[torqueusers] Nodes that pbs reports are busy which are actually running a job

Rahul Nabar rpnabar at gmail.com
Thu Aug 12 13:37:15 MDT 2010


On Thu, Aug 12, 2010 at 2:32 PM, Coyle, James J [ITACD] <jjc at iastate.edu> wrote:
>  The check to see if the node is dedicated is simply a count of the number of
> times the node is comntained in $PBS_NODEFILE.  If that is the same as np
> for that node, the node is dediacted to the batch jobs. In that case it is
> OK to kill runaway processes.  I also call node_cleanup from the prologue, in case
> errant processes were left over from a previous non-dedicated job.

Thanks! A very prudent check indeed! I'll make sure I check for this
before I issue my pkill.

One of the main reasons I had decided to do dedicated nodes on this
latest cluster was the ease of killing rogue processes. Just was lucky
that I didn't have so many so far.

-- 
Rahul


More information about the torqueusers mailing list