Some systems are experiencing issues

Stickied Incidents

Friday 17th May 2019

Critical System Patches

A new vulnerability with Intel processors, "Zombieload", was recently discovered and discussed widely in the news. MSI will be rolling out patches to all systems starting Friday afternoon through the end of Monday to protect against this bug. Patching systems will occur in a rolling fashion so as not to interrupt running jobs. Once a job finishes on a node, the node will be patched then added back into the scheduling queue. As a result of this work, jobs may take longer to start, and some services may be temporarily unavailable. Also, during patching, there will be some performance penalties.

Past Incidents

Tuesday 21st May 2019

No incidents reported

Monday 20th May 2019

No incidents reported

Sunday 19th May 2019

No incidents reported

Saturday 18th May 2019

No incidents reported

Friday 17th May 2019

Critical System Patches

A new vulnerability with Intel processors, "Zombieload", was recently discovered and discussed widely in the news. MSI will be rolling out patches to all systems starting Friday afternoon through the end of Monday to protect against this bug. Patching systems will occur in a rolling fashion so as not to interrupt running jobs. Once a job finishes on a node, the node will be patched then added back into the scheduling queue. As a result of this work, jobs may take longer to start, and some services may be temporarily unavailable. Also, during patching, there will be some performance penalties.

Thursday 16th May 2019

No incidents reported

Wednesday 15th May 2019

Mesabi Scheduler Outage

Mesabi's job scheduler is currently not allowing eligible jobs to begin running on the cluster. Our operations team is working to investigate the source of this outage; in the meantime, no eligible jobs will be able to start. Jobs that are already running have not been affected by this issue.