Site not loading
A ordered checklist for a site that will not load: DNS, TLS, the app process, the proxy and the origin server.
Work top to bottom. Each step rules out one layer, so you do not restart services that were never broken.
1. Is it just you?
- Try the site on a mobile connection, not the same Wi-Fi.
- Try
curl -I https://yoursite.comfrom a terminal. A browser may be showing a cached error. - Check whether other sites on the same account load.
If the site works elsewhere, the problem is local DNS, a VPN, or a browser cache. Flush DNS and try an incognito window.
2. Does DNS resolve?
dig +short yoursite.com
nslookup yoursite.com 8.8.8.8
- No answer — the record is missing or the delegation is broken. See Nameservers and delegation.
- Wrong address — a stale record or the wrong zone. Compare it with the address in the console.
- Correct address — continue.
Remember that a cached answer can persist for the record's TTL. Test with a public resolver and from a different network before changing DNS.
3. Is TLS working?
curl -vI https://yoursite.com
Look at the certificate subject, expiry and the TLS handshake. A certificate error is covered in SSL certificate issues.
4. What HTTP status comes back?
| Status | Meaning | Where to look |
|---|---|---|
| 502 Bad Gateway | The proxy reached the platform but not your app | App process, port, bind address |
| 503 Service Unavailable | No backend is available for the hostname | App stopped, domain not attached, deploy failed |
| 504 Gateway Timeout | The app accepted the connection but did not answer in time | Slow query, blocked event loop, deadlock |
| 404 | The request reached a server that has no route for it | Wrong domain, wrong path, static file missing |
| 521/522 (Cloudflare) | Cloudflare cannot reach the origin | Origin down or blocking Cloudflare |
A 503 with a short plain-text body is often the proxy's own “no available server” response, which means the hostname is not attached to a running app. Check the app's Domains card and its status.
5. Is the app running?
Cloud hosting:
- Open the app. The status should be running.
- Open Deployments — the latest deployment should be
success. - Open Logs — did the process crash on start?
- Confirm the app listens on
0.0.0.0and the expected port. - Restart once. If it crashes again, read the log rather than restarting repeatedly.
VPS or dedicated server:
systemctl status nginx # or your web server
ss -lntp # what is listening, and on which address
journalctl -u nginx -n 100 # recent errors
The console's Servers page tells you whether the machine itself is active.
6. Did something change?
Most outages follow a change. Check, in order:
- A deploy in the last hour
- An environment variable edit (saving one triggers a deploy)
- A DNS record edit
- A certificate or domain change
- A firewall rule on a VPS
- A provider-side incident
If a deploy caused it, redeploy the previous commit. If a variable caused it, restore the old value and save.
7. Escalate with evidence
If the checks above do not explain it, open a support ticket with:
- The exact URL and the time it started (with timezone)
- The output of
curl -vIanddig +short - The app or server name
- The latest deployment status and the last 50 lines of its log
- What you changed before it broke
That list turns a week of back-and-forth into a single reply.
Quick reference
| Symptom | First thing to check |
|---|---|
| Nothing resolves | DNS record and delegation |
| Resolves but times out | Firewall or origin down |
| Certificate warning | DNS points here? Certificate issued? |
| 502 | App port and bind address |
| 503 | App running and domain attached |
| 504 | Slow application code or database |
| Works on Zoophost hostname, not custom domain | DNS and the Domains card |
| Works for some users only | Cached DNS, geo routing, local proxy |
Next
- down
- 502
- 503
- 504
- dns
- ssl
- troubleshooting
- outage
Ready to use this
If the guide did not fix it, tell us what you already tried.
- A real support thread, not a bot loop
- Support in IST

