[COMPLETED] 2024-02-21: Upgrade of the queue system on Colossus

[2024-02-21 17:49 update] All affected jobs have been requeued. 85 jobs had to be cancelled, so please inspect the output of your jobs to see if they're affected.

[2024-02-21 15:45 update] Several jobs that were running at the start of the upgrade did not successfully resume. We're trying to resolve the issue. New jobs are not affected.

[2024-02-21 10:35: update] The upgrade is now done, and seems to have gone well.

[2024-02-21 10:00: update] The upgrade has now started

The queue system on Colossus will be upgraded on Wednesday (February 21) at 10:00.  During the upgrade, running jobs will be suspended, and slurm commands (squeue, sbatch, etc) will not work.  We expect the upgrade to take no more than 20 minutes.

Published Feb. 15, 2024 3:32 PM - Last modified Feb. 21, 2024 5:50 PM