31st January 2018

Virtual instances - Los Angeles, DC4 Node v1107 hardware faiure

18:07 PST Our monitoring system detected ECC memory errors on server v1107.

20:01 PST We will have to decommission this server completely to perform repairs & stability testing.

We will be migrating clients onto different nodes over the next 5-8 hours.

21:24 PST While migrating the clients the server died completely and will not boot. We are investigating.

21:37 PST Bad memory module was removed from server and we are continuing migrating clients.

23:28 PST All VMs have been successfully migrated off of this node & restarted. Node will be decommissioned.