I had a problem with the defaults determining suspended jobs failed, and saw your comment. I'll weight in and say suspended jobs should be considered running. On many clusters I've encountered jobs are frequently suspended for higher priority qos or partitions. Or suspended during maintenance.
I had a problem with the defaults determining suspended jobs failed, and saw your comment. I'll weight in and say suspended jobs should be considered running. On many clusters I've encountered jobs are frequently suspended for higher priority qos or partitions. Or suspended during maintenance.