Updates on recent issues

Thank you for your patience with recent issues with CHPC systems. We'd like to share what happened and thank CHPC staff and partners on campus for their assistance and commitment to keeping systems running.

Helpdesk email issue on July 30 (now resolved)

The issue with the helpdesk email, [email protected], has been resolved. At approximately 4:30 p.m. on July 30, the helpdesk system stopped receiving and sending emails. The issue was related to the platform used for helpdesk tickets, and it appears to have affected other departments in addition to the CHPC. CHPC staff immediately engaged the support team on campus, who identified the issue and contacted the vendor. Emails are now being sent and received appropriately, and it appears that no information was lost; unsent emails were saved in a queue until the issue was resolved.

notch472 through notch501 were offline briefly on July 30 (now resolved)

At approximately 10:30 a.m. on July 30, notch472 through notch501 went offline, which resulted in Slurm job failures. The nodes lost power due to an issue with the power distribution units supporting the racks. CHPC staff quickly responded and resolved the issue on-site; all affected nodes were available again by approximately 12:00 p.m.

Network issues on July 30–31 (now resolved)

Network issues on July 30 and 31 caused issues with connections to some CHPC systems. Several users indicated that connections were intermittent or slow, or that nodes were inaccessible. CHPC staff worked with the university's network team to identify and address the problem, which was resolved by approximately 11:30 a.m. on July 31.

If you are still encountering issues with CHPC systems, please contact us by emailing [email protected].