Am Fri, 20 Jan 2017 19:58:20 +0100 schrieb "Dr. Thomas Orgis" <[email protected]>:
> So I know my check works, but the failure in the prolog is without > consequence. It is not reported anywhere, apparently. I need to correct that. The prolog failure indeed is reported, but without any acting on it: [2017-01-20T19:58:50.493] sched: Allocate JobID=xxxxxx NodeList=nodexxx #CPUs=32 Partition=all [2017-01-20T19:58:51.605] error: Prolog launch failure, JobId=xxxxxx [2017-01-20T19:58:53.355] job_complete: JobID=xxxxxx State=0x1 NodeCnt=1 WEXITSTATUS 1 [2017-01-20T19:58:53.355] email msg to [email protected]: SLURM Job_id=xxxxxx Name=hello Failed, Run time 00:00:03, FAILED, ExitCode 1 [2017-01-20T19:58:53.364] job_complete: JobID=xxxxxx State=0x8005 NodeCnt=1 done [2017-01-20T19:58:54.560] error: job_epilog_complete: JobID=xxxxxx State=0x8005 NodeCnt=1 epilog error on nodexxx, draining the node [2017-01-20T19:58:54.560] drain_nodes: node nodexxx state set to DRAIN [2017-01-20T19:58:54.560] error: _slurm_rpc_epilog_complete: epilog error JobID=xxxxxx State=0x5 NodeCnt=0 Node=node376 Err=Job epilog failed So this shows rather clearly that is done: The prolog fails, the job is executed and then fails, too (because it executes the same health check that the proloc calls), followed by the epilog. How can I prevent running a job after prolog failure? Alrighty then, Thomas -- Dr. Thomas Orgis Universität Hamburg RRZ / Basis-Infrastruktur / HPC Schlüterstr. 70 20146 Hamburg Tel.: 040/42838 8826 Fax: 040/428 38 6270
smime.p7s
Description: S/MIME cryptographic signature
