This incident has been resolved.

The issue has been identified. We removed a faulty node from the cluster and rescheduled the workloads on the other compute nodes. All sites are back online again - We're continuing to work on getting the faulty node back into the cluster

We are currently investigating this issue.