[torquedev] pbsnodes -a shows long gone jobs

Craig Macdonald craigm at dcs.gla.ac.uk
Wed Jun 17 07:29:18 MDT 2009


Yes, we also see this in 2.1.8. They also show up in pestat by starred 
jobs. We just ignore them, the job processes arent running.

Craig

Glen Beane wrote:
> After upgrading from 2.1.x to 2.3.6, I've noticed that I see very old
> jobs listed in pbsnodes -a.   For example, the information returned
> for one of my nodes is as follows:
>
> cs-prod-14
>      state = free
>      np = 4
>      ntype = cluster
>      status = opsys=linux,uname=Linux cs-prod-14 2.6.18.2-34-default
> #1 SMP Mon Nov 27 11:46:27 UTC 2006
> x86_64,sessions=22388,nsessions=1,nusers=1,idletime=2314227,totmem=18550004kb,availmem=18312500kb,physmem=16445532kb,ncpus=4,loadave=0.00,netload=16020253916437,state=free,jobs=37005.wulfgar.jax.org
> 41061.wulfgar.jax.org 41062.wulfgar.jax.org 41310.wulfgar.jax.org
> 41313.wulfgar.jax.org,varattr=,rectime=1245244487
>
>
> You will notice the state is free, but this lists 5 jobs in the status
> string.  These jobs are long gone from the system, in the case of job
> number 37005 the job has been completed for OVER 2 MONTHS!
>
> Has anyone else seen this?
> _______________________________________________
> torquedev mailing list
> torquedev at supercluster.org
> http://www.supercluster.org/mailman/listinfo/torquedev
>   



More information about the torquedev mailing list