HTTP Status Code Lookup

Features

• Look up by number: meaning, 1xx–5xx class, typical uses and troubleshooting notes
• Reverse lookup by keyword, in English or Chinese
• Enter one or two digits to list a whole group — 40 lists the 4xx codes starting with 40
• Codes that get confused point at each other: 401 and 403, 301 and 308, 502 and 504
• Covers 62 common codes; an unlisted number gets its class explained instead of a blank
• Description language can be switched between English and Chinese

How to use

1. Type a three-digit code (404), or a keyword (rate limit)
2. Switch the description language or turn related codes off if you prefer
3. The result appears as you type — no button to press

FAQ

What exactly is the difference between 401 and 403?

A 401 means the request was not authenticated: no credentials, or invalid or expired ones. The fix is to log in again or refresh the token. A 403 means it was authenticated and the server still refuses — it knows who you are and will not let you do this, so logging in again changes nothing. Per spec a 401 must carry a WWW-Authenticate header saying how to authenticate; a 403 sends you to permission settings, IP or region blocks, or a WAF rule.

Is 301 vs 302 just permanent vs temporary?

Not quite. A 301 is cached long-term by browsers, and the caching is heuristic — it can be remembered even when no cache headers were sent, so a wrong 301 leaves users broken until they clear their cache. That is the classic 301 incident. On top of that, methods other than GET and HEAD are rewritten to GET by most browsers; when the method must survive, use 308 (permanent) or 307 (temporary).

What is the trap with 304?

A 304 only means something alongside cache headers: the server must have sent ETag or Last-Modified earlier, so the client can come back and ask whether its copy is still good. A 304 with no cache headers leaves the client with neither content nor a usable copy, which shows up as a blank page or an empty API response — a misconfiguration that never raises an error and is hard to trace back to a status code.

How do 502 and 504 differ?

A 502 is a gateway that received an invalid response from upstream: a crashed backend, a wrong port, an unparseable reply. A 504 is a gateway that timed out waiting for upstream: a slow backend or a stalled third-party call. With a 502, first check that the upstream process is alive and listening; with a 504, look at the backend’s own timings. Both error pages are usually emitted by Nginx, so the backend log may show nothing at all.

How many codes are covered?

62 common ones, based on RFC 9110, 9111 and 9112 plus documented extensions such as the WebDAV client errors, 429 for rate limiting and 418 as the April Fools joke. A valid but unlisted number, 599 for instance, gets its class explained and a note saying the number itself is not in the table, rather than pretending it was found.