Service: garderobe-backend

Domain: garderobe-backend.railway.app (Railway-generated native domain,

not a manually added custom domain)

GET /health on this domain works correctly and returns {"ok":true} (HTTP 200).

Every request to a path under /api/ (e.g. /api/catalog, /api/orders/test123)

on the SAME domain instead returns a generic 404 "Not Found"

(Content-Length: 9, Content-Type: text/plain), with response headers

Access-Control-Allow-Origin: https://railway.com and Via: 1.1 google — this

doesn't look like it's coming from my own Express app, but from Railway's

edge/ingress layer itself.

In this service's Network Logs (HTTP tab) I see no incoming HTTP traffic

at all, not even for the /health requests that did return a valid

response — suggesting some traffic isn't reaching the service instance

via the logged ingress layer.

Under Settings → Networking I'm also getting the message "Networking

info temporarily unavailable — Domains and TCP proxy details could not

be loaded."

This looks similar to an edge-ingress sync issue described in a previous

community post about custom domain edge routing failures, though in my

case it's not the whole domain that's affected but specifically the

/api/ path portion of the routing.

Already tried: manual redeploy, waiting several minutes, verified

environment variables are correctly present, direct requests bypassing

the frontend entirely (same result).

Could someone take a look at the edge ingress configuration for this

service?