Originally thought this was an issue with Fargate, but even with EC2 as the master 503 pages come up quite often. Back, refresh doesn't work. But if you wait around 60 seconds you can connect to the interface again. Almost seems like a restart is happening in the background.
Did a little troubleshooting and the path to the LB target needs to be changed to /login otherwise the healthcheck never passes and the server just keeps cycling every few minutes.
Originally thought this was an issue with Fargate, but even with EC2 as the master 503 pages come up quite often. Back, refresh doesn't work. But if you wait around 60 seconds you can connect to the interface again. Almost seems like a restart is happening in the background.