Although computers have hardware ports used as connection points, ports are also numbers used to map traffic to a particular process on a computer. These ports, also called TCP and UDP ports, are where programs transmit data. When a computer sends traffic over the Internet to a server or another computer, it uses an IP address to identify the server or remote computer, and a port number to identify the process on the server or computer that receives the data (If an IP address is like a street address, a port number is like an apartment number within that street address).
For example, suppose you want to see a particular web page. Your web browser attempts to create a connection on port 80 (the port used for HTTP traffic) for each element of the web page. When your browser receives the data it requested from the HTTP server, such as an image, it closes the connection.
Many ports are used for only one type of traffic, such as port 25 for SMTP (Simple Mail Transfer Protocol). Some protocols, such as SMTP, have ports with assigned numbers. Other programs are assigned port numbers dynamically for each connection. The IANA (Internet Assigned Numbers Authority) keeps a list of well-known ports. You can see this list at:
http://www.iana.org/assignments/port-numbers
Most policies you add to your Firebox configuration have a port number between 0 and 1024, but possible port numbers can be from 0 to 65535.
Ports are either open or closed. If a port is open, your computer accepts information and uses the protocol identified with that port to create connections to other computers. However, an open port is a security risk. To protect against risks created by open ports, you can block ports used by hackers to attack your network. For more information, go to About Blocked Ports.
You can also block port space probes: TCP or UDP traffic that is sent by a host to a range of ports to find information about networks and their hosts. For more information, go to About Port and IP Address Scans.