[torqueusers] Email notification flooding

Caird, Andrew J acaird at umich.edu
Mon Sep 24 09:13:47 MDT 2007


Try setting job_nanny=true in the server.  

I think that's what fixed it for us.

> -----Original Message-----
> From: torqueusers-bounces at supercluster.org 
> [mailto:torqueusers-bounces at supercluster.org] On Behalf Of Freya Nerve
> Sent: Monday, September 24, 2007 11:13 AM
> To: torqueusers at supercluster.org
> Subject: [torqueusers] Email notification flooding
> 
> Hi All,
> 
> I have seen others having this problem when searching the archives,
> but I don't see the solution.  We have a large cluster and compute
> nodes go down and get replaced on a regular basis.  When the compute
> node that is the MOM node for a job goes down and if the owner of the
> job attempts to delete the job with qdel, that user is treated to a
> flood of thousands of emails that say something like:
> ----------------------
> Date: Mon, 24 Sep 2007 09:30:48 -0400 (EDT)
> From: root at xxxx
> Subject: PBS JOB 481960.xxxxx
> To: xxxxx at xxxx.gov
> Precedence: bulk
> 
> PBS Job Id: 481960.xxxxxx
> Job Name:   test
> Execution terminated
> Exit_status=271
> resources_used.cput=00:02:36
> resources_used.mem=199104kb
> resources_used.vmem=693496kb
> resources_used.walltime=00:04:38
> -----------------------------------
> 
> How can I stop this by default?
> I know about the "#PBS -m bae" settings, but I don't want to rely on
> my users using them.  We are spamming our own mail system.
> 
> Thanks for any help!
> Jen
> _______________________________________________
> torqueusers mailing list
> torqueusers at supercluster.org
> http://www.supercluster.org/mailman/listinfo/torqueusers
> 


More information about the torqueusers mailing list