Does it start if the healthcheck is removed?

I haven't tried doing that yet as that too takes a PR to strike it -- but the 502 gateway error while still having an active 'good' deployment on staging has been what I've been investigating. My assumption has been that the server is fine, but nothing can resolve to it...

Was there any change made to the network configuration in the service settings, like changing the target port, or modifying the PORT variable of your service? Also, removing the healthcheck doesn't require a new PR, you just have to remove it from service settings, and hit apply changes, which will redeploy the service.

This error is caused when Railway's edge routes the traffic to your service, but your app doesn't receive it, because it has an incorrect port/host binding in most cases. Also, healthchecks require the PORT env variable to be set, so it knows which port to route the request to. Hopefully, this helps you in narrowing down what the issue is.

looking at the config, none of the network settings have changed in over 6 months... it is set to the correct port.

our config is set with a railway file, the web UI will not let you modify it in that case it needs to be a new code change.

I've disabled the health check and the system does come up fine now.

It doesn't really answer why the health checks are failing when the server is actually up. Any ideas or things do debug?

That means the app is binding correctly. I assume the PORT variable is set to the same port your app is listening on?

We've never had the PORT variable needing to be set (we use 8080), i tried setting per your suggestion while debugging but it made no difference either way

currently the only change made to the server was disabling the health check -- which has worked, and the route set for the health check works fine after being deployed...

Thank you for your suggestions darseen!