A page won't load. Before you assume the site crashed — or blame your Wi-Fi — there's a fast, reliable way to find out which it is. The single most useful question in web troubleshooting is “is it down for everyone, or just me?” and answering it takes about thirty seconds once you know how.
Down for everyone vs. just you
When a website fails to load on your screen, only two things can be true: the site is genuinely offline for the whole internet, or it's reachable everywhere except from where you're sitting. Almost every troubleshooting decision flows from telling these two situations apart.
The trick is to get a second opinion from outside your own network. Your browser, your operating system, your router, your ISP, and any VPN or proxy you use all sit between you and the server. Any one of them can be the reason a healthy site looks dead. An external down-checker sidesteps all of that: it fetches the URL from a remote server on a completely different network and reports what it sees. If the remote server loads the page, the site is up and the fault is local to you. If the remote server also fails, the site really is down.
The common causes and how each looks
“The site is down” is really a bucket of very different failures, and each one has a recognizable signature. Learning the fingerprints saves you from chasing the wrong fix.
- DNS failure — the domain name can't be turned into an IP address. You'll see errors like
DNS_PROBE_FINISHED_NXDOMAINor “server address could not be found.” The page never even starts loading. Often local (a stale cache) but sometimes the domain's DNS is genuinely broken or expired. - Server / 5xx errors — DNS resolves, the connection succeeds, but the server answers with a
500,502, or503. This is unambiguously the site's problem: the application crashed, a backend is unreachable, or it's overloaded. - Expired or invalid SSL certificate — the browser blocks the page with “Your connection is not private” (
NET::ERR_CERT_DATE_INVALID). The server is up; its certificate simply lapsed or is misconfigured. Affects everyone until renewed. - Network / routing problem — packets can't find a path between you and the server. Pages hang and then time out. This can be your ISP, an undersea cable, or a bad BGP route, and it may affect only certain regions.
- CDN / origin issue — the site sits behind a CDN like Cloudflare or Fastly, and the edge can't reach the origin server. You'll often see a branded error page (e.g. Cloudflare
521or522) rather than a blank screen. - DDoS or traffic overload — the server is swamped and either times out or returns
503. Common during viral spikes, ticket sales, or attacks. - Scheduled maintenance — a deliberate, temporary
503with a “be right back” message. Intended, and usually short.
The diagnostic ladder, step by step
Work these in order. Each rung either fixes the problem or narrows down where it lives, and you can stop the moment you get a clear answer.
- Check from an external checker first. Run the URL through an is-it-down tool. This single step tells you “everyone vs. just me” before you waste time on anything else.
- Test a second, unrelated site. If nothing loads anywhere, your internet connection is down — check the router and modem, not the website.
- Resolve the DNS. Confirm the domain still maps to an IP address with a DNS lookup. An
NXDOMAINresult means the name itself is failing to resolve — possibly an expired domain or a broken record. - Check the HTTP status code. If DNS resolves, look at what the server actually returns. A
200is healthy; a5xxpoints squarely at the server (see the table below). - Test the SSL certificate. If the browser is warning about security rather than failing to connect, run an SSL check to confirm the certificate is valid and unexpired.
- Try a different network or device. Switch your phone to cellular data, or use a different Wi-Fi. If it suddenly works, the fault is your original network, not the site.
- Flush your local DNS cache. Clear a stale record with
ipconfig /flushdnson Windows orsudo dscacheutil -flushcache; sudo killall -HUP mDNSResponderon macOS, then reload. - Trace the route. If everything above looks fine but the page still hangs, run
traceroute example.com(macOS/Linux) ortracert example.com(Windows) to see where packets stop — a dead hop reveals a routing problem between you and the server.
Is the site down right now?
Check any website from a remote server and get the live HTTP status and response time in seconds — free, no signup, answers “everyone or just me” instantly.
Open the Is It Down? Checker →What the HTTP status code tells you
Once DNS resolves and the connection succeeds, the server returns a three-digit HTTP status code. That number is the fastest way to pinpoint blame — it distinguishes “the server is broken” from “the page moved” from “you're not allowed in.”
200 OK— the server responded normally. If you still can't see the page, the trouble is in the browser, rendering, or client-side scripts, not availability.3xx(301, 302, 307, 308) — redirects. The site is up and sending you elsewhere; a redirect loop can look like an outage, so trace the chain.403 Forbidden/404 Not Found— the server is healthy but is refusing this request or can't find that specific page. The site itself is not down.500 / 502 / 503 / 504— server-side failures.500is an app crash,502is a bad gateway (a proxy got a broken response upstream),503is overloaded or in maintenance, and504is a gateway timeout waiting on a slow backend. All of these are the site owner's problem.
For the full reference of every code and what it means, see our HTTP status code guide.
Symptom → likely cause → next step
Match what you're seeing to the most probable cause, then jump straight to the right fix:
| Symptom | Likely cause | Next step |
|---|---|---|
“Server not found” / NXDOMAIN | DNS failure or expired domain | Run a DNS lookup; flush local DNS |
| “Connection is not private” | Expired / invalid SSL cert | Run an SSL check; wait for owner to renew |
500 / 502 error page | Server or app crash | It's their problem — retry later |
503 “be right back” | Overload or maintenance | Wait and retry; nothing to fix on your end |
Cloudflare 521/522 | CDN can't reach origin | Owner must revive the origin server |
| Page hangs, then times out | Network / routing problem | Run a traceroute; try another network |
| Works on phone, not computer | Local cache, VPN, or firewall | Flush DNS; disable VPN/proxy; check hosts file |
| Loads fine on external checker | Local device or ISP | Restart router; switch DNS resolver |
Their problem or yours? What to do next
When it's the website's problem
If an external checker confirms the site is down for everyone — a 5xx, an expired certificate, or a timeout from multiple locations — there is genuinely nothing you can do to load it. Your options are to wait and retry, check the company's status page or social accounts for an outage notice, and if it's your own site, look at server logs, the CDN dashboard, and deployment history for the trigger.
When it's your problem
If the outside world can reach the site and you can't, work through the local suspects: restart your router and modem, flush your DNS cache, disable any VPN or proxy, temporarily switch your device to a public DNS resolver like 1.1.1.1 or 8.8.8.8, clear the browser cache, and check that a security tool or an entry in your hosts file isn't blocking the domain. One of those almost always explains a “just me” outage.
Frequently asked questions
How do I know if a website is down for everyone or just me?
Why can I reach a site on my phone but not my computer?
hosts entry.What does a 503 error mean?
503 Service Unavailable means the server is reachable but temporarily can't handle the request — usually overloaded, deploying, or in maintenance. It's a server-side issue, so waiting and retrying is the only option for visitors.Is it my internet or the website that's down?
How do I check if a website is down?
200 means it's up, a 5xx means the server is failing, and a timeout or DNS error means it can't be reached at all.