script to check for system board/CPU failure

[Arun Kumar Rajamari]

Hi,

In one of our servers while restarting there was System Board failure.
While we were rebooting we could not get noticed one of the CPU failures.
Is there any existing script(command, which we can make into script) which monitors for SB failure checking for numbers of running CPU against expected?
We need to integrate with HP OVO either this failure logging an error for the respective component in any of the log files(using logfile template of HP OVO unix) or any script(command, monitor template of HP OVO Unix) to check for the availability as above.

Thanks,
Arun

[Steven E. Protter]

Shalom Arun,

If your CPU fails while the system is running you will get a High Priority Machine Check HPMC, your system will crash, even if you lose one of eight CPU's.

You don't need a script to detect that, its obvious and will be logged in your GSP card.

There are a number of good scripts that meet these needs in the following thread:

http://forums12.itrc.hp.com/service/forums/questionanswer.do?threadId=178514

There are two other threads and hundreds of scripts in there.

SEP

[Tim Nelson]

Have you looked at EMS ? This is what it is there for.

Can be reviewd by /etc/opt/resmon/lbin/monconfig

Can even send trap events to the server of your choice.