HP ML370 G3 hangs after several hours - hardware ?

[Richard Ledoigt]

I have 3 ML370 G3 servers in my lab.
One of them is acting weird: It hangs after several hours, no warning given, it just freezes.
I had the problem while running RHEL ES 4 Update 3 so I tried with memtest86+ and it hangs just the same. 1st wall time was 22 hrs, 2nd was 230 hrs and 3rd was 23 hrs.
Since memtest does not rely on disks or RAID controllers, I think my problem may be related to the hardware or BIOS.
The BIOS is unlikely because the other two servers run the same P28 dated 09/15/2004 without problem...

Server information: ML370 G3 Dual Xeon 2.8GHz@400MHz processors, 2048KB ECC RAM, Serial # 8057LDQ11D, HT activated, CPU performance optimized for I/O.
When the server freezes, with memtest86+ or RHEL, there is no mention whatsoever of an incident or a warning before it happens, the logs don't give any clue away.

Could you please help me and suggest the next step to troubleshoot my problem ? Thanks a bunch.

[Rob Leadbeater]

Hi Richard,

Have you looked in the IML (Integrated Management Logs) to see if that shows anything interesting ?

As you're running Linux, I'll guess that you haven't got a Proliant Support Pack installed. Grab it from this page:

http://h18023.www1.hp.com/support/files/server/us/locate/1116_4307.html

Once that's installed, you should be able to get to the Systems Management homepage for the server at:

https://:2381

You'll be able to see the various system logs from there...

Cheers,

Rob