11:28 PST we have received an alert from our monitoring system about node v5205 outage.
11:34 PST we have determined that the reason for service outage is a RAM module failure.
11:41 PST we were able to reboot the server with the affected module disabled. All VMs have started successfully.
11:44 PST our system began automatic migration of all VMs away from the affected node. As this issue is no longer impacting our clients, we will close this issue.