a month ago
Hi everyone,
My production backend has suddenly become unreachable, which is currently breaking the live frontend for my platform, It's Your Story.
All requests to my GraphQL API at https://its-your-story-prod.up.railway.app/api/graphql are failing with a DNS_PROBE_FINISHED_NXDOMAIN error, as seen in the attached screenshots.
This used to happen occasionally in the past, but it would always start working again by itself after a short while. Now, however, the domain seems completely broken and is keeping my production site down permanently.
Here are the details of my setup and what I have verified so far:
The service currently shows as "Online" and active in the dashboard (the last deployment was about 6 months ago).
Enable Serverless is toggled OFF, so it shouldn't be a cold-start sleep issue.
My account billing is active.
Since the container shows as active but the domain is failing to resolve entirely, it seems like the DNS record might have been dropped or the edge routing became detached on Railway's end.
Could someone please look into this or advise on how to force the edge network to recognize the domain again?
Thanks for your help!
Pinned Solution
a month ago
It works for me.
It’s possible that your ISP may be blocking traffic to Railway’s pregenerated domains. If that’s the case, then unfortunately, Railway isn’t able to do anything about it. You’d need to use a custom domain or a VPN.
6 Replies
a month ago
This thread has been opened as a bounty so the community can help solve it.
Status changed to Open Railway • about 1 month ago
a month ago
Try flushing your local DNS cache, or use a different device/network to make the requests. You can also try using a VPN as well.
0x5b62656e5d
Try flushing your local DNS cache, or use a different device/network to make the requests. You can also try using a VPN as well.
a month ago
Flushing DNS or telling production users to use a VPN is unacceptable. This is not a local issue on my end.
Yes, it works on a VPN. No, it doesn't work off it. That proves Railway's edge network dropped the route and regional ISPs are now caching a dead DNS record for your up.railway.app domain. A manual redeploy didn't fix it, and users on completely different cellular networks are getting the exact same NXDOMAIN error.
I am paying for this service to work out of the box. Escalate this to your engineering team and fix the routing drop on your end.
irfan106
Flushing DNS or telling production users to use a VPN is unacceptable. This is not a local issue on my end. Yes, it works on a VPN. No, it doesn't work off it. That proves Railway's edge network dropped the route and regional ISPs are now caching a dead DNS record for your up.railway.app domain. A manual redeploy didn't fix it, and users on completely different cellular networks are getting the exact same NXDOMAIN error. I am paying for this service to work out of the box. Escalate this to your engineering team and fix the routing drop on your end.
a month ago
It works for me.
It’s possible that your ISP may be blocking traffic to Railway’s pregenerated domains. If that’s the case, then unfortunately, Railway isn’t able to do anything about it. You’d need to use a custom domain or a VPN.
0x5b62656e5d
It works for me. It’s possible that your ISP may be blocking traffic to Railway’s pregenerated domains. If that’s the case, then unfortunately, Railway isn’t able to do anything about it. You’d need to use a custom domain or a VPN.
a month ago
I'm cancelling my Subscription right away. This is as stupid as it can be.
If your platform's pregenerated domains are failing regionally and your official response is "not our problem, tell your users to use a VPN or go buy a custom domain," then your product is fundamentally broken.
I paid for this service so I wouldn't have to deal with infrastructure drops. I'll be moving my backend to a provider that actually maintains its routing and takes responsibility for its downtime.
irfan106
I'm cancelling my Subscription right away. This is as stupid as it can be. If your platform's pregenerated domains are failing regionally and your official response is "not our problem, tell your users to use a VPN or go buy a custom domain," then your product is fundamentally broken. I paid for this service so I wouldn't have to deal with infrastructure drops. I'll be moving my backend to a provider that actually maintains its routing and takes responsibility for its downtime.
a month ago
Railway isn’t able to control what ISPs block.
0x5b62656e5d
Railway isn’t able to control what ISPs block.
a month ago
I get that you can't control what ISPs do, but the issue seems specifically tied to Jio and its related networks. Since they are such a massive provider, them blocking or dropping traffic to up.railway.app breaks the out-of-the-box experience for a huge number of users. Could your engineering team look into this and see if there is a way to request a delisting or routing fix with them?
Status changed to Solved 0x5b62656e5d • 7 days ago