10th January 2019

SLURM Scheduler Slurm (scheduler) issues

UPDATE: A confluence of regal being unresponsive and a large array cancellation made Slurm very slow/unresponsive. This also put some nodes in drain state. Cleanup is generally complete and any additional stuck nodes are being restarted.

The job scheduler (Slurm) is experiencing performance issues. We are investigating. updates to follow.

For issues not shown here, please contact FASRC via
https://portal.rc.fas.harvard.edu or email rchelp@rc.fas.harvard.edu