DL380 random reboots

[Jean-Michel Daviault]

I own 10 of these servers, all running Linux. in a rack mount and lately a few of them started doing random reboots.

I tought it was ram so I downloaded memtest86 and ran the ram test on it and it froze after 2sec of test, I was pretty sure it was a ram problem until I noticed if I downgraded the memtest to 3.4 instead of 3.5 the test did a full pass without problem. I also noticed that the 3.5 would do that only if I had more then 2 sticks of RAM.

I'm pretty sure I had the latest BIOS version I could find, ILO 1.92 and smart array 5i 2.72

It also randomly reboots when I try to run test with the smartstart CD, especially the ram test.

All my ram is same brand (PC2100 DDR 266 ECC), I tried multiple combination of sticks, I got over 20 of them so I higly doubt its a stick problem... what other then this could cause such a problem. I'd hardly beleive its the sockets as we got 2 servers so far that we noticed were doing that.

I'd appreciate any idea towards this as I am running out of idea.

thank you

[Jean-Michel Daviault]

forgot to say its a DL380 G3 server

[Pieter 't Hart]

did you look at the "integrated management log" from the ILO page for clueues?