Instances on the affected compute node are now coming back online. We will issue another update when service has been completely restored.
