[Linux-cluster] Cluster is slow

[Abhra Paul]

On Feb 5, 2008 11:43 AM, jim parsons <jparsons at redhat.com> wrote:
>
> On Tue, 2008-02-05 at 11:33 +0530, Abhra Paul wrote:
> > Hi,
> >
> >        I am using a cluster of 10 nodes. In node7,8,9,10  dirac is
> > running for 4 days. This dirac is also required huge amount of space.
> > Now when I am visiting individual node and issuing top ,o/p is not
> > showing and cursor is only blinking without any message or anything.
> > How I can solve this problem.
> >
> Can you provide some ancillary info, such as what version of cluster
> software you are running on what hardware? Could you post your
> cluster.conf file? Is this a new cluster, or has this cluster been
> around for awhile working OK and this is a sudden problem?
>
> -J
>
> --
> Linux-cluster mailing list
> Linux-cluster at redhat.com
> https://www.redhat.com/mailman/listinfo/linux-cluster
>
-------------------------------------------------------------------------------------------------------------
As a cluster software I am using PBS package(pbs_server,
pbs_mom,pbs_schedl), and to submit the job I am using Moab Access
Potral job schedular. But I dont know the version of those clustering
software. Please tell me how I find the version. In my system there is
no cluster.conf file, I have searched by whereis command. This cluster
is approx 9 months old. This problem I am facing first time.

H/W configuration is as follows:
-- Total 12 nodes, one is master and others are slave.
-- I am using Quad-core intel processor
-- In slave node there is 12G Ram , in master 24G.
-- Infinyban , GHz switch

It is a BeoWoulf cluster. Also tell me where I can find the cluster
configuration file.