Netcat is a utility that reads and writes data across network connections, using the TCP or UDP protocol. It is designed to be a reliable "back-end" tool that can be used directly or driven by other programs and scripts. At the same time, it is a feature-rich network debugging and exploration tool since it can create almost any kind of connection you would need and has several interesting built-in capabilities. Common uses include:

• Simple TCP proxies
• Shell-script based HTTP clients and servers
• Network daemon testing
• A Socks or HTTP ProxyCommand for ssh

nc [-46bCDdhklnrStUuvZz] [-I length] [-i interval] [-O length] 
   [-P proxy_username] [-p source_port] [-q seconds] [-s source] 
   [-T toskeyword] [-V rtable] [-w timeout] [-X proxy_protocol] 
   [-x proxy_address[:port]] [destination] [port]

Options

destination can be a numerical IP address or a symbolic hostname (unless the -n option is given). In general, a destination must be specified, unless the -l option is given (in which case the local host is used). For UNIX-domain sockets, a destination is required and is the socket path to connect to (or listen on if the -l option is given).

port can be a single integer or a range of ports. Ranges are in the form nn-mm. In general, a destination port must be specified, unless the -U option is given.

Client/Server model

It is quite simple to build a very basic client/server model using nc. On one console, start nc listening on a specific port for a connection. For example:

nc -l 1234

nc is now listening on port 1234 for a connection. On a second console (or a second machine), connect to the machine and port being listened on:

nc 127.0.0.1 1234

There should now be a connection between the ports. Anything typed at the second console will be concatenated to the first, and vice-versa. After the connection was set up, nc does not really care which side is being used as a ‘server’ and which side is being used as a ‘client’. The connection may be terminated using an EOF (‘^D’).

There is no -c or -e option in modern netcat, but you still can execute a command after connection being established by redirecting file descriptors. Be cautious here because opening a port and let anyone connected execute arbitrary command on your site is DANGEROUS. If you really need to do this, here is an example:

On ‘server’ side:

rm -f /tmp/f; mkfifo /tmp/f

cat /tmp/f | /bin/sh -i 2>&1 | nc -l 127.0.0.1 1234 > /tmp/f

On ‘client’ side:

nc host.example.com 1234

(shell prompt from host.example.com)

By doing this, you create a fifo at /tmp/f and make nc listen at port 1234 of address 127.0.0.1 on ‘server’ side, when a ‘client’ establishes a connection successfully to that port, /bin/sh gets executed on ‘server’ side and the shell prompt is given to ‘client’ side.

When connection is terminated, nc quits as well. Use -k if you want it keep listening, but if the command quits this option won't restart it or keep nc running. Also, don't forget to remove the file descriptor once you don't need it anymore:

rm -f /tmp/f

Data transfer

The example in the previous section can be expanded to build a basic data transfer model. Any information input into one end of the connection will be output to the other end, and input and output can be easily captured to emulate file transfer.

Start by using nc to listen on a specific port, with output captured into a file:

nc -l 1234 > filename.out

Using a second machine, connect to the listening nc process, feeding it the file that is to be transferred:

nc host.example.com 1234 < filename.in

After the file is transferred, the connection will close automatically.

Talking to servers

It is sometimes useful to talk to servers "by hand" rather than through a user interface. It can aid in troubleshooting, when it might be necessary to verify what data a server is sending in response to commands issued by the client. For example, to retrieve the homepage of a website:

printf "GET / HTTP/1.0\r\n\r\n" | nc host.example.com 80

Note that this also displays the headers sent by the web server. They can be filtered, using a tool such as sed, if necessary.

More complicated examples can be built up when the user knows the format of requests required by the server. As another example, an e-mail may be submitted to an SMTP server using:

nc [-C] localhost 25 << EOF
HELO host.example.com
MAIL FROM:
RCPT TO:
DATA
Body of e-mail.
.
QUIT
EOF

Port scanning

It may be useful to know which ports are open and running services on a target machine. The -z flag can be used to tell nc to report open ports, rather than initiate a connection. Usually, it's useful to turn on verbose output to stderr by use this option in conjunction with -v option.

For example:

nc -zv host.example.com 20-30

Connection to host.example.com 22 port [tcp/ssh] succeeded!
Connection to host.example.com 25 port [tcp/smtp] succeeded!

The port range was specified to limit the search to ports 20 - 30, and is scanned by increasing order.

You can also specify a list of ports to scan, for example:

nc -zv host.example.com 80 20 22

nc: connect to host.example.com 80 (tcp) failed: Connection refused 
nc: connect to host.example.com 20 (tcp) failed: Connection refused 
Connection to host.example.com port [tcp/ssh] succeeded!

The ports are scanned by the order you given.

Alternatively, it might be useful to know which server software is running, and which versions. This information is often contained in the greeting banners. To retrieve these, it is necessary to first make a connection, and then break the connection when the banner was retrieved. This can be accomplished by specifying a small timeout with the -w flag, or perhaps by issuing a "QUIT" command to the server:

echo "QUIT" | nc host.example.com 20-30

SSH-1.99-OpenSSH_3.6.1p2 Protocol mismatch. 
220 host.example.com IMS SMTP Receiver Version 0.84 Ready

nc -p 31337 -w 5 host.example.com 42

Opens a TCP connection to port 42 of host.example.com, using port 31337 as the source port, with a timeout of 5 seconds.

nc -u host.example.com 53

Opens a UDP connection to port 53 of host.example.com.

nc -s 10.1.2.3 host.example.com 42

Opens a TCP connection to port 42 of host.example.com using 10.1.2.3 as the IP for the local end of the connection.

nc -lU /var/tmp/dsocket

Creates and listens on a UNIX-domain stream socket.

nc -x10.2.3.4:8080 -Xconnect host.example.com 42

Connects to port 42 of host.example.com via an HTTP proxy at 10.2.3.4, port 8080. This example could also be used by ssh.

nc -x10.2.3.4:8080 -Xconnect -Pruser host.example.com 42

The same as the above example, but this time enabling proxy authentication with username "ruser" if the proxy requires it.