5 months ago
Hi Railway team,
I'm experiencing intermittent DNS resolution issues with my Railway-generated domain (not a custom domain).
Issue:
- Domain: be-your-future-copywriting.up.railway.app
- Error: DNS_PROBE_FINISHED_NXDOMAIN
- The service shows "Online" in the dashboard, logs show normal execution
- Deployments complete successfully
- Redeploy temporarily fixes the issue, but it comes back after some time
Environment:
- Region: us-west2
- Service: n8n (Node.js application)
- Architecture: Primary + Worker + Redis + Postgres
What I've tried:
- Flushing local DNS cache (ipconfig /flushdns)
- Accessing via incognito/private window
- Accessing from mobile on 4G
- Redeploy (works temporarily)
Observations:
- The issue happens multiple times per day
- When it fails, all tests (incognito, mobile, different networks) fail
- Redeploy fixes it temporarily but it returns
- No errors in deploy logs when this happens
This is affecting production workflows. Is there any known issue with DNS for .up.railway.app domains in us-west2, or something I should check on my end?
1 Replies
5 months ago
Update - RESOLVED:
Found and fixed two separate issues:
Issue 1: X-Forwarded-For / Rate Limiter Error
- Cause: n8n wasn't configured to trust the Railway proxy
- Solution: Added environment variables to the Primary service:
N8N_PROXY_HOPS=1
N8N_TRUST_PROXY=true Issue 2: Redis/Postgres Connection Loss → DNS_PROBE_FINISHED_NXDOMAIN
- Cause: A pinned workflow was causing instability in the Redis/Postgres connections, which made the service appear offline (DNS error)
- Solution: Unpinned the problematic workflow
Both issues are now resolved. The service is stable.
Thanks for the support resources — leaving this here in case it helps others with similar symptoms.
Status changed to Open medim • about 2 months ago
Status changed to Solved medim • about 2 months ago