During a regular update of the management agent the SQL host indicated that the connection was lost. The SQL host crashed and a failover was initiated. Where this normally brings the redundant cluster up within 30 seconds the failover actually lasted > 6 minutes. As this led to loss of connection to databases it was decided to manually power down the failed cluster instead of waiting. Which resulted in the start up of the required VM’s. The total downtime of the SQL servers was 40 minutes.

To avoid this issue in the future also the management agent will now only be updated in the regular maintenance windows.