2 months ago
I'm experiencing intermittent connectivity failures between Cloudflare and my Railway-hosted API. Requests fail before reaching my application - no backend logs are generated during these incidents.
Error Types Observed:
- 520 - Web server returning unknown error
- 522 - Connection timed out
- 525 - SSL handshake failed
Timeline (all times PST):
- Jan 12, 2026 @ 11:42 AM - First occurrence
- Jan 13, 2026 @ 10:07 AM
- Jan 13, 2026 @ 1:27 PM
- Jan 14, 2026 @ 10:27-10:55 AM - Multiple failures in this window
Monitoring Setup:
DataDog Synthetic browser tests ping my /ready endpoint every 5 minutes. These failures are being caught by that monitoring.
Infrastructure:
- Cloudflare DNS with proxy enabled → Railway
- This setup has been stable for approximately 1 month
- No recent changes to DNS, Cloudflare settings, or Railway configuration
Cloudflare Ray IDs (for correlation):
- 9bdf395f654eb637 (520)
- 9bdf37f154deb637 (525)
- 9bdf2d7491eab637 (525)
- 9bdf2c36e440fccc (525)
- 9bdf294a455a596d (522)
Cloudflare Edge Locations:
Newark, Ashburn
Key Observation:
No corresponding logs appear in my Railway application during these failures, confirming the connection is failing at the Railway edge/proxy layer before reaching my server.
Request:
Can you investigate what's happening on Railway's side during these timestamps? Happy to provide additional information. I'll continue monitoring and can share further incident timestamps as they occur.
8 Replies
2 months ago
I've also noticed a lot of slowness in the past day or so. I think it's related to this thread:
https://station.railway.com/questions/edge-routing-going-through-asia-instead-17b353fb?utm_campaign=Community+Thread+Notification&utm_content=Community+Thread+Notification&utm_medium=email_action&utm_source=customer.io
I'm still seeing a lot of request go through Singapore. Here's an example:
x-railway-edge: railway/asia-southeast1-eqsg3a
x-railway-request-id: vaNWpG72QlmOtlE2V7rehQ
cf-ray: 9be7f6ba3a127678-SEA
I'm starting to see some requests go through us-east (maybe about 25%) of them. But my service is configured in us-west so it's still slow and CloudFlare is routing through Seattle, WA which is correct.
I'm thinking this might be why I've been seeing these intermittent connectivity issues and slowness that have been going on the last couple of days.
2 months ago
Hello,
We fixed the issue where some users were being routed to the Asia edge. Please try again.
Status changed to Awaiting User Response Railway • about 2 months ago
2 months ago
I'm still seeing issues as of this morning. Now everything is routing through us-east. I would expect it to route through us-west.
This request was made at 8:27 AM PST
cf-ray: 9beefa695b6b18f4-SEA
x-railway-edge: railway/us-east4-eqdc4a
x-railway-request-id: kSRaTIbDQdKO2HloozsQ6Q
My service is configured to be in us-west and Cloudflare is routing through Seattle, WA.
Status changed to Awaiting Railway Response Railway • about 2 months ago
2 months ago
Hello!
We're acknowledging your issue and attaching a ticket to this thread.
We don't have an ETA for it, but, our engineering team will take a look and you will be updated as we update the ticket.
Please reply to this thread if you have any questions!
2 months ago
Hi - this should be resolved now. Let us know if you are still seeing any routing issues.
The Railway Team
Status changed to Awaiting User Response Railway • about 2 months ago
2 months ago
Yes, I've confirmed all requests are now coming from railway/us-west2. Thanks!
Status changed to Awaiting Railway Response Railway • about 2 months ago
Status changed to Solved Anonymous • about 2 months ago
2 months ago
✅ The ticket Edge routing optimization has been marked as completed.
2 months ago
✅ The ticket Edge routing issue for non-US-East traffic has been marked as completed.