Portfolio › Learn ›Month 02 Networking Basics

🔌 Day 2: Ports, IP, TCP, UDP

Networking
Ports

A Port is just a door through which different services and your device communicates. Earlier, we talked about NAT and PAT which translates the IPs and Ports. 🚪

🚪 Ports

Every Port is of 16 bits, ranges from 0 to 65535, so there are a total of 65536 Ports.

📊 Port Ranges

Port Type Range Usage
Well-Known 0 - 1023 Public ones. Used commonly.
Registered 1024 - 49151 Registered ones, for specific Services.
Dynamic 49152 - 65535 Those which OS chooses randomly.

But what's the real explanation? How is it done? 🤔

Here's how:

  1. When you open your browser, OS assigns that connection (a single tab may contain dozens of connections) a random Port from Dynamic Ports.
  2. Then PAT translates that to a public Port.
  3. The server then talks to your public Port and all the communication happens here.

Example:
You have 3 tabs open in a browser. Every tab will be assigned a unique random Port from Dynamic Ports by your OS. When you make a search, for instance, "Hello", the browser itself binds Port like 443 to the packet. Then it goes through PAT, and the server serves it on Port 443 (HTTPS).

📋 Common Ports

Of course, you don't have to remember all these ports, but here is the list of some Ports you must be familiar with:

Port Service Description
21 FTP File Transfer Protocol
22 SSH Secure Shell (Secure Login)
23 Telnet Unsecure Text Communications
53 DNS Domain Name System
80 HTTP HyperText Transfer Protocol
443 HTTPS HTTP Secure
445 SMB Server Message Block
3389 RDP Remote Desktop Protocol
139 NetBIOS Network Basic Input/Output System
3306 MySQL Database Server
8080 HTTP-Alt Alternative HTTP Port

📜 Protocols

We have rules for the English language, like grammar. Imagine if we spoke English without caring about grammar—would anyone understand us? 🗣️

Similarly, in the realm of the internet, there are rules, like grammar, which are called Protocols. 📜

Examples: HTTPS, SMTP, FTP, etc.

📦 Internet Protocol (IP)

From a hacker's perspective, this is important. 🕵️

Internet Protocol (IP) only cares about the source and destination IPs.
As we know, data flows over the internet in chunks, and these chunks are called packets.
Every packet usually has two main parts (with a third optional one):

  1. Packet Header
  2. Payload

🏷️ Packet Header (IP Header)

Also called the IP Header. Imagine a letter in an envelope. The envelope has sender and receiver addresses. Here, the envelope represents the IP Header.

Important Fields:

  1. Identification: Uniquely identifies each packet. Used for reassembling fragmented packets.
  2. IP Flags: Indicates whether a packet is fragmented (M) or not (D).
  3. Fragment Offset: Shows the position of the fragmented part. Like a page number in a book (page 10 comes after page 9).

  4. TTL (Time To Live): Tells us whether the packet successfully reached its destination.
    Full Explanation of TTL: Despite its name, TTL does not measure real time, it counts "hops" or cycles instead. Here’s what happens:

    • Every packet is assigned a TTL, usually 64 on Linux/macOS and 128 on Windows.
    • The packet leaves your router and travels through multiple routers to reach its destination (it does not go through every router on the internet).
    • Each router the packet passes through subtracts 1 from the TTL.
    • If the packet reaches its destination before TTL reaches zero, it is successfully delivered.
    • If TTL reaches 0 before the packet reaches its destination, the router discards the packet and sends a "TTL Expired" message back to the sender.
    • TTL is crucial because without it, undelivered packets could loop endlessly, increasing traffic and making the network inefficient.

  5. Protocol: Defines which protocol is used (e.g., 6 for TCP, 17 for UDP).

  6. Source and Destination: The IP addresses of sender and receiver.

🤝 TCP (Transmission Control Protocol)

There are many protocols, but TCP is the most reliable.
Before any data is sent, TCP says: "Let's establish a connection."

🧠 TCP Header

Just like the IP Header, the TCP Header ensures data reaches the correct service.

Important Fields:

  1. Source & Destination Port: Where the comms started and where it's going.
  2. Sequence Number: Number given to each data segment to keep order.
  3. Acknowledgement Number: Tells the sender: "I received packet X, send the next one." (Reliability).
  4. Flags:
    • SYN: Start connection.
    • ACK: Acknowledge message.
    • RST: Reset connection (something went wrong).
    • FIN: Finish/Close connection.
  5. Window Size: Flow control. Receiver says "I can only handle this much data."
  6. Checksum: Ensures data integrity.

👋 3-Way Handshake

  1. SYN: "Hello, can you hear me?" (Sender)
  2. SYN-ACK: "I can hear you. Can we talk?" (Server)
  3. ACK: "Acknowledged! Let's begin." (Sender)

Connection States:

What's the difference between IP Header's ID and TCP Header's Sequence number? 🤔

The IP Header gives the packet a unique ID. TCP segments the data and gives it a Sequence Number.

🚀 UDP (User Datagram Protocol)

UDP is used where speed/optimization matters more than reliability.

Imagine you’re standing on a mountain and shout "HELLOOOOO!" 🏔️

You don’t:

You just shout. That’s UDP.

⚠️ Why is UDP tricky?

UDP is silent.

Because open and filtered ports both look like "silence", tools like nmap often have to wait or retry, making UDP scanning slower and less reliable.

← Day-01, All about IP Addresses Day-03, OSI Model →