a month ago
Hi Railway team,
My Chatwoot service suddenly stopped working.
Environment:
- Railway Hobby Plan
- Chatwoot Community template
- Postgres healthy
- Valkey healthy
What I've already done:
- Upgraded from Free to Hobby
- Increased memory to 8 GB
- Removed stale Rails PID issue
- Added startup command:
rm -f /app/tmp/pids/server.pid && bundle exec rails server -b 0.0.0.0 -p $PORT
Current status:
- Deployment completes
- Healthcheck fails
- Service never becomes healthy
- Railway shows:
"Deployment failed during network process"
"Network > Healthcheck"
"Healthcheck failure"
Recent logs previously showed:
"A server is already running (pid: 29, file: /app/tmp/pids/server.pid)"
Now deployment reaches healthcheck but still fails.
Additional information:
I changed the startup command to:
rm -f /app/tmp/pids/server.pid && bundle exec rails server -b 0.0.0.0 -p $PORT
The previous PID error disappeared, but deployments still fail during the Railway network healthcheck.
Current healthcheck path tested:
- /api
- /installation/onboarding
Both fail.
Could you check whether the container is actually binding to the expected Railway port and whether this could be related to the current AWS incident affecting US West workloads?
Could you help determine why the service is failing healthchecks and whether this is related to the current AWS US West incident?
Thank you.
2 Replies
a month ago
This thread has been opened as a bounty so the community can help solve it.
Status changed to Open Railway • 26 days ago
a month ago
Try removing the healthcheck for now, see if the application starts.
a month ago
Additional diagnostics:
The latest deployment is marked as Completed, but the application remains unavailable.
Current observations:
-
HTTP logs return 502 responses for all requests
-
Response times are consistently around 15 seconds
-
Metrics show:
- CPU usage: 0
- Memory usage: 0
- Public network traffic: 0
-
Console shows:
- "No running instances"
- "Disconnected"
-
Deploy logs only show:
Mounting volume...
Starting Container
-
No Rails, Puma, Chatwoot, or application logs appear after startup
We already resolved the previous PID issue and removed the healthcheck as suggested. The deployment now completes successfully, but there still appears to be no active running instance serving requests.
The service is deployed in US West, and I also noticed the current Railway status banner mentioning AWS workload issues in that region.
Could you please verify whether the container is actually starting and remaining attached to a running instance, and whether this behavior could be related to the current US West infrastructure incident?
Project:
heartfelt-simplicity / Chatwoot / production