redirect

Redirect is a mechanism that sends a user or client from one URL or resource to another. In web contexts, redirects occur when a server responds with a redirect status code, guiding the client to a different location. Redirects can also refer to input/output redirection in command-line interfaces, where data is redirected from or to files or devices.

HTTP redirects use 3xx status codes. The most common are 301 Moved Permanently and 302 Found (temporary). Other codes include 303 See Other, 307 Temporary Redirect, and 308 Permanent Redirect. A 301 indicates a permanent move and is often used during site migrations; it may transfer search engine ranking to the new URL. A 302 historically meant temporary, though modern implementations may treat it as a temporary or conditional redirect. 307 and 308 preserve the original request method. Redirects can be configured on servers (for example, Apache or Nginx) or via meta refresh or JavaScript. Important considerations include avoiding redirect chains and loops, ensuring the target URL is correct, and handling canonicalization for search engine optimization.

Redirect targets can be absolute URLs or relative. A well-managed redirect strategy uses stable, crawlable URLs so that bookmarks and links remain valid. DNS-level redirects, such as CNAME records, alias a hostname and are not HTTP redirects; they do not convey a move for crawling. Security risks include open redirects, where an attacker can cause a user to land on a malicious site, so validating redirect targets is recommended.

In Unix-like systems, redirection refers to directing input or output streams, using operators such as > to create or overwrite files, >> to append, < to read input, and 2> to redirect standard error. Redirection enables simple data routing between commands and files.