[torqueusers] Problems with queueing

David Chin david.w.h.chin at gmail.com
Thu Jan 11 16:05:50 MST 2007


On 1/11/07, Jerry Smith <jdsmit at sandia.gov> wrote:
> You may want to try only having the following 2 lines in your config
> $pbsserver master.dfci.harvard.edu  ##is this the internal address name?
> $usecp *.dfci.harvard.edu:/home /home

Yes: master.dfci.harvard.edu is the internal address.

>[snip]
>
> Everything else looks ok.  What does your $PBS_HOME/server_priv/nodes
> look like?


    master:ts np=1 headnode
    n01 np=1 odd
    n02 np=1 even
    n03 np=1 odd
    n04 np=1 even
    n05 np=1 odd
    n06 np=1 even
    n07 np=1 odd
    n08 np=1 even
    n09 np=1 odd
    n10 np=1 even
    n11 np=1 odd
    n12 np=1 even
    n13 np=1 odd
    n14 np=1 even
    n15 np=1 odd
    n16 np=1 even
    n18 np=1 even

I just tried it with the changes you suggested, submitting trivial "sleep 30"
to the queue while logged in as myself on the headnode. All jobs still go to
the headnode, with job IDs that look like 73.jvneumann.dfci.harvard.edu
(jvneumann is its external name).

Thanks for looking at this,
  Dave

-- 
Email: david.w.h.chin at gmail.com    dwchin at lroc.harvard.edu
Public key: http://gallatin.physics.lsa.umich.edu/~dwchin/crypto.html
      pub   1024D/1C557DDF 2006-07-21 [expires: 2007-07-21]
      Key fingerprint = 4EEB A409 5010 3679 4EA7  D420 4E52 202A 1C55 7DDF


More information about the torqueusers mailing list