PBS Forum Has Closed 06/12/17The PBS Works Support Forum is no longer active. For PBS community-oriented questions and support, please join the discussion at http://community.pbspro.org. Any new security advisories related to commercially-licensed products will be posted in the PBS User Area (https://secure.altair.com/UserArea/).
Search the Community
Showing results for tags 'rerun'.
Found 1 result
NNN posted a topic in TroubleshootingHi guys, I'm rather new to PBS but I'm dealing with a stable PBS deployment in a private company. Here job submission is done exclusively from the Compute Manager which recalls PAS that is installed upon PBS Pro 12.2. The only available queue is workq. What we want to do is to check, for a specific PAS application, if a job has failed due to a specific software reason, by testing the output or error log of the software (which reduces to parsing a text file) and, in that case, re-queue the job in workq so that it will be the first one to be run when resources available. No delays or modifications to PAS resources are needed; just requeue (o rerun). The whole thing should be transparent to the users. I guess that something similar could be done by writing a hook to be triggered in the execjob_end phase. Am I right? Is this feasible? Is there any other simpler way to do it? We currently don't have any hooks enabled and so we lack of experience with them. Our deployment and workflow is rather "pas_application centric". All the logic and rationales are implemented inside the strat.py and presubmit.py scripts in PAS. Thanks a lot for your help NN