I added PROFILE=1 to the params of sched_conf in order to measure where the scheduler was spending its time. I discovered a fortuitous side effect is that any "job BLAH should have finished since" lines appear between the PROF: sge_mirror and PROF: static urgency lines. We have a script that processes the message file to report these problem jobs to us and up till now we've been using a fudge factor on the time to determine if the problem is current or not. It looks like we can use the profiling lines to separate problems that were still current on the last scheduling cycle from earlier ones as they appear in the same place regardless of how long the scheduling cycle took. Can anyone see any problems with this approach? Are there any negative side effects to having PROFILING enabled all the time (presumably it doesn't eat much CPU or it wouldn't be much use)?
William _______________________________________________ users mailing list [email protected] https://gridengine.org/mailman/listinfo/users
