Reduced capacity on Zeus nodes

Incident Report for Pawsey Supercomputing Centre

Resolved

One bank of outlets on a PDU have failed. A workaround has been implemented and the compute nodes remain operational

Posted Jun 03, 2021 - 10:56 AWST

Identified

One of the power distribution units (PDUs) in the rack had failed. Nodes are still running jobs but without power redundancy.

Posted Jun 01, 2021 - 10:59 AWST

Investigating

Automated monitoring indicates that zome of the zeus nodes are offline or running on non-redundant power supplies. Staff onsite will investigate.

Posted Jun 01, 2021 - 10:33 AWST

This incident affected: Zeus (Zeus Compute nodes).