عناوين IP والمنافذ

Written Explanation

What is an IP address?

An IP address is a number that identifies a device on a network that uses the Internet Protocol (IP). MDN's glossary defines it as a number used to give each device on an IP network a unique address. Without addresses, data would have no destination.

Every packet you met in Lesson 1 carries two addresses in its header: where it came from and where it is going. Routers along the way read the destination address and pass the packet onward, like postal sorters reading the street name.

IPv4 addresses

The older and still very common format is IPv4. An IPv4 address is a 32-bit number. You almost never see it in binary (BEG-001-L03). Instead it is written as four numbers separated by dots, each between 0 and 255:

192.0.2.10

Why 0 to 255? Each of the four numbers is one byte (8 bits), and a byte can hold 256 different values. Four bytes give 32 bits in total. That is about 4.3 billion possible addresses, which turned out not to be enough for a world full of phones, laptops, and sensors.

The addresses 192.0.2.0 to 192.0.2.255 are a special case: they are reserved for use in documentation and examples, so you can safely use them in teaching. This lesson uses them, which means they are illustrative: they do not belong to any real server.

IPv6 addresses

To solve the shortage, a newer format called IPv6 uses 128 bits. It is written as groups of hexadecimal digits separated by colons:

2001:db8:85a3::8a2e:370:7334

You do not have to learn hexadecimal to understand the idea. IPv6 simply provides vastly more addresses. Two rules shorten it: leading zeros in a group can be dropped, and one run of all-zero groups can be replaced with ::. The address above uses the second rule. The prefix 2001:db8::/32 is reserved for documentation, like the 192.0.2.x block.

Both formats are in use today. A single domain can have both an IPv4 and an IPv6 address, and your computer picks whichever works.

Public and private addresses

Not every address is reachable from the whole internet.

• A public address is unique across the internet. Servers that serve the public have these.
• A private address is used inside a local network such as your home Wi-Fi or an office. The common private IPv4 ranges are 10.x.x.x, 172.16.x.x to 172.31.x.x, and 192.168.x.x. Many homes have devices numbered 192.168.1.x.

Your home router typically has one public address facing the internet and gives each device a private one inside. When your phone fetches a page, the router translates between the two (a technique called NAT, network address translation). You do not need to configure this, but it explains why the address shown in your phone's Wi-Fi settings and the address a website sees can be different.

The special address for "this computer"

The address 127.0.0.1 always means "this computer itself". It is called the loopback address, and its usual name is localhost. A program that connects to localhost talks to another program on the same machine, without leaving it. Developers use it constantly: when you start a web service on your laptop to test it, you open localhost in a browser.

For example, the address http://localhost:8080 means: this computer, port 8080.

What is a port?

One computer usually runs many network programs at once. A web server, a database, and a mail program may all live on the same machine, and they may all be waiting for connections. If the IP address only names the computer, how do incoming packets reach the right program?

A port is a number that picks one communication endpoint on a computer. MDN's glossary calls a port a communication endpoint designated by a number. Each program that wants to receive connections listens on a port, and each incoming connection says which port it wants.

Port numbers are whole numbers from 0 to 65535 (16 bits). The Internet Assigned Numbers Authority (IANA) divides them into ranges:

The famous ones for this course:

IANA's registry confirms that port 443 is HTTPS, and MDN states that the defaults are 80 for HTTP and 443 for HTTPS.

Developer-made services commonly use higher numbers such as 3000 or 8080, which are conventions, not rules. A backend you write in a later module will probably start on a port like 8080 so that it does not need special permissions.

Putting address and port together

When you want to reach a service, you give both:

192.0.2.10:8080
 |          |
 |          +-- port: which service on that computer
 +------------- IP address: which computer

The colon separates them. You write the address first, then a colon, then the port.

In a URL, the port comes after the host:

https://example.com:8443/path

If the port is left out, the browser uses the default for the scheme: 80 for http and 443 for https. That is why https://example.com/ and https://example.com:443/ mean the same thing, and why you rarely see a port in everyday URLs.

The apartment building, again

Two things follow from the analogy:

1. Two different flats in the same building (two programs on one computer) have the same street address but different door numbers.
2. The same door number in a different building (port 80 on another server) is a completely different place.

That is why it takes both parts to name a service.

A peek at the sender's side

The sender also has an address and a port. When your browser connects to a server on port 443, your computer picks a temporary port for its own side so that replies find their way back to the right program on your machine. You rarely see this, but it is why many browser tabs can talk to the same server at once without their answers getting mixed up.