University of Cambridge Computer Laboratory - Some GPU cluster VMs crashed – Incident details

All systems operational

Show current status

Some GPU cluster VMs crashed

ResolvedPartial outage59 minutes

Affected

Virtual Machine Hosting

Partial outage from 2:17 AM to 3:16 AM

GPUs

Partial outage from 2:17 AM to 3:16 AM

  • ResolvedUTC

    This incident has been resolved.

  • IdentifiedUTC

    Due to a problem during network maintenance, VMs on the departmental GPU cluster briefly lost access to their disks. This caused some VMs to crash. Affected VMs will be rebooted (if CPU VMs) or shut down (if GPU VMs); the latter can be started again from XO.