httpd Busybox Linux Web Server Tutorial #1

httpd Busybox Linux Web Server Tutorial #1

Video is ready, Click Here to View ×

This video was sponsored by:
Karl Arvid
John Tedesco –
Vinay Sud

For help:

Intro Video by Milan Ristic



Http uses tcp, and you can think of tcp as a pipe between two computers. What you put into the pipe in one side will appear on the other side in the exactly same shape and order. No duplicated part, no dropped part or no reordering. That is what tcp are for. You can use udp instead of tcp to connect two computers, but then you don't get a pipe. You throw envelopes and hopes the envelope will reach the other computer. But no guarantees.
Back to tcp. When you connect to another computer you identify which pipe to use on each computer by a 16-bit number, the port number. Each pipe have two ends, which have a unique port number. So each end can have different port number, but still connect.
Some port numbers are well known, port 80 is one for http. Low port numbers are reserved for those. So when you want to contact a well known port, you choose one high numbered for your side. And that is why you also get a new high number on the Web side on each new connection, refresh in the web browser.
It makes a new pipe to port 80 from the web browser, but because a port number are quarantined a while after used and closed, you get a new free port number.

That is why you get new port numbers in the web servers log.

Leave a Reply

Your email address will not be published. Required fields are marked *