Re: [nottingham] Clusters...

From: Tom Allender (nottingham-lug@sminkybang.com)
Date: Mon 08 Jul 2002 - 14:26:12 BST


On Mon, 2002-07-08 at 14:15, Ben Blackburne wrote:
> I'm at the Computational Chemistry unit at the University of Nottingham.
> We have just installed a 48-node dual Athlon cluster. Its running Redhat
> with the SCore software (http://pccluster.org/) and Sun Grid Engine. 16
> nodes have myrinet, the rest have only ethernet.

Very cool. I nearly took Computational Chem. at Nottingham...
 
> It is however, at the moment, remarkably unstable :-(. The main node
> crashes every so often with no messages in the log. The suspects include
> dodgy hardware and not enough cooling. The hardware of the main node has
> just been replaced so we should know soon if that was the problem.

Naive shot in the dark. Is it Red Hat 7.3 with 2.4.18-3? There was an
errata kernel released as 2.4.18-3 had a repeatable panic with ext3 on
SMP systems.

Tom.
--------------------------------------------------------------------
http://www.lug.org.uk http://www.linuxportal.co.uk
http://www.linuxjob.co.uk http://www.linuxshop.co.uk
--------------------------------------------------------------------



This archive was generated by hypermail 2.1.3 : Mon 08 Jul 2002 - 14:26:28 BST