Reuti, Thanks that's exactly the issue. I just remembered I had:
set -e and then ran grep which found a result, returned non-zero, and caused the script to exit. -Bill On Jun 27, 2011, at 12:01 PM, Reuti wrote: > Hi, > > Am 27.06.2011 um 20:47 schrieb William Deegan: > >> Greetings, >> >> I have a short running job, which doesn't exceed memory or runtime limits >> specified. >> Here's the qacct output: >> qname all.q >> hostname a13.company.com >> group contr >> owner user123 >> project NONE >> department defaultdepartment >> jobname veri_MM_11.1_82 >> jobnumber 18323 >> taskid undefined >> account sge >> priority 0 >> qsub_time Mon Jun 27 11:30:58 2011 >> start_time Mon Jun 27 11:31:48 2011 >> end_time Mon Jun 27 11:33:05 2011 >> granted_pe NONE >> slots 1 >> failed 0 >> exit_status 1 > > it's the result of your job. Something like: > > #!/bin/sh > exit 1 > > will produce it. From SGE's point of view the job ran successfully. Whether > there is any application error because of wrong input data or whatever it > can't decide. > > NB: There are two special error codes which the user can use to trigger a > special behavior in SGE: a job exiting with 99 will be rescheduled and with > 100 set to application error. > > -- Reuti > > >> ru_wallclock 77 >> ru_utime 65.078 >> ru_stime 3.581 >> ru_maxrss 0 >> ru_ixrss 0 >> ru_ismrss 0 >> ru_idrss 0 >> ru_isrss 0 >> ru_minflt 896944 >> ru_majflt 0 >> ru_nswap 0 >> ru_inblock 0 >> ru_oublock 0 >> ru_msgsnd 0 >> ru_msgrcv 0 >> ru_nsignals 0 >> ru_nvcsw 14382 >> ru_nivcsw 1306 >> cpu 68.660 >> mem 23.400 >> io 0.101 >> iow 0.000 >> maxvmem 1.338G >> arid undefined >> >> >> Why does the exit status=1 if the job didn't fail? >> >> -Bill >> _______________________________________________ >> users mailing list >> [email protected] >> https://gridengine.org/mailman/listinfo/users > _______________________________________________ users mailing list [email protected] https://gridengine.org/mailman/listinfo/users
