serac icon indicating copy to clipboard operation
serac copied to clipboard

Published 20 hours ago • verified publisher icon LLNL

Investigate allocation messages on TOSS3

Open white238 opened this issue 2 years ago • 1 comments

There seems to be something slightly wrong with the allocation commands

29: srun: Job 9479148 step creation still disabled, retrying (Requested nodes are busy)
26: srun: Job 9479148 step creation still disabled, retrying (Requested nodes are busy)
51: srun: Job 9479148 step creation still disabled, retrying (Requested nodes are busy)
50: srun: Job 9479148 step creation still disabled, retrying (Requested nodes are busy)
43: srun: Job 9479148 step creation still disabled, retrying (Requested nodes are busy)
49: srun: Job 9479148 step creation still disabled, retrying (Requested nodes are busy)
52: srun: Job 9479148 step creation still disabled, retrying (Requested nodes are busy)
53: srun: Job 9479148 step creation temporarily disabled, retrying (Requested nodes are busy)

white238 avatar May 04 '22 17:05 white238

Try one of the following:

  • setting environment variable: SLURM_OVERLAP=1
  • adding flag --overlap to srun commands

white238 avatar Aug 01 '22 20:08 white238