A single host in the NGD1 availability zone crashed. Customer VMs were automatically restarted on alternative hosts in the cluster.
Our engineers have been made aware of an issue affecting a single host in the NGD1 availability zone. Customer virtual machines running on the host should have automatically restarted on an alternative node.
Our engineers have confirmed that a single host in the NGD1 Availability Zone crashed at 00:42. The VMware environment detected this failure and the High-Availability functionality kicked in to restart affected virtual machines on alternative hosts.
We have identified the affected virtual machines and will be raising tickets directly with the affected customers.
Working with the vendors, the fault was identified as a problematic driver. An update has been issued and applied to the affected host which has been monitored for stability. We have seen stable results and return the host to production, resolving this issue.