Hi. I have looked through multiple documents but cannot find an explanation for the meaning of the state codes in the slurm control daemon log file,eg job_complete: JobID=846 State=0x8003 NodeCnt=1 done job_complete: JobID=222 State=0x8005 NodeCnt=1 done job_complete: JobID=224 State=0x1 NodeCnt=1 WEXITSTATUS 13 I'm assuming 0x1 means success, but what do 0x8003 and 0x8005 mean and what other values might I expect to see? Thanks. Mark.
Hi Mark. The job states can be found in slurm/slurm.h.in. 0-15 are mutually exclusive "states" the job can be in; the rest are bit flags that may be set in addition to the regular state. State 0x8003 is COMPLETE (state 3) with the flag JOB_COMPLETING (0x8000). State 0x8005 is FAILED (state 5) with the flag JOB_COMPLETING (0x8000). State 1 is RUNNING. Does that help?
*** Ticket 3731 has been marked as a duplicate of this ticket. ***
(In reply to Marshall Garey from comment #1) > Hi Mark. The job states can be found in slurm/slurm.h.in. 0-15 are mutually > exclusive "states" the job can be in; the rest are bit flags that may be set > in addition to the regular state. > > State 0x8003 is COMPLETE (state 3) with the flag JOB_COMPLETING (0x8000). > State 0x8005 is FAILED (state 5) with the flag JOB_COMPLETING (0x8000). > State 1 is RUNNING. > > Does that help? Yes, thankyou. Please go ahead and close the case. Thanks. Mark.