2 months ago
Verified:
- Both services in same project:
5faa6ac9-9c40-4bfc-a046-57a116c6dbb1✓ - Both services in same environment:
production✓ - Gateway calling correct internal hostname:
http://frontend.railway.internal:80✓ - Frontend listening on correct port:
80(Nginx) ✓ - DNS resolves: Yes ✓
- Service is online: Yes ✓
- Service is serving correct content: Yes ✓
The issue is confirmed as a Railway private-network routing failure.
Since redeploy didn't fix it, you need to open a Railway infrastructure ticket with:
- Project ID:
5faa6ac9-9c40-4bfc-a046-57a116c6dbb1 - Deployment IDs:
- Gateway:
7ccf9b8f-975c-4667-9bfb-718ead8f0ef8 - Frontend:
ef47c3fc-4bd2-4f78-9a7b-d197ecb66728
- Gateway:
- Issue: Private network TCP connection from gateway to frontend times out despite DNS resolution and both services being online in the same project/environment
- Evidence: Frontend is serving correct HTML/assets, but gateway cannot establish connection to
frontend.railway.internal:80
6 Replies
Status changed to Awaiting Railway Response Railway • about 2 months ago
2 months ago
This thread has been marked as public for community involvement, as it does not contain any sensitive or personal information. Any further activity in this thread will be visible to everyone.
Status changed to Open Railway • about 2 months ago
2 months ago
What exactly is the issue?
It seems like all you posted was a response from an AI...
Are you trying to access your frontend service from your local machine?
2 months ago
Getting a blank white page on accessing my url
2 months ago
What URL are you trying to access? And are you accessing it from your local machine?
2 months ago
Before treating this as purely a Railway infrastructure issue, there are a couple of things worth confirming from inside the containers:
- Check how the frontend is binding
Even if Nginx is on port 80, it must be listening on
0.0.0.0, not127.0.0.1.
Run inside the frontend service:
netstat -tulnp | grep 80or
ss -tulnp | grep 80You should see:
0.0.0.0:80If it is bound to 127.0.0.1, the gateway will not be able to reach it over the private network.
- Test from inside the gateway Open a shell in the gateway service and run:
curl -v http://frontend.railway.internal:80If this times out, it confirms either a binding issue or a private network routing issue.
- Try hostname variations Sometimes Railway resolves slightly differently, so also try:
http://frontend.railway.internal
http://frontend:80- Redeploy both services in order Stop both services, deploy frontend first, then deploy gateway. This can refresh internal service discovery.
- Temporary workaround If this is blocking you right now, expose the frontend publicly and point the gateway to its public URL. That bypasses the private network entirely.
Based on everything you shared, if the frontend is correctly bound to 0.0.0.0 and internal curl still times out, then it does look like a Railway private networking issue.
In that case, your ticket is valid, but it will help to include:
- Output of the curl test from gateway
- Output of netstat or ss from frontend
- Whether the timeout is consistent or intermittent
That should give Railway enough to investigate faster.