|
Hi Everyone
This particular problem is causing me some hassle, I'm sure it's hardware rather than software related.
The server will run full tilt at ~100% over all cores for a couple of weeks on a particular task (statistical genetics) Then when the box is almost idle will keel over with MBE errors - Dell Poweredge 2970, 64GB ram RHEL 5.3 all bang up to date.
I'm trying to get a feel for what's happening just before it pegs out as it doesn't log anything just crashes the box - I was thinking something along the lines of top at regular intervals to try and narrow it down and I'd be grateful of any input from anyone.
Thanks Bryan
|