Ravindra BagaleCourses & study guides मराठी Track your progress

Chapter 1: Networking Fundamentals

1.4 IPv4 in detail

By Last updated:

Friends, pay attention. This is the most important section of the chapter, so focus completely. Once IPv4 is clear, CIDR and subnets in AWS will feel very natural. Don't worry – take a pen and paper and do the calculations with me.

Structure

An IPv4 address is 32 bits, written as four octets (8 bits each) in dotted-decimal form. Each octet ranges from 0 to 255.

     192    .    168    .     1     .    10
  11000000   10101000   00000001   00001010
  |<-8 bits->|
  |<---------------- 32 bits ---------------->|

Total possible IPv4 addresses = 2^32 ≈ 4.3 billion — not enough for today's world of phones, laptops and IoT devices. That is the main reason for NAT, private IPs and IPv6.

Every IPv4 address has two parts:

  • Network portion — identifies the network (like the street name).
  • Host portion — identifies the device on that network (like the house number).

The subnet mask tells which bits are network and which are host. Example: 255.255.255.0 means the first 24 bits are network bits.

Classful addressing (historical but asked in exams)

Class First octet range Leading bits Default mask Networks / Hosts per network Use
A 1 – 126 0 255.0.0.0 (/8) 126 / ~16.7 million Very large networks
B 128 – 191 10 255.255.0.0 (/16) 16,384 / 65,534 Medium networks
C 192 – 223 110 255.255.255.0 (/24) ~2 million / 254 Small networks
D 224 – 239 1110 — — Multicast
E 240 – 255 1111 — — Experimental / reserved

What about 127?

127.0.0.0/8 is reserved for loopback. 127.0.0.1 (also called localhost) always means "this same machine". When an app listens on 127.0.0.1:3000, it is reachable only from the server itself — that is exactly what we want behind a reverse proxy.

CIDR (Classless Inter-Domain Routing)

Classes waste addresses (a company needing 500 hosts would get a whole Class B of 65,534). CIDR replaced classes. We write the address followed by /n, where n = number of network bits.

CIDR Subnet mask Total addresses Usable hosts (traditional) Usable in AWS subnet
/16 255.255.0.0 65,536 65,534 65,531
/20 255.255.240.0 4,096 4,094 4,091
/24 255.255.255.0 256 254 251
/25 255.255.255.128 128 126 123
/26 255.255.255.192 64 62 59
/27 255.255.255.224 32 30 27
/28 255.255.255.240 16 14 11
/32 255.255.255.255 1 single host —

Formula: total addresses = 2^(32 − n), usable hosts = 2^(32 − n) − 2 (network address and broadcast address are reserved).

AWS reserves 5 addresses

In every AWS subnet the first four and the last IP are reserved: network address, VPC router (.1), DNS (.2), future use (.3) and the last (broadcast) address. So a /24 subnet gives 251 usable IPs. The smallest subnet AWS allows is /28 and the largest VPC is /16.

0.0.0.0/0 means "every IPv4 address" — you will see it in security groups ("anywhere") and route tables (default route).

Suppose the Mauli Dairy office Wi-Fi network is 192.168.1.0/24 – that means the first 24 bits belong to the network and the last 8 bits to the devices, giving 254 usable hosts. If the same block were an AWS subnet, you would get 251 usable IPs, because AWS reserves 5 addresses in every subnet.

Worked subnetting example

Come on, let's work through an example together now – take a pen and paper, don't just read.

Problem: Your college lab is given the network 192.168.10.0/24. Divide it into 4 equal subnets for four labs. Find each subnet's network address, usable range and broadcast address.

Step 1 — borrow bits. 4 subnets need 2 extra bits (2² = 4). New prefix = 24 + 2 = /26.

Step 2 — block size. Addresses per subnet = 2^(32−26) = 64. Mask = 255.255.255.192.

Step 3 — list subnets (increase the last octet in steps of 64):

Subnet Network address First usable Last usable Broadcast Usable hosts
Lab 1 192.168.10.0/26 192.168.10.1 192.168.10.62 192.168.10.63 62
Lab 2 192.168.10.64/26 192.168.10.65 192.168.10.126 192.168.10.127 62
Lab 3 192.168.10.128/26 192.168.10.129 192.168.10.190 192.168.10.191 62
Lab 4 192.168.10.192/26 192.168.10.193 192.168.10.254 192.168.10.255 62

Step 4 — check with binary. Host 192.168.10.100: last octet 100 = 01100100. The first two bits (01) are the subnet bits → subnet 1 (counting from 0) → it belongs to Lab 2 (192.168.10.64/26).

Quick trick for block size

Block size = 256 − (last non-255 octet of the mask). For /26 the mask is 255.255.255.192, so block = 256 − 192 = 64. Subnets start at 0, 64, 128, 192.

AWS example: A VPC 10.0.0.0/16 is commonly split into subnets 10.0.1.0/24 (public, AZ-a), 10.0.2.0/24 (public, AZ-b), 10.0.11.0/24 (private, AZ-a) and 10.0.12.0/24 (private, AZ-b).

You can check subnet maths on Linux with the ipcalc tool (package ipcalc):

ipcalc 192.168.10.64/26

Ravindra Bagale's Tip

Subnetting comes with practice, not with reading. For one week, solve five problems every day – using the block-size trick (256 − mask octet) – and check every answer with ipcalc. In interviews you are very often asked to split a /24 into four subnets on the whiteboard – you will be ready.

Ravindra Bagale's Tip

Many students say a /26 has 64 usable hosts. Wrong! 64 is the total number of addresses – subtract the network and broadcast addresses and you get 62, and in an AWS subnet 59. Remember the formula: 2^(32 − n), then −2 (−5 in AWS).

Lab

Chala, 10.0.0.0/24 che 4 equal subnets kara: navin prefix, block size, pratyek subnet cha network address, usable range aani broadcast liha. Mag ipcalc 10.0.0.64/26 chalvun tumchya dusrya subnet che uttar match kara.

Got it? IPv4 = 32 bits, CIDR /n = network bits, and the block-size trick – once these three things are solid, subnetting will feel easy. Now let's move on to IPv6.