[slurm-users] Jobs that may still be running at X time?
novosirj at rutgers.edu
Fri Apr 16 21:33:10 UTC 2021
Does anyone have a particularly clever way, either built-in or scripted, to find out which jobs will still be running at such-and-such time? I bet anyone who’s made the mistake of not entering a maintenance reservation soon enough knows the feeling.
I know that jobs /may/ end earlier, but we’re looking for the potential.
I originally started looking for jobs with an EndTime= that overlapped with the maintenance, using "scontrol show job” and excluding jobs that had a StartTime= that was after the maintenance. Today I had a minor panic as many many new jobs had an EndTime= that overlapped with the maintenance, but a StartTime= that was after we put in the maintenance reservation but before the maintenance period. Come to find they were still pending. Not sure why they have a StartTime= that is before the reservation, but it appears as if they won’t be running.
Anyway, I figure this is something people probably need to know often enough. Any tips?
|| \\UTGERS, |---------------------------*O*---------------------------
||_// the State | Ryan Novosielski - novosirj at rutgers.edu
|| \\ University | Sr. Technologist - 973/972.0922 (2x0922) ~*~ RBHS Campus
|| \\ of NJ | Office of Advanced Research Computing - MSB C630, Newark
More information about the slurm-users