The server went down again, but this time came back after a reboot. We already have database replication running to the new servers and this outage certainly boost the migration priority.