Website Down Checker

In order of how often each works: hard refresh (Ctrl/Cmd+Shift+R) to bypass your cache; try a private/incognito window (isolates extensions and cookies — if this works, an extension or stale cookie is your culprit); try another device on the same network (separates device from network); switch DNS to 1.1.1.1 or 8.8.8.8 (stale ISP DNS causes a surprising share of phantom outages); finally restart the router. If mobile data works where Wi-Fi doesn't, it's your network or ISP — and now you know exactly what to tell them.

200–399: healthy — the server answered normally. 403: usually not down — many sites block automated checkers while serving humans fine (our checker identifies itself honestly as a robot, and bot-protection walls like Cloudflare challenge it); if you can load it in a browser, it's up. 404: the server is up but that specific page doesn't exist. 429: alive but rate-limiting. 500/502/503: the meaningful ones — their application or upstream is failing; 503 often accompanies deploys and overload. Timeout or connection failure: the closest thing to truly down.

No, and this tool won't pretend: one vantage point proves reachability from here, which settles the 'just me?' question — the thing you actually asked — but can't rule out regional outages (a CDN region failing, a country-level block, an ISP routing problem). If a major site fails our check, that's usually big news within minutes on outage aggregators, which crowdsource thousands of reports. The honest decision tree: down here + down for you = really down; up here + down for you = fix your side with the checklist.

Because 'down' is rarely binary at scale. Giant services fail partially — one region, one API, one product — so a checker hitting the homepage can get a healthy 200 while Gmail chokes, or vice versa. Big outages also frequently trace to shared infrastructure (a CDN, DNS provider, or cloud region), which is why unrelated sites fall over together — those incidents make 'half the internet is down' afternoons. When a titan fails our check, assume infrastructure event, check an outage tracker, and give it twenty minutes before troubleshooting anything of yours.