24 days ago
Hello Team,
My deployed project was accessible via browser from mac desktop and iphone but now it stopped working without any new code being pushed. I tried to redeploy it but it didn't help.
frontend - frontend-production-22a3.up.railway.app
badkend - backend-production-8e8df.up.railway.app ( regenerated multiple times to fix the issue but no luck )
Container is running (Uvicorn on 0.0.0.0:8000) .Public domain returns ERR_NAME_NOT_RESOLVED
It started happening suddenly . I upgraded the plan but still it didn't happen. I was using to personally and has some important data as well and I don't want to loose the data as it's going to help me in further fine tuning of the app. Please help fixing the error.
Pinned Solution
23 days ago
If you need friends to test the application, I would stop relying on Railway's generated domain and attach a custom domain.
For example:
app.yourdomain.com
api.yourdomain.com
pointed via CNAME to Railway.
Benefits:
More reliable DNS propagation
Easier troubleshooting
Independent of Railway-generated subdomain changes
Better for collecting user feedback
One thing I'd still verify
Open a terminal and run:
dig frontend-production-9697.up.railway.app
and
dig @8.8.8.8 frontend-production-9697.up.railway.app
Post the full output.
The exact status code (REFUSED, NXDOMAIN, SERVFAIL, etc.) and the authoritative nameserver chain will reveal whether this is:
ISP resolver failure,
DNSSEC issue,
Railway zone delegation problem,
regional DNS filtering.
Given the data you've shown so far, my estimate is roughly:
80% ISP/recursive DNS issue
15% Railway DNS delegation issue affecting some resolvers
5% application/deployment issue
The fact that Google DNS resolves the hostname is the strongest piece of evidence that the application itself is not the root cause.
12 Replies
24 days ago
This thread has been opened as a bounty so the community can help solve it.
Status changed to Open Railway • 24 days ago
24 days ago
Hey, I can reach both services from my computer. Please give your devices some time to update their DNS records.
24 days ago
I have been trying since morning . I'm unable to login using my credential to the app ,neither I'm able to register new users. Why did it happen at first place? how long do I need to wait
24 days ago
My backend container is fully healthy (confirmed via console: curl localhost:8000/health returns 200 OK) but the public domain backend-production-56af.up.railway.app returns ERR_NAME_NOT_RESOLVED
My backend container is fully healthy (confirmed via console: curl localhost:8000/health returns 200 OK) but the public domain backend-production-56af.up.railway.app returns ERR_NAME_NOT_RESOLVED
24 days ago
I can also access both of your domains, please try on mobile data and or flushing your computer DNS.
23 days ago
Use cloudflare one client to access your application when you are in url: xxx.up.railway.app. If you don't want to use vpn then use custom domain
valuemoretec
Use cloudflare one client to access your application when you are in url: xxx.up.railway.app. If you don't want to use vpn then use custom domain
23 days ago
I wasn't using vpn. I was directly accessing it for last 15 days and all of a sudden i faced issues. I'll try cloudflare but I'm still clueless what went wrong
23 days ago
Pretty sure this is a DNS thing on your side, not Railway. ERR_NAME_NOT_RESOLVED means your device can't even look up the domain, so it's never actually reaching the server. Makes sense given the container's healthy and the mods can load it fine.
Quick things to try:
Switch to mobile data and see if it loads. If it does, it's your network's DNS.
Flush your DNS cache. ipconfig /flushdns on Windows, sudo killall -HUP mDNSResponder on Mac.
Change your DNS to 1.1.1.1 or 8.8.8.8 and restart your router. Your ISP's resolver is probably the culprit.
Run nslookup your-domain 1.1.1.1. If it resolves there but not normally, that confirms it's your ISP.
If you're putting this in front of real users, honestly just grab a custom domain and point a CNAME at it. The default railway subdomains can be flaky with some ISPs.
23 days ago
- Test on mobile data first
Turn off WiFi on your phone and try loading the URL on mobile data. If it works, your home network/ISP's DNS is the culprit.
- Flush your DNS cache (Mac)
bashsudo killall -HUP mDNSResponder
Then try again in a fresh browser tab.
- Switch to a public DNS resolver
Change your DNS to Cloudflare or Google instead of your ISP's:
Cloudflare: 1.1.1.1 and 1.0.0.1
Google: 8.8.8.8 and 8.8.4.4
On Mac: System Settings → Network → your connection → DNS → add those IPs. Then restart your router.
- Confirm it's DNS with nslookup
bashnslookup backend-production-56af.up.railway.app 1.1.1.1
If this resolves but your normal browsing doesn't, that proves your ISP's DNS is the problem.
pagetree
1. Test on mobile data first Turn off WiFi on your phone and try loading the URL on mobile data. If it works, your home network/ISP's DNS is the culprit. 2. Flush your DNS cache (Mac) bashsudo killall -HUP mDNSResponder Then try again in a fresh browser tab. 3. Switch to a public DNS resolver Change your DNS to Cloudflare or Google instead of your ISP's: Cloudflare: 1.1.1.1 and 1.0.0.1 Google: 8.8.8.8 and 8.8.4.4 On Mac: System Settings → Network → your connection → DNS → add those IPs. Then restart your router. 4. Confirm it's DNS with nslookup bashnslookup backend-production-56af.up.railway.app 1.1.1.1 If this resolves but your normal browsing doesn't, that proves your ISP's DNS is the problem.
23 days ago
I switched to mobile data and the problem still exists. I want to access the UI from 4-5 devices (friends) i.e mac,window,android and iphone to get initial feedback. The solution you're proposing for my own mac only. How can it be accessed from all devices ?? It was working till 12th of June starting from the day of my first deployment and suddenly stopped working without any new changes being pushed. I'm clueless now. cmd1 : nslookup frontend-production-9697.up.railway.app , output1:** server can't find frontend-production-9697.up.railway.app: REFUSED, cmd2 : nslookup frontend-production-9697.up.railway.app 8.8.8.8 , output2: output looks good, and same for backend.
I switched to mobile data and the problem still exists. I want to access the UI from 4-5 devices (friends) i.e mac,window,android and iphone to get initial feedback. The solution you're proposing for my own mac only. How can it be accessed from all devices ?? It was working till 12th of June starting from the day of my first deployment and suddenly stopped working without any new changes being pushed. I'm clueless now. cmd1 : nslookup frontend-production-9697.up.railway.app , output1:** server can't find frontend-production-9697.up.railway.app: REFUSED, cmd2 : nslookup frontend-production-9697.up.railway.app 8.8.8.8 , output2: output looks good, and same for backend.
23 days ago
If you need friends to test the application, I would stop relying on Railway's generated domain and attach a custom domain.
For example:
app.yourdomain.com
api.yourdomain.com
pointed via CNAME to Railway.
Benefits:
More reliable DNS propagation
Easier troubleshooting
Independent of Railway-generated subdomain changes
Better for collecting user feedback
One thing I'd still verify
Open a terminal and run:
dig frontend-production-9697.up.railway.app
and
dig @8.8.8.8 frontend-production-9697.up.railway.app
Post the full output.
The exact status code (REFUSED, NXDOMAIN, SERVFAIL, etc.) and the authoritative nameserver chain will reveal whether this is:
ISP resolver failure,
DNSSEC issue,
Railway zone delegation problem,
regional DNS filtering.
Given the data you've shown so far, my estimate is roughly:
80% ISP/recursive DNS issue
15% Railway DNS delegation issue affecting some resolvers
5% application/deployment issue
The fact that Google DNS resolves the hostname is the strongest piece of evidence that the application itself is not the root cause.
pagetree
If you need friends to test the application, I would stop relying on Railway's generated domain and attach a custom domain. For example: app.yourdomain.com api.yourdomain.com pointed via CNAME to Railway. Benefits: More reliable DNS propagation Easier troubleshooting Independent of Railway-generated subdomain changes Better for collecting user feedback One thing I'd still verify Open a terminal and run: dig frontend-production-9697.up.railway.app and dig @8.8.8.8 frontend-production-9697.up.railway.app Post the full output. The exact status code (REFUSED, NXDOMAIN, SERVFAIL, etc.) and the authoritative nameserver chain will reveal whether this is: ISP resolver failure, DNSSEC issue, Railway zone delegation problem, regional DNS filtering. Given the data you've shown so far, my estimate is roughly: 80% ISP/recursive DNS issue 15% Railway DNS delegation issue affecting some resolvers 5% application/deployment issue The fact that Google DNS resolves the hostname is the strongest piece of evidence that the application itself is not the root cause.
23 days ago
PFA the output. Thank you for your suggestion btw. I'll try with custom domain option and let you know .
Attachments
PFA the output. Thank you for your suggestion btw. I'll try with custom domain option and let you know .
23 days ago
Thank you so much for you suggestions mate. I created domain in cloudflare and updated the custom domain for both frontend and backend (app.yourdomain.com and api.yourdomain.com) along with few variables and CORS in the main class . Now it started working fine for me from my mac and iphone as well 😊 would you mind explaining what's went wrong and why it stopped working while I was using up.railway.app? Thank you once again
Status changed to Solved passos • 17 days ago
