Urgent! Proliant ML350G5 resets random (daily up to every 5 minutes)

[Mark Verheijen]

Hello,

I have a ML350 running for 2½ years now. It suddenly reset and a week later again. The PSU is replaced and everything was fine for 2 weeks then the resets started again. After replacement of the mainboard it was fine for 1½ day....now still resets.

I was able to exchange the Xeon QC and memory, disconnected the streamer with SCSI PCI, cdrom. So the only thing not exchanged is the 8 x 2,5" drive cage. It contains a little electronics could it be possible this is the cause? It isn't software related because it will even reset during the BIOS (infact one time before any firmware was loaded)...

Anyone had this before? It's an production server so the client loses his patience in not a long time from now :(

[Glenn Matthys]

Can you disable the watchdog in the BIOS setup?

[Navid Baradaran]

If you can post the latest NMI events or IML event of the system, we may better investigate the problem.
You may access IML log via iLO, configure your iLO and open it from a browser to see its content.

[Graeme Bray]

What is the iLo2 firmware on the server as well as the BIOS level?

If you're saying its resetting its possible that the iLo2 firmware needs to be updated to 1.78 as well as the BIOS to HP ProLiant ML350 G5 (D21) Servers Version 2009.05.12

Also, does this server have an e200i controller? If so, update that firmware to at least 1.82 (I have not tested 1.84 so I cannot speak for it).

[Johan Guldmyr]

Do you have a DVD reader connected to it?
Might be that one.

Also, do you see any error leds on the server or on the system board?

[Mark Verheijen]

I have run the latest firmware CD. At the moment we have reached level 2 of HP support and are waiting for someone to arrive with spareparts, which are not in stock! It seems like a hardware component is making short-circuit now and then. We try'd to clone the server to a backup server yesterday but even outside the OS (bootable USB stick with clone software) we weren't able to clone the server succesfully because off the resets. We have to wait for the spareparts and in the meantime try to backup as often as posssible and keep our fingers crossed...

I must say i'm a bit dissapointed in HP, all of their carepacks with X hours / days response says nothing about the availability off the needed parts. At the moment i have a ML150G6 that is waiting on a mainboard exchange for a week now... We are active in the SMB market so 95% is working with only 1 server which is a disaster when it breakdown...

TBC

Mark