SYS_RESET in the errpt

[jpn1]

We had a box go offline this morning and this is all I see in the errrpt. I'm researching with IBM now, but always like to hear other real world scenarios and discussions.

LABEL: SYS_RESET
IDENTIFIER: 1104AA28

Date/Time: Fri Oct 19 04:48:49
Sequence Number: 72968
Machine Id: 0006761F4C00
Node Id: localhost
Class: S
Type: TEMP
Resource Name: SYSPROC

Description
SYSTEM RESET INTERRUPT RECEIVED

Probable Causes
SYSTEM RESET INTERRUPT

Detail Data
KEY MODE SWITCH POSITION AT BOOT TIME
normal
KEY MODE SWITCH POSITION CURRENTLY
normal

 

[RMGBELGIUM]

hmz, this means that someone pressed the magic reset button ...

rgds,

R.

 

[jpn1]

Here is a snip of the errpt. Why does the SYS_RESET showup in the middle of these other entries with a different time?
Is there a different clock used to create the timestamp when the reset button is pushed?

813FE820 1019081007 U S MQSeries SOFTWARE PROGRAM ERROR
0BA49C99 1019080507 T H scsi0 SCSI BUS ERROR
BD797922 1019080507 P H enclosure1 SUBSYSTEM FAILURE
BD797922 1019080507 P H enclosure0 SUBSYSTEM FAILURE
071F4755 1019080407 P H sysplanar0 ENVIRONMENTAL PROBLEM
071F4755 1019080307 P H sysplanar0 ENVIRONMENTAL PROBLEM
1104AA28 1019044807 T S SYSPROC SYSTEM RESET INTERRUPT RECEIVED
9DBCFDEE 1019080507 T O errdemon ERROR LOGGING TURNED ON
BD797922 1019040007 P H enclosure1 SUBSYSTEM FAILURE

 

[p5wizard]

Because such events are first stored in NVRAM - the system does not have time to write it to disk when someone RESETs the machine...

Then on next reboot, somewhere during initialization, NVRAM log entries are read and written to the error log.


HTH,

p5wizard

 

[jpn1]

The system stopped reponding to network pings at 5am and then the reset button was pushed at 0805 by another person. But when I look at this log it tells me the reset button was pushed at 0448.

So if the reset button was pushed at 0448, but it doesn't get writen to the log until the next reboot, which would have happened by pressing the reset button at 0805. The I should see this in the log again when the next reboot occurs.
I'm trying to figure out if there really was someone in the datacenter at 0448 pressing the button. Or if my clocks are not in sync. Security is checking the card reader logs now, I just don't believe it.

 

[p5wizard]

Well, does errpt -a give you more detail? I'm not sure the reset-button is the only source of such events.

Another possibility is: (never seen this before though, and I think it is unlikely) the system "hung" itself alltogether and the system's software clock stopped ticking, hence the "old" timestamp when the button was pressed at 8:05 ?
Then on reboot, the RTC-chip time is copied to the software clock again...


HTH,

p5wizard

 

[jpn1]

here is the errpt -a:
---------------------------------------------------------------------------
LABEL: SYS_RESET
IDENTIFIER: 1104AA28

Date/Time: Fri Oct 19 04:48:49
Sequence Number: 72968
Machine Id: 0006761F4C00
Node Id: localhost
Class: S
Type: TEMP
Resource Name: SYSPROC

Description
SYSTEM RESET INTERRUPT RECEIVED

Probable Causes
SYSTEM RESET INTERRUPT

Detail Data
KEY MODE SWITCH POSITION AT BOOT TIME
normal
KEY MODE SWITCH POSITION CURRENTLY
normal
---------------------------------------------------------------------------
LABEL: ERRLOG_ON
IDENTIFIER: 9DBCFDEE

Date/Time: Fri Oct 19 08:05:49
Sequence Number: 72967
Machine Id: 0006761F4C00
Node Id: localhost
Class: O
Type: TEMP
Resource Name: errdemon

Description
ERROR LOGGING TURNED ON

Probable Causes
ERRDEMON STARTED AUTOMATICALLY

User Causes
/USR/LIB/ERRDEMON COMMAND

Recommended Actions
NONE

---------------------------------------------------------------------------