[torqueusers] how to browse the stdout and stderr files of a running batch job

Gustavo Correa gus at ldeo.columbia.edu
Fri Nov 2 12:05:51 MDT 2012


Hi Maura

As an alternative while you are dubugging your code, 
you could also launch your jobs in interactive mode, with the -I (capital "i") switch
(unless your site restricts the use of interactive mode).
Say (first switch is letter "i", the second is letter "L", see 'man qsub' for details):

qsub -I -l nodes=2:ppn=8

This would give you two 8-processor nodes and 
open an interactive session on one of those nodes for you.
You may need to wait until the nodes become available.

From there you could launch your program, say:

cd /my/work/directory
mpiexec -np 16 ./my_parallel_monte_carlo_program

and watch its stderr and stdout on the terminal window,
or redirect them to a file and "more" or "tail -f " that file.

I hope this helps,
Gus Correa

On Nov 2, 2012, at 11:44 AM, Burkhard Bunk wrote:

> Hi,
> 
> you can redirect stdout and stderr files to your HOME directory with
> 
> 	qsub -k oe ...
> 
> and watch them during job execution.
> 
> Regards,
> Burkhard Bunk.
> ----------------------------------------------------------------------
> bunk at physik.hu-berlin.de      Physics Institute, Humboldt University
> fax:    ++49-30 2093 7628     Newtonstr. 15
> phone:  ++49-30 2093 7980     12489 Berlin, Germany
> ----------------------------------------------------------------------
> 
> On Thu, 1 Nov 2012, mauede at alice.it wrote:
> 
>> I submit long Monte Carlo simulations to the PBS scheduler. I redirect each
>> simulation stdout and stderr to a uniquely named file. Unluckily, as far as I
>> know, PBS does not allow me to peep at such a file before the simulation is
>> finished, whether it has completed OK or has aborted.
>> Some super-computer centers have developed commands, like qcat, qpeep, that
>> allow for monitoring number-cruncher programs and browse their stdout and
>> stderr. This feature helps saving a lot of CPU time when the submitted jobs
>> are not doing what is expected.
>> Is there a free-ware implementation of qcat or qpeep  from some websites ?
>> Thank you in advance.
>> Regards,
>> maura
> _______________________________________________
> torqueusers mailing list
> torqueusers at supercluster.org
> http://www.supercluster.org/mailman/listinfo/torqueusers



More information about the torqueusers mailing list