Sven/Notes
Contents
30 March 2007
Topics:
Client Server model
How a client side web browser interacts with a web server. See http://web-sniffer.net for an example of how http headers interact between the server and client. The Apache HTTP Server uses a listener process on port 80 (and ports 21, 443) to communicate with the client using the Transmission Control Protocol TCP/IP. The listener uses threads to fork child processes as client requests demand content from the web server.
Client browsers allow caching of downloaded web content to save on bandwidth. The browser uses HTTP Response Header information to determine if a cached or new copy of the body of the web content needs to be transmitted to the client browser. The crital part in the HTTP Response Header are the fields;
Cache-Control: no-cache (completely dynamic), no-store, max-age=0 (check header everytime), must-revalidate Date: Fri, 30 Mar 2007 09:11:02 GMT |
GET and POST requests are essentially similar, they differ in the amount of content that can be transfered, GET requests are restricted to 4Kb. An analogy in the context of an email message, is that a GET request is like all the information being in the subject line, and a POST request is like an email attachment.
See also
Debugging
There are three ways to debug, use print and echo statements to print I/O to STDOUT, use debugging software available within a program, and use functions which create log files of content at specific points of execution.
