I can assure you, you can not trust any AWS health checks to be a primary alert for something down. You have to do it all yourself, on host, or inside the conta

I can assure you, you can not trust any AWS health checks to be a primary alert for something down. You have to do it all yourself, on host, or inside the container. AWS/Rackspace support just say: "It's your problem as we don't manage what is inside the AWS service".