What Is an IP Address and How Does It Work?

What is an IP address and what does it do?

At 3 a.m., an IP address is rarely an abstract networking concept. It is usually the address in a firewall rule, a DNS record pointing at the wrong VPS, or the reason an SSH connection is not reaching the machine you expected.

An IP address identifies a device or network interface on a local network or the internet. Routers use it to deliver packets toward the right destination. When you open a website, DNS resolves its domain name to one or more IP addresses, then your device connects to the selected address and receives the response.

When I connect to a server, the IP address is one of the first things I check. Running the right command on the wrong machine can be just as expensive as running the wrong command on the right one. An IP address identifies a computer, phone, server, or printer on a network, but it does not identify a particular person by itself.

How does an IP address work?

An IP address is not just a string of numbers. It tells the network where a destination belongs and which interface should receive traffic. Routers inspect destination IP addresses and forward packets between networks.

When you visit a website hosted on a VPS, the usual sequence looks like this:

1. Your browser asks a DNS resolver for the domain’s IP address.
2. Your operating system and router determine a route to that destination.
3. The client communicates using a transport protocol such as TCP or UDP.
4. The web server processes the request and sends the response back to the client’s IP address.

The IP address gets you to the machine, not to a particular service on it. HTTP and HTTPS commonly use ports 80 and 443, while SSH commonly uses port 22. In VPS administration, this distinction appears constantly: the address selects the host, and the port selects the service.

That distinction saved me from chasing the wrong problem more than once. A reachable server does not prove that nginx is listening, and an open port does not prove that the application behind it is healthy.

What is the difference between IPv4 and IPv6?

IPv4 is still the version you encounter most often. It uses 32-bit addresses written as four dot-separated numbers:

192.0.2.15

Each section can range from 0 to 255. IPv4 provides roughly 4.3 billion possible addresses in theory, although reserved, private, and otherwise unavailable ranges mean that not all of them can be assigned directly on the public internet.

IPv6 was developed to provide a much larger address space. It uses 128-bit hexadecimal addresses:

2001:db8:1234:5678::25

The 2001:db8::/32 range is reserved for documentation and examples, so it does not represent a real server. Consecutive zero groups in IPv6 addresses can be shortened with ::. That is why a full address can often be written in a much shorter form.

When you order a VPS, checking only IPv4 support is not enough for every setup. If IPv6 is enabled, check DNS records, firewall rules, and web server configuration for both protocols. Forgetting IPv6 while writing rules only for IPv4 can leave an unexpected path into the server.

That last part is easy to miss because an IPv4-only test still passes. The server appears secure until somebody connects over IPv6.

What is the difference between a public and private IP address?

A public IP address is routable on the internet and can be reached from external networks, subject to firewall rules. The public IPv4 address assigned to a VPS or hosting account usually falls into this category. An A record for a website commonly points to that address.

A private IP address is used inside a local network. The common private IPv4 ranges are:

• 10.0.0.0/8
• 172.16.0.0/12
• 192.168.0.0/16

Private addresses are not routed directly across the public internet. A home router usually uses NAT to let several private addresses share one public address. From outside, your phone, laptop, and smart TV may all appear to use the same public IP.

Loopback is another special case. In IPv4, 127.0.0.1 refers to the local machine; in IPv6, the equivalent is ::1. An application listening only on localhost cannot be reached from outside. That is a useful default for databases and other internal services.

I have caught myself testing a service on 127.0.0.1 and assuming the network path was working. It only proved that the local process could answer. Small distinction, large difference.

What are static and dynamic IP addresses?

A static IP is expected to remain unchanged. It is useful for web servers, VPN endpoints, mail servers, and systems managed remotely because DNS can continue pointing to the same address.

A dynamic IP can change when a connection is established or when a DHCP lease is renewed. Home internet connections commonly use this model, so an internet provider may assign a different public address after a reconnect.

Dynamic DNS can help with a home server. An update client changes a DNS record when the public IP changes, but the client still needs to be secured, its API key protected, and DNS caching taken into account.

It is not magic. If the update client has an API key with more permissions than it needs, a convenient home-server fix can become a domain-management problem.

How can you find your IP address?

The answer to “what is my IP address?” depends on which address you mean. Your device’s local address and the public address visible to websites are often different.

Finding a local IP address on Linux

On a Linux server, I usually start with ip addr. Older tutorials often use ifconfig, but the iproute2 tools are the normal default on current Debian, Ubuntu, and many other distributions.

ip addr show

Lines beginning with inet show IPv4 addresses, while inet6 lines show IPv6 addresses. The 127.0.0.1 entry belongs to loopback. For the actual network interface, look for names such as eth0 or ens3.

For a shorter view:

ip -br addr

The -br flag gives a brief summary. It is useful when I need to see which interface has which address without reading every detail of the interface configuration.

Short is good here.

Finding a public IP address on Linux

To see which address a server uses when it reaches the internet, query an external service:

curl -4 https://icanhazip.com
curl -6 https://icanhazip.com

The first command checks IPv4 and the second checks IPv6. If the server has no IPv6 connectivity, the second command may fail without indicating a wider server problem. A DNS-based query is another option for IPv4:

dig +short myip.opendns.com @resolver1.opendns.com

When DNS is involved, I go to the terminal before opening a provider panel. I still compare the result with the provider’s information and the addresses configured on the server rather than treating one external response as final proof.

For IPv6, the curl -6 check is usually clearer because it explicitly tests an IPv6 connection.

Finding an IP address on Windows

Open Command Prompt and run:

ipconfig

The IPv4 Address line shows the local IPv4 address, while Default Gateway shows the router used to leave the network. For more DNS and DHCP details:

ipconfig /all

In PowerShell, you can use:

Get-NetIPAddress

You may see several virtual adapters. Docker, Hyper-V, VPN clients, and WSL can all create their own interfaces, so check the interface name and address scope together.

Finding an IP address on macOS

macOS shows the address in System Settings under the relevant network connection. For Wi-Fi, the usual terminal command is:

ipconfig getifaddr en0

Wired connections may use en1 or another interface name. Check ifconfig instead of assuming the name, especially on a different Mac model.

Interface names are clues, not guarantees.

How do you find a phone’s IP address?

On Android, open the details for the connected Wi-Fi network. On an iPhone, go to Settings > Wi-Fi and tap the information icon beside the network.

The address shown there is usually in the 192.168.x.x or 10.x.x.x range. To see the public address, open an IP-checking service in the browser. With mobile data, that address may be shared with other customers through the carrier’s NAT infrastructure.

What is the difference between an IP address and a domain name?

An IP address is a numerical network address. A domain name is a human-friendly name that DNS maps to one or more IP addresses. When you enter example.com, DNS finds the address or addresses where the service can be reached.

This is different from the topic covered in Domain Extensions Explained: .com vs .net vs .org: the extension is the final part of a domain name, while the IP address identifies the machine or service endpoint used for the connection.

A domain can have several IP addresses for load balancing, geographic distribution, or redundancy. Conversely, several domains can share one IP address. The web server distinguishes them using the HTTP Host header and, for TLS, SNI.

That is why opening an IP address directly can show the wrong website or a certificate warning. The server may need the domain name to select the correct virtual host and certificate.

How does DNS find an IP address?

DNS is a distributed naming system that maps domain names to IP addresses. A browser may first check the operating system cache, then ask the local resolver. If the resolver has no cached answer, the request continues toward the domain’s authoritative DNS servers.

To query an IPv4 record:

dig +short A example.com

For IPv6, query the AAAA record:

dig +short AAAA example.com

An empty response may mean that the record does not exist. A response alone does not prove that the website works: the server must listen on ports 80 or 443, the firewall must allow the traffic, and the web server must have the correct virtual host configuration.

After changing DNS, you may continue seeing the old address because of cached records and their TTL. I ran into this while changing the WireGuard endpoint in my Proxmox lab. I updated DNS, but a client continued using its cached answer for several minutes. Comparing resolvers with dig +short showed that the problem was caching, not the server. I now check DNS changes from more than one resolver and, when possible, from another network.

That small check has saved me from undoing a working server configuration just because one resolver was behind.

How do you check a server’s IP configuration?

A provider panel shows the address assigned to a VPS, but it does not prove that the operating system configured the address correctly or has a usable route. I check both.

ip -br addr
ip route

The default via line in ip route shows the default gateway. Without a default route, a server may see its local interface but still be unable to reach the internet.

To see which interface and source address would be used for a destination:

ip route get 1.1.1.1

This is particularly useful on machines with multiple interfaces or both IPv4 and IPv6 enabled. Choosing an unexpected source address can make an application connect from an address the remote service does not allow.

When I first configured WireGuard at home, the tunnel looked active but services on the internal network were unreachable. I had written the peer configuration correctly and stopped there. Looking at ip route alongside wg show exposed the missing route:

ip route get 192.168.1.10
wg show

An interface being UP does not mean traffic is going where you expect. For VPNs, multiple network cards, or IPv6, I check the source address, route, and firewall together.

The mistake was mine: I had checked the handshake and treated it as proof that routing worked. It was only proof that the peers could exchange WireGuard traffic.

What does an IP address mean for security?

An IP address is not a password. When you connect to a website, API, or SSH service, the other side normally sees the source address as part of the connection. It can be used to identify a traffic source, apply rules, or target a service.

An address may reveal limited information such as:

• The internet provider or data-centre network that owns the range.
• An approximate geographic region, though not reliably a building or exact address.
• Which ports are exposed on the address.
• Whether the address appears in abuse or spam databases.

An IP address alone does not reveal a person’s name or home address. A precise identification requires additional information such as provider records, timestamps, and legal processes. A VPN or proxy changes the visible exit address, but it does not guarantee complete anonymity.

For a server, sensible basics include:

• Close unused ports and allow only required traffic through the firewall.
• Disable SSH password authentication where practical and use keys.
• Do not mistake moving SSH away from port 22 for real security; it mainly reduces log noise.
• Use Fail2ban or suitable nftables rules to limit repeated failed logins.
• Do not depend on hiding an IP address to secure a web application.

IP allowlists are useful for administration panels and database services. For a home connection with a changing address, giving management access through WireGuard is often easier to maintain than constantly changing an allowlist. My Raspberry Pi handles that VPN, and I keep the internal services off the public internet.

Changing the SSH port is fine if it makes logs quieter. It is not a security plan.

What happens when an IP address changes?

A public IP change can affect every service tied to the old address. If DNS still points to the previous address, visitors may reach the wrong server or fail to connect. Firewall allowlists, licensing systems, webhooks, and remote database rules may also contain the old address.

During a VPS migration or provider change, I check the following:

1. Record the new IPv4 and IPv6 addresses.
2. Compare the DNS zone’s A and AAAA records.
3. Update firewall rules, application settings, and external allowlists.
4. Use dig +short with different resolvers.
5. Test web, API, email, and SSH access separately.

A lower DNS TTL can make a transition faster, but it cannot force every client to forget its cached answer immediately. For an important migration, leaving the old server available for a while is safer than shutting it down the moment the DNS record changes.

If you are new to VPS hosting, VPS Bandwidth Explained: How Much Do You Need? is useful when thinking about traffic alongside network addressing. For the server itself, the checks in 10 Essential Steps to Secure and Harden Your Linux Server belong beside the network configuration, not after it.

What mistakes do people make with IP addresses?

The first mistake is confusing a local IP with a public one. Trying to reach 192.168.1.20 from the internet will not work unless the router has suitable port forwarding or a VPN configuration.

The second is testing a DNS change only in your own browser. Browser and local resolver caches can mislead you. Use dig, test from another network, and check the server logs together.

The third is treating an IP change as a security measure. Moving SSH from port 22 may reduce scanning noise, but it does not repair outdated software, weak passwords, or overly broad firewall rules. Restarting a server without reading its logs is not diagnosis; changing its IP and hoping the problem disappears is not diagnosis either.

The fourth is forgetting IPv6. A firewall rule written for IPv4 does not automatically protect IPv6. If you do not use IPv6, make that an intentional provider and operating-system configuration. If you do use it, write and test rules for both protocols.

I have made the second mistake myself: I trusted what my browser displayed and started looking at the application before checking the DNS answer. The terminal gave me the answer in one line. I should have started there.

When an IP address looks correct but a local connection still fails, I check how the address maps to a device on the network; my guide, What Is ARP? Address Resolution Protocol Explained, walks through that process and the Linux commands I use to troubleshoot it.

Frequently asked questions

Does an IP address reveal my personal information?

It can usually indicate an internet provider and an approximate region, but it does not reveal your name or exact address by itself. More data, including provider records and timestamps, is needed for a precise association.

Why does my IP address keep changing?

Home and mobile providers often assign dynamic addresses. A reconnect, DHCP lease renewal, or change in the provider’s network can result in a different public IP.

Why are my local and public IP addresses different?

The local address identifies your device inside a home or office network. The router uses NAT to let that device reach the internet through a shared public address.

Can I hide my IP address?

A VPN or proxy usually makes the visited service see the VPN or proxy address instead of your original one. It changes the visible exit point, but it does not guarantee complete anonymity or erase every trace of a connection.

When an IP-related problem appears, I start with three things: the address on the interface, the route selected for the destination, and the DNS answer the client actually receives. That order is less exciting than restarting the server, which is exactly why it tends to work.