[torqueusers] Fwd: job dieing immediately, 0 byte output file being produced

Sabuj Pattanayek sabujp at gmail.com
Tue Feb 23 16:50:46 MST 2010


Nice, so my next question is whether it's safe to upgrade from 2.4.3
to 2.4.5 with jobs running, i.e. the binaries will get replaced after
I do make install. This is usually how I restart pbs_server and maui
after config changes while keeping jobs running:

qterm -t quick pbsserver
schedctl -k
sleep 5

on the nodes:

killall -2 pbs_mom
pbs_mom -p

Will running jobs be ok?

Sabuj Pattanayek

---------- Forwarded message ----------
From: David Beer <dbeer at adaptivecomputing.com>
Date: Tue, Feb 23, 2010 at 5:26 PM
Subject: Re: [torqueusers] job dieing immediately,	0 byte output file
being produced
To: Sabuj Pattanayek <sabujp at gmail.com>

If you are running a more recent version of TORQUE (2.4.4 or later)
then you can set

$spool_as_final_name true

in your mom config. Obviously this will be very slow if you don't have
a fast connection to the output location, but I'm assuming that isn't
your situation since you are looking for this.

----- "Sabuj Pattanayek" <sabujp at gmail.com> wrote:

> Is there anyway to get the output that usually goes to
> /var/spool/torque/spool to go directly into the output file specified
> by the user in #PBS -o ? Or if the user doesn't specify a file, into
> the current working directory of the .pbs file as
> $JOB_ID.serverName.OU ?
> _______________________________________________
> torqueusers mailing list
> torqueusers at supercluster.org
> http://www.supercluster.org/mailman/listinfo/torqueusers

David Beer | Senior Software Engineer
Adaptive Computing

More information about the torqueusers mailing list