We have completed the recovery of most instances affected by this event. A very small number of instances are still not responding. We will continue to attempt to restore all instances or notify users directly if the instances remain degraded.