Purdue University

The Negishi cluster will be partially unavailable Wednesday, May 28, 2025 from 8:00am - 5:00pm EDT for scheduled maintenance. The cluster will return to full production by Wednesday, May 28th, 2025 at 5:00pm EDT.

During this time, Negishi will have its operating system patched, and upgrades will be made to SMB network drive connectivity. Filesystems will remain available on the cluster, but existing network drive connections may be disrupted. Login sessions via ThinLinc, SSH, and Open OnDemand will terminate as the servers reboot.

Any Slurm jobs submitted during the maintenance will not start and will remain in the queue until after the maintenance is completed.

What Users Can Expect

• You can log into the clusters and submit jobs during the maintenance, though new jobs will not start.
• Login nodes will reboot, terminating existing sessions, but you will be able to log back in without waiting for the end of the maintenance window.
• While you will have login node access, the ability to check job status or submit new jobs will be briefly interrupted while the scheduler reboots.
• While you will have access to the Open OnDemand interface through Gateway, you may experience a brief interruption when it is rebooted. If that happens, simply reload the web page. If you are running an interactive job, you will be able to reconnect to it.
• The Slurm scheduler will reboot, during which time new Slurm queries and job submissions will be interrupted. Any Slurm jobs running before the maintenance will continue to run to completion under the existing OS.