A500 processor failed

[Andres Valentin Lopez]

Hi Guys...

Recently I had the necessity to expand the /var file system in my HP9000 A500, I did it without any problem, but when the server re-start the next message appears

DH test 3160
platform test 616A
processor test 1146
processor INIT 1701
unknown, no source stated machine check 0624
processor machine check 1632

NEXT

SYSTEM NAME: adaia
DATE: 29/10/2005 TIME: 02:40:29
ALERT LEVEL: 2 = Non-Urgent Operator attention requierd
SOURCE: 0 = unknown, no source stated
SOURCE DETAIL: 0 = unknown, no source stated SOURCE ID: 0
PROBLEM DETAIL: 5 = unexpected
CALLER ACTIVITY : 0 = unspecified STATUS : 0
CALLER ACTIVITY : 00 = implementation dependent
REPORTING ENTITY TYPE : 0 = system firmware REPORTING ENTITY ID = 00

I cancel the boot and the processor summary appears

proccessor 0 440 Mhz stopped:unknown
proccessor 1 440 Mhz OK

the server start and is ok, but I do not know what happend if the processor 0 is stopped, and I need to know how to do to solve this problem....

Thanks a lot for your recomendations....

[Sameer_Nirmal]

Hi,
As indicated the event has been reported by system firmware.
The event happened during processor testing during startup , which caused
HPMC with processor 0 making it stopped. I guess the event details given by you are incomplete as there is no chassis code which should be there below "Reporting activity"

I would advise to raise a H/W call to HP and provide them the system event logs SL from MP menu.
Be aware you need to provide all those events not just the one.
Moreover you need to provide a copy /var/tombstones/ts99 & ts98
/var/adm/syslog/OLDsyslog.log
/var/adm/syslog/syslog.log
dmesg

Is it the currnet status of processor 0 is stopped?
Do you see the processor in sar -uM or in Glance?

[Bjoern Gothe_1]

Hi,

I agree to the advice open a HW case at your local HP Support Center. They need the HPMC (/var/tombstones/ts99 file) and the last relevant error chassis codes from GSP --> SL --> view error logs E.

The most probable cause is the processor itself, but also the power pods of CPU can be affected.
You can do "ioscan -fnC processor" on HP-UX to check how many processors are in use. If one is missing means that this one was deconfigured by BCH.

Kindest regards, Bjoern Gothe

[Andres Valentin Lopez]

Thanks ...

I do not found out in the server the file var/tombstones/ts99 & ts98 and the result of the command ioscan -fnC processor is
Class I H/W Path Driver S/W State H/W Type Description
===================================================================
processor 0 162 processor CLAIMED PROCESSOR Processor

indicates that I have one proccesor online. And the result of the command sar -uM is sar: Can't open /var/adm/sa/sa03

Thanks again....

[Sameer_Nirmal]

I did suspect that one processor won't be there in OS environment.

I would do the shutdown of the system and have a look at the event logs in MP and capture it using MP -> SL -> D
Take a look at Console log as well
MP -> CL

I would do the total power down of the system ( pulling out the power cord ) and start it again and see the status of the processors again.

If no success , raise a h/w call to HP.