[Aces-support] jobs in queue not running

[Florian Bleibinhaus]

Hello Greg,

Sorry, but I don't know what "headnodes" are.
I start the jobs with
#PBS -l nodes=1:ppn=1,mem=2GB,walltime=24:00:00
I used a template for the job from your webpage, and modified it.
I just submitted some test jobs called "FWI_t1" (68801) and "FWI_t2" (68802), and as you can see, the second one is still in the queue, although I submitted them almost simultaneously.
The corresponding scripts "qinv_test1" and "qinv_test2" are in my home-dir.

Florian
  ----- Original Message ----- 
  From: aces-admin at techsquare.com 
  To: ACES-support at mitgcm.org 
  Sent: Monday, August 27, 2007 11:17 AM
  Subject: Re: [Aces-support] jobs in queue not running


  hello bleibi-

  can you give a specific jobid as an example,
  or at least spec which of the three (3) headnodes
  you were using ? 

  [greg]

  > From: "Florian Bleibinhaus" <bleibi at mit.edu>
  > Date: Mon, 27 Aug 2007 09:53:38 -0400
  > MIME-Version: 1.0
  > Reply-To: ACES-support at mitgcm.org
  > 
  > Hi,
  > I noticed that some of my jobs in the long queue stay there for a very long time (several hours) before they start running, but only when there are already jobs running. Does anybody know if this might be related to requesting 2GB memory for the jobs, or what could it be?
  > 
  > Thanks,
  > Florian
  > _______________________________________________
  > Aces-support mailing list
  > Aces-support at acesgrid.org
  > http://acesgrid.org/mailman/listinfo/aces-support
  _______________________________________________
  Aces-support mailing list
  Aces-support at acesgrid.org
  http://acesgrid.org/mailman/listinfo/aces-support
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://forge.csail.mit.edu/pipermail/aces-support/attachments/20070827/86ce7891/attachment.html