The HTTP request–response cycle

Every PHP web page sits inside the HTTP request–response cycle. Understanding that cycle explains superglobals, sessions, headers, and why PHP cannot "push" data to a closed tab without another request or WebSockets elsewhere.

Steps in order

1. Client request — browser or API client sends method (GET/POST), URL, headers, optional body
2. Web server — routes the URL to a .php file or front controller (index.php)
3. PHP bootstrap — autoloaders, config, session start (if used)
4. Your logic — read input, validate, query DB, choose view
5. Response — status line, headers (Content-Type, Set-Cookie), body (HTML/JSON)
6. Process ends — unless using long-running workers (RoadRunner, FrankenPHP), the PHP process typically exits

Statelessness

HTTP is stateless: each request is independent. PHP remembers users via sessions (server storage + session cookie) or JWT/tokens in API designs. Do not assume variables from request A exist in request B.

Important interview questions and answers

1. Q: What happens between clicking a link and seeing HTML?
   A: DNS → TCP/TLS → HTTP request → server runs PHP → HTML response → browser parses and renders.
2. Q: Why is HTTP called stateless?
   A: The protocol does not inherently tie requests together; apps add cookies, sessions, or tokens to simulate state.
3. Q: Where does PHP run in this chain?
   A: On the server after the web server receives the request and before the response is sent to the client.

Self-check

1. What carries form data in a POST request—the URL or the body?
2. Why must sessions use cookies or equivalent identifiers?