On 1 Feb 2012, at 11:05, Reuti wrote: > Am 01.02.2012 um 11:21 schrieb Hugh Macdonald: > >> Thanks Ben and Reuti, >> >> I think (and I was reading up more on this after sending my initial email) >> that using prologs and epilogs will actually do exactly what I'm after - >> it's not so much specific job cleanup, as sending messages for job >> start/finish, which needs to happen on absolutely everything. >> >> >> One question I did have about prolog/epilog scripts was how they deal with >> exit codes... The prolog script would, on the whole, exit with a code of 0, >> but how about the epilog? If the main job script exits with, say, 99, will >> the epilog run, and should that also return 99, or will Grid see the job >> script's 99 exit code, and still reschedule, even though the epilog script >> returns 0? > > It will also run, and could even exit with 0. The exit code of any of them > will be taken to reschedule the job or put it into error state. > > job 0, epilog 99 => reschedule but exit code 0 (i.e. not easy to spot why it > was rescheduled) > job 99, epilog 0 => reschedule but exit code 99 > > You will get multiple entries in the accounting file for such a job (same for > jobs in error state, when you remove the error with `qmod -cj <wc_job>` to > let it run again).
Great - thanks for that - that sounds fine - I would imagine that my prolog and epilog scripts would always return 0, and the main job scripts are already returning the correct thing. Thanks Hugh Macdonald nvizible – VISUAL EFFECTS [email protected] +44(0) 20 3167 3860 +44(0) 7773 764 708 www.nvizible.com > -- Reuti > > >> Hugh Macdonald >> nvizible – VISUAL EFFECTS >> >> [email protected] >> +44(0) 20 3167 3860 >> +44(0) 7773 764 708 >> >> www.nvizible.com >> >> On 31/01/12 23:27, Reuti wrote: >>> Am 31.01.2012 um 23:10 schrieb Ben De Luca: >>> >>> >>>> http://gridscheduler.sourceforge.net/htmlman/htmlman5/queue_conf.html >>>> >>>> terminate_method >>>> >>> This will be used in case you issue `qdel`. >>> >>> Not for a normal termination. >>> >>> -- Reuti >>> >>> >>> >>>> looks like it might be simplest. >>>> >>>> >>>> >>>> On Tue, Jan 31, 2012 at 9:51 PM, Ben De Luca >>>> <[email protected]> >>>> wrote: >>>> >>>>> Strangely im pondering this issue at the moment. If a python process >>>>> is killed any process started with subprocess does not die. >>>>> >>>>> The two methods im following, >>>>> >>>>> 1. a reaper, the grid job starts a python process (parent) that starts >>>>> two other jobs, the task, and the reaper. >>>>> A. The task is the process I want to run. >>>>> B. The reaper, sits waiting to see if the parent python process >>>>> disappears, if it does it cleans up process A. >>>>> >>>>> When Parent sees A finishes normally it kills B. >>>>> If Parent is killed B kills A. >>>>> >>>>> 2. I think an epilog can be run to clean up. >>>>> >>>>> >>>>> On Tue, Jan 31, 2012 at 8:00 PM, Hugh Macdonald >>>>> >>>>> <[email protected]> >>>>> wrote: >>>>> >>>>>> Hi, >>>>>> >>>>>> I've got a python script here that is run through Grid. When I requeue >>>>>> the >>>>>> job, I want to catch the signal in the script, to do a bit of cleanup >>>>>> before >>>>>> exiting. I don't seem to be able to catch the signal, making me assume >>>>>> that >>>>>> it's SIGKILL that is being sent. Am I right in this assumption, and, if >>>>>> so, >>>>>> is there any way to change it to SIGTERM? >>>>>> >>>>>> >>>>>> Thanks >>>>>> >>>>>> >>>>>> Hugh Macdonald >>>>>> nvizible – VISUAL EFFECTS >>>>>> >>>>>> >>>>>> [email protected] >>>>>> >>>>>> +44(0) 20 3167 3860 >>>>>> +44(0) 7773 764 708 >>>>>> >>>>>> >>>>>> www.nvizible.com >>>>>> >>>>>> >>>>>> >>>>>> _______________________________________________ >>>>>> users mailing list >>>>>> >>>>>> [email protected] >>>>>> https://gridengine.org/mailman/listinfo/users >>>>>> >>>>>> >>>>>> >>>> _______________________________________________ >>>> users mailing list >>>> >>>> [email protected] >>>> https://gridengine.org/mailman/listinfo/users >>>> >>>> >>>> >> _______________________________________________ >> users mailing list >> [email protected] >> https://gridengine.org/mailman/listinfo/users >
_______________________________________________ users mailing list [email protected] https://gridengine.org/mailman/listinfo/users
