Website Down? A Checklist Before Contacting Hosting Support

Your website won't load. The instinct is to open a support ticket immediately — but five minutes of structured checking will often reveal the problem is something you can fix yourself, or at least tell you which support team to contact. Even when you do need help, arriving with the right diagnostics gets your issue fixed far faster than "my site is down, please help."

Work through these steps in order. Each one narrows down where the problem lives.

Step 1: Is It Down for Everyone, or Just You?

This is always the first question, because the answer completely changes the diagnosis.

Use an external uptime checker

Services like DownForEveryoneOrJustMe check your site from their own servers, independent of your connection. If the checker says the site is up but your browser can't reach it, the problem is on your end — your device, your network, or your DNS. If the checker also fails, the problem is real and lives with the site itself.

Try a different device and connection

Load the site on your phone with Wi-Fi turned off (mobile data instead), or ask a friend in another city to try. Same result across devices and networks? It's the site. Works on mobile data but not your office Wi-Fi? Your local network or ISP's DNS is the likely culprit.

What "just me" usually means

If it's only down for you, the most common causes are: a stale DNS cache on your device (your computer remembers an old, wrong address), your ISP's DNS server having a bad day, a VPN routing you through a problematic server, or a browser extension interfering. Steps 3 and 4 cover these — don't skip them even if they sound boring. They fix a surprising number of "outages."

Step 2: Read the Error Message (It Usually Tells You Who's at Fault)

The exact error message is the single most useful piece of information. Don't just say "it's broken" — read what the screen actually says.

Browser connection errors: this site can't be reached, DNS_PROBE

Messages like "This site can't be reached", DNS_PROBE_FINISHED_NXDOMAIN, or "Server not found" mean your browser couldn't even find the server's address. "NXDOMAIN" is technical shorthand for "this domain doesn't exist" as far as your device can tell. These errors point to DNS problems (Step 4) or to a client-side issue like a typo in the URL, a disconnected network, or a browser DNS cache that needs clearing. Cloudflare's DNS documentation notes these can also come from the domain's DNS being misconfigured at the source — so keep both possibilities open.

Cloudflare error numbers: 521, 522, 523, 524 — server-side

If you see a Cloudflare-branded error page with a number, the CDN could reach you fine but couldn't reach your hosting server. Per Cloudflare's official troubleshooting docs:

• Error 521 — web server is down: the origin server actively refused the connection. The server process may have crashed, or a firewall is blocking Cloudflare.
• Error 522 — connection timed out: Cloudflare couldn't complete a connection to the server in time. Often server overload or a firewall silently dropping packets.
• Error 523 — origin is unreachable: the network path to the server failed entirely.
• Error 524 — a timeout occurred: Cloudflare connected, but the server took too long to respond (slow scripts, overloaded database).

All four mean the problem is between Cloudflare and your hosting server — your visitors' browsers and DNS are fine. This is a hosting-side issue.

HTTP 500, 502, 503, 504 — server errors

• 500 Internal Server Error: the server hit something it couldn't handle — often a broken script, a bad plugin, or a misconfigured .htaccess file.
• 502 Bad Gateway / 504 Gateway Timeout: a server in the chain (like a proxy) couldn't get a valid response from the one behind it.
• 503 Service Unavailable: the server is up but refusing requests — overloaded, in maintenance mode, or rate-limiting.

Write down the exact code and any extra text. Support teams triage on these codes.

Step 3: Rule Out Your Own Setup

Before blaming the server, spend two minutes eliminating your own side. These are quick and they genuinely fix things.

Clear browser cache and try incognito/private mode

A corrupted or stale cache can make a working site look broken. Open the site in an incognito window (no extensions, fresh cache). If it loads there, clear your normal browser cache and cookies for the site.

Flush your DNS cache

Your operating system caches DNS answers, and a stale entry can point you at an old or dead IP address for hours.

• Windows: open Command Prompt and run ipconfig /flushdns
• macOS: open Terminal and run sudo dscacheutil -flushcache; sudo killall -HUP mDNSResponder

Then retry the site.

Disable VPN, ad-blockers, and extensions briefly

VPNs route your traffic through third-party servers that can be blocked, overloaded, or geographically restricted. Ad-blockers and security extensions occasionally break legitimate sites too. Turn them off temporarily, test, then re-enable. If the site works with the VPN off, the problem was the VPN's exit server — not your site.

Step 4: Check DNS — the Most Common Silent Cause

DNS is the phone book of the internet: it translates your domain into the server's IP address. It's also the cause of a huge share of "my site is down" reports.

Did you just change nameservers or DNS records?

If you recently pointed your domain at new nameservers, changed hosting, or edited DNS records, wait before panicking. DNS changes need to propagate — the updated information has to spread across DNS servers worldwide, and that can take anywhere from a few minutes to 48 hours, though most changes settle within a few hours. During propagation, some visitors see the new site and some see nothing at all. This is normal, not an outage.

You can watch propagation in real time with tools that check your DNS from servers around the world — if most locations show the new records and a few don't, just wait.

How to spot a DNS problem

Strong hints that DNS is the issue:

• The domain shows NXDOMAIN / "server not found" but the hosting control panel and server IP respond fine.
• It works when you type the raw server IP but not the domain name.
• You recently transferred the domain, changed nameservers, or let the domain expire and renewed it.
• Email for the domain also stopped working at the same time (both rely on DNS).

If DNS is the cause and you didn't change anything, check that the domain hasn't expired and that the nameservers at your registrar still point where they should.

Step 5: Check Whether the Server Itself Is Responding

If DNS looks fine and the error points at the server, the next question is whether the server is alive at all.

Quick server checks you can do yourself

• Hosting control panel: can you log in? If the panel loads but the site doesn't, the server is running and the problem is likely the website's code or configuration, not the machine.
• Other sites on the same account: if you host multiple sites and they're all down, it's the server or account. If only one is down, it's that site's files or database.
• FTP/SSH: if you can connect, the server is up — the web server software (or your site's code) is what's failing.

Signs the hosting plan is overwhelmed

On shared hosting, sites can go down not because the server died but because the account hit its resource limits — CPU, memory, or concurrent processes. Telltale signs: the site is intermittently slow or throws 503/508 errors at busy times but recovers on its own, and the pattern started as traffic grew. That's a capacity problem, and no amount of code-tweaking fully fixes it — it's the signal to look at choosing the right hosting plan and understand how shared hosting works.

Step 6: Think About What Changed Recently

A surprisingly reliable diagnostic rule: what changed in the last 48 hours?

The last 48 hours rule

Websites rarely break spontaneously. Walk back through recent changes:

• WordPress/plugin/theme updates (a failed or incompatible update is a classic 500-error cause)
• New plugins installed
• DNS or nameserver changes
• SSL certificate installed or renewed
• Files edited (especially .htaccess, wp-config.php)
• Domain renewed after expiring

If you can name the change, you can usually undo it or narrow the search dramatically. If your host offers backups, knowing when it broke tells you which restore point to use.

Common triggers people forget

• Billing: an unpaid hosting or domain invoice can suspend service. Check your account's billing status before anything else — it's the most embarrassing and the fastest to fix.
• SSL renewal: certificates expire. If the site broke around the renewal date and browsers now show certificate warnings, the renewal didn't complete properly.
• Traffic spikes: a mention on social media or a sale can push a small plan past its limits (see Step 5).

Step 7: Gather What Support Will Ask For (Do This Before You Contact Them)

If you've worked the checklist and the problem is genuinely on the hosting side, now contact support — but come armed. Cloudflare's own troubleshooting guidance lists exactly what support teams need, and providing it upfront can cut resolution time dramatically. A good ticket includes:

1. The exact error message — screenshot it, or copy the text verbatim (e.g. "Error 522, Ray ID 8a9f...").
2. The Ray ID (for Cloudflare error pages) — it's printed at the bottom of the error page and lets support pull the exact failed request from their logs.
3. The affected URLs — which pages fail? All of them, or just some?
4. When it started — date, time, and your timezone. "Around 2pm PKT today" beats "recently."
5. What you've already tried — list the checklist steps you completed. This stops support from asking you to repeat them.
6. Whether it's down for everyone — include the result of your external uptime check.

Tickets with this information get routed to the right technician immediately. Tickets that say "site down plz fix" go to the back of the queue while someone asks you all of the above.

Frequently Asked Questions

My website is down but my internet works fine. What should I check first?

Check whether it's down for everyone or just you (Step 1) — use an external uptime checker and try mobile data. If it's just you, flush your DNS cache and try incognito mode. Most "down for me only" cases are DNS or browser cache, fixed in minutes.

How do I know if it's a DNS problem or a hosting problem?

DNS problems show as "site can't be reached" / NXDOMAIN errors and often follow nameserver or record changes; the hosting control panel usually still works. Hosting problems show 500-series errors, Cloudflare 521–524 pages, or timeouts — and the control panel may also be unreachable. Read the exact error message (Step 2); it almost always tells you which one it is.

I just changed my nameservers. How long should I wait before worrying?

Up to 48 hours for full global propagation, though most locations update within a few hours. Check propagation from multiple locations — if the majority show your new records, the stragglers will catch up. Only start troubleshooting after 48 hours, or if no location shows the new records (which suggests the change itself was wrong).

What does a Cloudflare 522 error mean, and is it my fault?

Error 522 means Cloudflare couldn't establish a connection to your hosting server in time — the server is overloaded, down, or a firewall is blocking Cloudflare. It's a server-side issue, not a visitor problem. Check whether the server is running and whether anything changed recently; if you can't resolve it, it's a legitimate hosting support ticket — include the Ray ID from the error page.

Should I contact my domain registrar or my hosting provider?

It depends on where the fault is. Domain/DNS issues (expired domain, wrong nameservers, NXDOMAIN errors) go to whoever manages your DNS — often your registrar. Server errors (500s, timeouts, site files broken) go to your hosting provider. If you're unsure, the error message from Step 2 usually points the right way.

What information makes a support ticket get fixed faster?

The exact error message (screenshot or verbatim text), the Ray ID if it's a Cloudflare error page, the affected URLs, when it started (with timezone), what you've already tried, and whether it's down for everyone. Support teams triage on these details — a complete ticket skips the entire back-and-forth of them asking you for basics.

Still stuck after running the checklist? You've already done the hard part — you know the error code, whether it's DNS or hosting, and when it started. Take those diagnostics to support and you'll get a real answer instead of a round of "have you tried clearing your cache?" And if the pattern keeps repeating — slow at busy times, 503s under load — it may be time to revisit choosing the right hosting plan rather than fighting the same battle every month.