Any web developer would know about HTTP buy may not be familiar with the exact syntax of it because it’s abstracted by the network libraries.

HTTP is just a protocol which a client and a server use for data communication. In order to comply with this protocol, every HTTP message should be formed in accordance with a certain standard.

In this post, I’m going to give a brief overview of HTTP messages, and build a simple web server with netcat.

HTTP Messages

There are two types of HTTP messages, requests and responses. Each has its own format.

Requests

An example HTTP request message:

GET / HTTP/1.1
Host: google.com
  • GET: Method.
  • /: Path.
  • HTTP/1.1: The Protocol version.
  • Host: google.com: Headers.
  • Body. (optional for some methods like POST or PUT)

Responses

An example HTTP response message:

HTTP/1.1 301 Moved Permanently
Location: http://www.google.com/

<HTML>301 Moved</HTML>
  • HTTP/1.1: The Protocol version.
  • 301: Status code.
  • Moved Permanently: Status message.
  • Location: http://www.google.com/: Headers.
  • <HTML>301 Moved</HTML>: Body.

netcat

netcat is a networking utility for sending and listening packets using TCP or UDP connections. The command is nc. For more information, run man nc.

To test the both server and client models, I’m going to use two terminals - term1 and term2.

Server Model

On term1, start nc listening on port 80 which is a typical port number for HTTP.

# term1
$ nc -l 80

Client Model

On term2, connect to the machine of term1.

# term2
$ nc 127.0.0.1 80

Now a connection between the port 80 has been made, you can send any text message from both terminals. For example, send hello world from term1, then in term2, you will see the message received from term1.

# term1
$ nc -l 80
hello world
# term2
$ nc 127.0.0.1 80
hello world

Sending HTTP request messages using netcat

Using netcat, you can actually send HTTP messages in pure text. No browser, no library, no programming. Let’s send the same HTTP request in the example above.

$ nc google.com 80
GET / HTTP/1.1
Host: google.com

HTTP/1.1 301 Moved Permanently
Location: http://www.google.com/
Content-Type: text/html; charset=UTF-8
Date: Sat, 01 May 2021 04:32:47 GMT
Expires: Mon, 31 May 2021 04:32:47 GMT
Cache-Control: public, max-age=2592000
Server: gws
Content-Length: 219
X-XSS-Protection: 0
X-Frame-Options: SAMEORIGIN

<HTML><HEAD><meta http-equiv="content-type" content="text/html;charset=utf-8">
<TITLE>301 Moved</TITLE></HEAD><BODY>
<H1>301 Moved</H1>
The document has moved
<A HREF="http://www.google.com/">here</A>.
</BODY></HTML>

You will get an HTTP response message with a bunch of headers and body from google server. The other way of sending the same HTTP request is:

$ printf 'GET / HTTP/1.1\r\nHost: google.com\r\n\r\n' | nc google.com 80

Note that there is a blank line(\r\n) at the end of the message. It indicates all information for the request has been sent.

Buliding a web server using netcat

Let’s run a simple web server using netcat and connect to it with a web browser. The web server in this example will redirect any incomming request to https://google.com.

$ printf 'HTTP/1.1 302 Moved\r\nLocation: https://google.com/' | nc -l 80

Then open the web broswer and connect to 127.0.0.1 or localhost. The browser will redirects to https://google.com and the HTTP request that the browser sent is printed on the console.

$ printf 'HTTP/1.1 302 Moved\r\nLocation: https://google.com/' | nc -l 80
GET / HTTP/1.1
Host: 127.0.0.1
Connection: keep-alive
...

The request messages may differ depending on which web browser you use.

Conclusion

An important takeaway here is that netcat doesn’t know anything about HTTP layer. It just opens a connection to a port and sends a sequence of characters - a string - over TCP layer. It is the web server that decides whether the information is a valid HTTP request or not and responds to it, and that is the protocol. This simple and intuitive demonstration gave me a clearer understanding of how HTTP works.