I've used something similar under Condor-G which doesn't( didn't
?) have
this functionality. I create a time stamp file before the start
of
application then after running the app use
$ find -newer time_stamp_file_name
to work out which files have been created/modified and transfer them
back. If
you use DAGman then a POST script could automatically unzip/untar
the output on the
submit host to make things easier.
-ian.
From:
condor-users-bounces@xxxxxxxxxxx [mailto:condor-users-bounces@xxxxxxxxxxx] On Behalf Of carl langlois
Sent: 04 February 2008 14:57
To: Condor-Users Mail List
Subject: Re: [Condor-users]
retrieving output subdirectories
Hi Javier,
I have the same problem as you, what i'm planning to do is wrap the
application in a .bat file or .sh if on unix and after the jobs is complete
juste zip everything and tell condor to retrived the zip file.
On Feb 4, 2008 5:37 AM, Javier Forment <jforment@xxxxxxxxxxxx> wrote:
Hi, all... I'm trying to run an application in Condor which creates an
output subdirectory and write output files on it. I know that Condor has
the feature of retrieving (and sending back to submit node) only the
output files created/modified in the plain execute directory at the
corresponding execute node.
Is there any way to cope with this? I've browsed the archives at the
mailing list, but I only find messages telling that it is a well-known
issue, and the solution is to modify the application or writing a custom
wrapper for condor to tar (and maybe gzip) the output subdirectory/ies.
Since I cannot modify the application, where can I find documentation to
write such a custrom wrapper for condor?
Thanks a lot,
Javier.
_______________________________________________
Condor-users mailing list
To unsubscribe, send a message to condor-users-request@xxxxxxxxxxx
with a
subject: Unsubscribe
You can also unsubscribe by visiting
https://lists.cs.wisc.edu/mailman/listinfo/condor-users
The archives can be found at:
https://lists.cs.wisc.edu/archive/condor-users/