Issue: Service stuck on old deployment, new builds auto-rollback
aurora-sva
HOBBYOP

24 days ago

The service is hard-pinned to deployment 09fc5d8b-834a-41f5-b2a0-874eade27cee (uptime ~2h44m). Every new deployment: - Builds successfully ✓ - Container starts ✓ - Gets killed immediately (Stopping Container) ✗ - Service rolls back to 09fc5d8b This is not a code issue — the same code runs fine elsewhere. The platform is auto-rolling back new deployments. Please investigate and unstick this service.

$10 Bounty

1 Replies

Status changed to Open Railway 24 days ago


imsushant2005
FREE

22 days ago

This is usually caused by a failed health check, not a stuck deployment.

Railway will keep the last healthy deployment active if the new deployment starts but does not become healthy. That is why it keeps rolling back to deployment 09fc5d8b-834a-41f5-b2a0-874eade27cee.

Check these:

  1. Make sure the app listens on Railway’s $PORT, not a hardcoded port.
  2. Bind to 0.0.0.0, not localhost.
  3. Confirm the health-check path returns HTTP 200 quickly.
  4. Check startup logs right before Stopping Container.
  5. Remove or fix any custom health check that points to the wrong route.
  6. Restart/redeploy after updating the start command.

Example:

gunicorn app:app --bind 0.0.0.0:$PORT

The platform is rolling back because the new container is considered unhealthy. Once the new deployment passes health checks, it will stop falling back to the old deployment.


Welcome!

Sign in to your Railway account to join the conversation.

Loading...