Holland Computing Center - Swan GPU access degraded for security update – Incident details

All systems operational

Status of Holland Computer Center Resources


Swan GPU access degraded for security update

ResolvedDegraded performance117 hours 35 minutes

Affected

HPC resources

Degraded performance from 10:40 PM to 8:16 PM

Swan

Degraded performance from 10:40 PM to 8:16 PM

  • ResolvedUTC

    The majority of Swan GPU nodes are updated and back in service, with the remaining few to be updated as running jobs complete.

  • IdentifiedUTC

    Swan GPU access will be degraded, delaying job start times, to drain nodes for a high severity NVIDIA driver security update. Running jobs are unaffected. GPU nodes will be placed back online as they are drained of running jobs and updated.