Currently if the server is deemed unhealthy, it starts returning an error 500 for all requests. It would be better if only a "/healthcheck" endpoint would be affected by that and leave up to AWS to disconnect the server from the load balancer. The problem is, AWS routes to all targets if all of them are unhealthy - this should be addressed as well.
refi93
commented
5 years ago
Currently if the server is deemed unhealthy, it starts returning an error 500 for all requests. It would be better if only a "/healthcheck" endpoint would be affected by that and leave up to AWS to disconnect the server from the load balancer. The problem is, AWS routes to all targets if all of them are unhealthy - this should be addressed as well.