What Is an IP Address? IPv4 and IPv6 Explained for Server Beginners

An IP address is one of the basic pieces of information that makes network communication possible. When a server sends a packet to another system, the network needs an address that identifies where that packet came from and where it should go.

For server administrators, understanding IP addresses is more useful than simply memorizing what an address looks like. You need to know why an address belongs to a particular network, what the prefix length means, why a server can have several IP addresses, and why IPv4 and IPv6 look completely different.

🌐 What Is an IP Address?

IP stands for Internet Protocol. At a basic level, IP provides addressing and packet-delivery functions between interconnected networks.

An IP address gives the network layer an identifier that can be used as the source or destination of IP packets. This is why an IP address is fundamental to servers: a remote client needs a network-level destination for traffic intended for that server.

Think of the difference between a website name and an IP address this way:

A hostname such as example.com is a human-friendly name. DNS can map names to IP addresses, while IP itself deals with addressing packets. The two concepts are related, but they are not interchangeable.

🧩 What Does an IP Address Identify?

A common beginner shortcut is to say that an IP address identifies a computer. That is useful for a first approximation, but it is not always precise.

IP addressing is associated with network interfaces and IP communication. A single physical or virtual machine can have multiple interfaces and multiple IP addresses. A server may therefore have a public IPv4 address, one or more private addresses, and one or more IPv6 addresses at the same time.

🔢 IPv4: The Familiar Four-Part Address

IPv4 uses a 32-bit address. It is commonly represented as four decimal octets separated by dots.

192.0.2.10

Each octet represents 8 bits, so the four parts together make 32 bits.

The address format is easy to read, but the four numbers do not independently tell you which part is the network and which part identifies a host. That depends on the prefix length or subnet mask.

🧱 What Does /24 Mean?

Modern IP networks commonly use CIDR, or Classless Inter-Domain Routing, notation. You will often see an address written like this:

192.0.2.10/24

The /24 is the prefix length. It says that the first 24 bits belong to the network prefix, leaving the remaining 8 bits outside that prefix.

You can think of the notation as:

That visual is simplified, but it captures the core idea: the prefix defines which network an address belongs to, while the remaining bits distinguish addresses within that prefix.

CIDR is important because routers and operating systems use prefixes to understand network boundaries and routing decisions. The same basic idea applies to IPv4 and IPv6.

🔐 Public vs Private IPv4 Addresses

Not every IPv4 address is globally routable on the public Internet.

IPv4 has address space reserved for private networks. The three private-use ranges defined by RFC 1918 are:

Private addresses are intended for use inside private networks and are not globally unique Internet addresses. A home router, office network, or internal server network can use them without coordinating those addresses globally.

A public IP address, by contrast, is address space intended for use where global Internet routing is appropriate. The important distinction is not simply “one is safe and one is unsafe”; it is about addressing scope and routability.

🔄 One Server Can Have Multiple IP Addresses

This is normal in server environments.

Consider a Linux server with:

eth0    203.0.113.20
eth1    10.10.0.20
eth0    2001:db8:10::20

The example shows three addresses associated with the server’s network configuration:

• 203.0.113.20 represents an IPv4 address from a documentation range used for examples.
• 10.10.0.20 is a private IPv4 address.
• 2001:db8:10::20 is an IPv6 documentation address.

The actual routing and interface configuration determines how traffic uses these addresses. A server does not automatically send every packet through every address it owns.

🚀 IPv6: Why Is It So Much Longer?

IPv6 was designed as the successor to IPv4 and uses 128-bit addresses instead of IPv4’s 32 bits. The larger address size provides a vastly larger address space and supports a different addressing architecture.

A conventional IPv6 address looks like this:

2001:0db8:0000:0000:0000:0000:0000:0020

IPv6 represents its 128 bits as eight groups of hexadecimal digits. Leading zeroes can be omitted within a group, and one consecutive run of all-zero groups can be compressed using ::.

The previous example can therefore be written as:

2001:db8::20

The compression is only a notation rule. It does not make the address shorter internally; the underlying address remains 128 bits.

📊 IPv4 vs IPv6

IPv6 is not simply “IPv4 with bigger numbers.” It changes the address format and includes different addressing mechanisms. For example, IPv6 defines unicast, anycast, and multicast addressing, and IPv6 does not use broadcast addresses in the IPv4 sense.

📍 Important IPv6 Address Types

IPv6 addresses have different roles and scopes. Beginners do not need to memorize every special range, but three concepts are especially useful when administering servers.

IPv6 addressing is defined by several standards, so server administrators should rely on the actual address configuration and prefix rather than trying to infer an address’s role from appearance alone.

🧭 What Happens When a Server Sends a Packet?

An IP address does not contain the entire route to a destination. It is an addressing component used by the IP layer. Routing systems determine where packets should be forwarded next.

A simplified model looks like this:

Suppose a server wants to communicate with a remote destination. The IP layer places source and destination IP addresses into the IP packet. The local system then uses its routing information to determine how the packet should leave the machine.

Routers along the path examine the destination address and make forwarding decisions. The packet can therefore cross multiple networks without every intermediate router needing to know the complete end-to-end application context.

🧪 Mini Lab: Inspect IP Addresses on Linux

Goal: identify the IP addresses currently configured on a Linux system and see which interfaces own them.

The standard ip utility can display addresses assigned to network interfaces. The following command only reads configuration; it does not change the system.

ip address

Look for interface sections containing lines beginning with inet and inet6.

A simplified example might look like:

inet 192.168.1.20/24
inet6 2001:db8:1::20/64

From this output, you can identify:

• Address family: inet indicates IPv4 and inet6 indicates IPv6.
• Address: the value before the slash.
• Prefix length: the number after the slash.
• Interface: the network interface section containing the address.

You can also ask the system which route it would use for a particular destination:

ip route get 1.1.1.1

This is useful because it moves the investigation from “What addresses does this server have?” to “Which route would the kernel choose for this destination?” The exact output depends on the server’s routing configuration.

⚠️ Common Mistakes

💡 Practical Notes for Server Administrators

When looking at an IP address on a server, avoid asking only “What is this number?” A more useful investigation asks several questions:

This way of thinking is especially useful when diagnosing connectivity problems. If a service is unreachable, the IP address itself is only one part of the investigation. You may also need to examine routing, firewall rules, transport ports, DNS, and the service listening on the destination.

✅ Knowledge Check