SUNBT6743183 "OPL SC FMA DE fails to report expected ereport, fault-event in response to the power failure event."

A Fujitsu engineer writes:
...
It seems we misinformed Sun about behaviour of power failure event.

Current implementation is that there is no ereport/fault event for input power failure, instead of ereport/fault event we logs event logs and make SNMP trap as a notification.

Since input power failure is not defect of FRU, we do not issue fault event.
...

Solution

Install a system management system like Oracle Enterprise Manager.

Attachments

This solution has no attachment