[GE users] GridEngine and RHEL/Centos

Sean Dilda agrajag at dragaera.net
Thu Oct 14 14:39:52 BST 2004


On Mon, 2004-10-11 at 21:47, Geoff Galitz wrote:
> G'day.
> 
> I'm having a little trouble getting Gridengine 5.3 running
> properly on a cluster using Centos 3.3 (a rebuilt Redhat
> Enterprise 3 distro).
> 
> Everything appears okay, but my jobs hang in the queue
> with the gridengine complaining that node is overloaded,
> but there is nothing (unusual) running on the node and
> the load is 0.
> 
> Has anyone else run into similar problems?

I'm running SGE 6.0u1 on CentOS 3.3 and haven't noticed problems like
that.  What does SGE report the load of that node at?  (qhost -h
<node>)  And what do you have your load thresholds set to? (qconf -sq
<queue-name>|grep load_threshold)

If you're interested in trying 6.0u1 instead of 5.3, I have some rpms up
at http://www.linux.duke.edu/~agrajag/RPMS/


---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe at gridengine.sunsource.net
For additional commands, e-mail: users-help at gridengine.sunsource.net




More information about the gridengine-users mailing list