The root has been identified and the issue resolved.

We have identified the root cause of the issue and are working on a resolution. As the issue requires reboots we will evacuate other workloads without downtime first.

A driver update cause the underlying GPUs to fail and shelve your Virtual Servers.

The matter is escalated and will keep you posted with the actions accordingly. Please try not to unsheleve the VMs currently as they are automatically isolated to avoid data loss due to the driver issue.