2 months ago
- Project ID:
5faa6ac9-9c40-4bfc-a046-57a116c6dbb1 - Environment:
production - Issue: Service-to-service communication over private network is broken
- Evidence:
- DNS resolves correctly (e.g.,
frontend.railway.internal→[fd12:a5b6:c391:1:8000:e0:65d8:a664]and10.216.166.100) - All TCP connection attempts timeout with
ConnectTimeoutException - Tested with IPv4, IPv6, short names, full
.railway.internalnames - Frontend Nginx is running and listening on both IPv4 and IPv6
- Gateway has IPv6 egress enabled
- Latest deployment:
88ab95fe-b218-4698-b8e9-6bb8be798e97
- DNS resolves correctly (e.g.,
This is not a configuration issue - it's a platform networking problem that only Railway's infrastructure team can fix.
1 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
The likely reason is a Railway private-network routing failure rather than DNS: Railway private networking is scoped to the same project and environment, should be called as http://SERVICE.railway.internal:<port> over the service’s listening port, and supports TCP traffic; so if DNS resolves, the service is listening, and connections still time out, that points to a broken internal network path on Railway’s side.
The short fix is: verify both services are in the same project/environment and call the internal hostname over http on the exact listening port; if it still times out, open a Railway infra ticket with the project ID and deployment ID because there is no public app-side fix for a failed private-network route.