Networking for Sysadmins 3: IPv4 — Class Notes

2026-09-27 · class-notes, networking, ipv4, subnetting, netmask, nat, troubleshooting

Full class notes for Networking for Sysadmins ch. 3: IPv4 addressing, subnets/netmasks, viewing config on Windows/Debian/FreeBSD, private addressing and NAT, and troubleshooting.

How these notes were made: the material supplied for this lesson was the chapter's own heading list — "IPv4 Addresses", "Special IPv4 Addresses", "The Localhost Address", "Subnets", and so on — with an explicit instruction to draft the content from general knowledge rather than a transcript. These notes follow that heading structure, but the explanations are written from general networking knowledge, not reproduced or paraphrased from Michael W. Lucas's actual book text, which was not supplied. Treat this as original coverage of the same topics, organised the way the book organises them — worth checking against your own copy of the chapter for Lucas's specific examples, wording, and any details particular to his presentation.

Class notes · Networking for System Administrators (2nd ed.), Michael W. Lucas · Chapter 3

Quick reference: the short version of this lesson lives in the Networking for Sysadmins resource sheets. This follows Chapter 2 (Ethernet) — Ethernet gets frames onto the local wire; this chapter is about the addressing layered on top of that, IPv4.

Learning objectives

By the end of these notes you should be able to:

1. Describe the structure of an IPv4 address and identify special/reserved addresses, including localhost.
2. Explain subnetting, netmasks, and CIDR notation, and identify the unusable addresses in a subnet.
3. Explain the role of a router and default gateway, and distinguish a LAN from a routed network.
4. View IP configuration on Windows, Debian and FreeBSD.
5. Explain multiple network interfaces and IP aliasing.
6. Explain private addressing and NAT.
7. Apply a basic IP troubleshooting approach.

1. IPv4 addresses

An IPv4 address is a 32-bit number, conventionally written as dotted decimal notation: four numbers from 0–255, separated by periods — for example 192.168.1.10. Each of those four numbers represents one octet (8 bits), and together the four octets make up the full 32-bit address.

An IPv4 address by itself does not say which part identifies the network and which part identifies the host on that network — that division is supplied separately, by the subnet mask (Section 4).

2. Special IPv4 addresses

Certain addresses and ranges are reserved for specific purposes rather than being assigned to ordinary hosts:

3. The localhost address

127.0.0.1 is the best-known loopback address — a way for a machine to address itself over IP, without traffic ever leaving the machine or touching a real network interface. In fact, the entire 127.0.0.0/8 range (127.0.0.1 through 127.255.255.254) is reserved for loopback use, though 127.0.0.1 is the address almost universally used in practice.

Loopback is heavily used by services that run and communicate locally on one machine — a database only accepting connections from 127.0.0.1, for instance, is deliberately refusing any connection that did not originate on that same machine.

4. Subnets

A subnet is a logically distinct portion of a larger IP network. Rather than treating every device as part of one enormous flat address space, subnetting divides the address space into smaller segments — useful for separating traffic (a staff subnet vs a guest subnet), for limiting the size of broadcast domains, and for organising a network in a way that mirrors an organisation's actual structure.

5. Netmasks and subnet size

A subnet mask (or netmask) marks which bits of an IP address belong to the network portion and which belong to the host portion. It is written either in the same dotted-decimal form as an IP address (e.g. 255.255.255.0) or as CIDR notation — a slash followed by the number of network bits (e.g. /24).

The larger the CIDR number (more network bits), the fewer hosts a subnet can hold. A /24 (255.255.255.0) is a very common size for a small office LAN, giving 254 usable host addresses.

6. Unusable IPv4 addresses

Within any subnet, two addresses cannot be assigned to a host:

• The network address — every host bit set to 0. Identifies the subnet itself, not a device on it. For 192.168.1.0/24, that is 192.168.1.0.
• The broadcast address — every host bit set to 1. Used to reach every host on that specific subnet at once. For 192.168.1.0/24, that is 192.168.1.255.

That is why a /24 subnet, with 256 total addresses, yields only 254 usable host addresses — the first and last are reserved.

192.168.1.0/24:

  192.168.1.0    <- network address (unusable)
  192.168.1.1    <- first usable host  (often the gateway)
  ...
  192.168.1.254  <- last usable host
  192.168.1.255  <- broadcast address (unusable)

7. Routers and the default gateway

A router forwards traffic between different networks — it is the device that lets a packet leave the local subnet and reach somewhere else. When a host needs to send traffic to an address outside its own subnet, it sends that traffic to its default gateway: the router's IP address on the local subnet, configured on every host that needs to reach beyond its own network.

Without a correctly configured default gateway, a host can talk to other devices on its own subnet, but nothing beyond it.

8. Networks vs LANs and gateways

It helps to keep two related ideas distinct:

• A LAN (Local Area Network) is the local segment a host's traffic can reach directly, without needing a router — typically the same as its subnet.
• A network, more broadly, can mean the LAN, or it can mean the wider set of interconnected networks (potentially spanning many LANs joined by routers) that a host can ultimately reach.

The gateway is the specific device — a router's local interface — that bridges a host's LAN to everything beyond it. "The network" a host can reach is therefore its own LAN, plus whatever its gateway (and the routers beyond that) can forward traffic toward.

9. Viewing IP configuration

The command to check a machine's IP configuration differs by operating system.

9.1 Windows

ipconfig

For full detail, including DNS servers and adapter descriptions:

ipconfig /all

9.2 Debian (and most modern Linux)

The modern tool is ip, from the iproute2 package:

ip addr show

or shortened:

ip a

The older ifconfig command (from the net-tools package) still works on many systems, though it is considered legacy on modern Linux distributions:

ifconfig

9.3 FreeBSD

FreeBSD uses ifconfig as its standard, current tool for viewing (and configuring) interfaces — unlike on Linux, this is not a legacy command on FreeBSD:

ifconfig

Note (beyond this lesson): on FreeBSD, ifconfig with no arguments lists every interface and its assigned addresses; a specific interface can be queried with ifconfig <interface_name> (e.g. ifconfig em0).

10. Multiple network interfaces

A machine can have more than one network interface — for example, a server with both a wired Ethernet port and a separate management network port, or a router with an interface on each network it connects. A host with multiple interfaces, each potentially on a different subnet, is sometimes called multihomed.

Each interface typically gets its own IP address (and can have its own default gateway, though a host generally uses only one at a time for general-purpose routing).

11. IP aliasing

IP aliasing is assigning more than one IP address to a single network interface. A single physical (or virtual) interface can therefore answer to several addresses at once — useful for hosting multiple services that each need their own IP, without needing a separate physical interface for each.

On Linux, this is done with the ip command, adding a further address to an existing interface:

ip addr add 192.168.1.50/24 dev eth0

Note (beyond this lesson): older Linux tooling historically used a labelled sub-interface naming convention such as eth0:0, eth0:1 for aliases; modern iproute2-based tooling (as shown above) simply attaches multiple addresses to the same interface name without that labelling scheme.

12. Private addresses and NAT

Certain IPv4 ranges are set aside as private — not routable on the public internet, and free for any organisation to reuse internally without coordinating with anyone else. These are defined in RFC 1918:

Because private addresses cannot be routed on the public internet directly, a device called a NAT (Network Address Translation) gateway — typically built into a home or office router — translates between a host's private address and a single public address when traffic needs to leave the local network. This is what lets many devices on a private network share one public internet connection.

12.1 Worked example — a home network behind NAT

A laptop on a home network has the private address 192.168.1.20. It sends a request to a public web server.

1. The laptop sends the request to its default gateway, the home router, at 192.168.1.1.
2. The router's NAT function rewrites the packet's source address from 192.168.1.20 to the router's own public IP address before sending it onward to the internet.
3. The web server's response comes back addressed to the router's public IP.
4. The router's NAT table remembers which internal private address made the original request, and rewrites the response's destination back to 192.168.1.20 before delivering it to the laptop.

The laptop never has a directly internet-routable address of its own — NAT makes that unnecessary for outbound connections.

13. Troubleshooting IP

A basic, layered approach to IP-level troubleshooting:

1. Confirm the host's own configuration. Check the IP address, subnet mask and default gateway are what you expect (Section 9's commands).
2. Test the local subnet. Can the host reach another device on its own subnet? A failure here usually points to a local link or configuration problem, not a routing issue.
3. Test the gateway. Can the host reach its own default gateway? If not, nothing beyond the local subnet will work either.
4. Test beyond the gateway. Can the host reach something further away — another subnet, or the internet? A failure here, with a working gateway, points toward a routing problem beyond the local network.
5. Check DNS separately from IP reachability. A host can be fully IP-reachable and still fail to "get to a website" if DNS resolution is broken — rule this out as a distinct step rather than assuming it is an IP problem.

In the real world: working through these steps in order — local config, then local subnet, then gateway, then beyond — narrows down where a connectivity problem actually lives far faster than guessing at the far end first.

14. Security perspective

IPv4 addressing and subnetting decisions carry real security weight, beyond pure connectivity:

• Subnet boundaries are a natural place to enforce segmentation. Putting different trust levels — staff, guest Wi-Fi, IoT devices, servers — on separate subnets means a firewall or router between them can actually filter traffic; devices sharing one flat subnet can typically reach each other directly with nothing in the way.
• NAT is not a firewall, even though it feels like one. NAT exists to conserve public addresses and enable address translation, not to provide security — a host behind NAT that has a port forwarded to it, or that initiates an outbound connection, is just as reachable as if it had a direct public address for that traffic. Do not rely on "it's behind NAT" as a security boundary on its own.
• Loopback-only services are a real, simple control. A service configured to listen only on 127.0.0.1 cannot be reached from anywhere else on the network at all — a genuinely strong restriction for anything that only needs to be used locally (a local admin interface, a database only accessed by an app on the same host).
• IP aliasing and multihomed hosts widen what needs reviewing. Each additional address or interface on a host is another thing that might be listening for connections, and another thing to include when reasoning about that host's actual exposure — an audit that only checks "the" IP address of a multihomed or aliased host can miss real attack surface.
• APIPA addresses (169.254.0.0/16) are a diagnostic signal. A host showing a self-assigned link-local address has failed to reach DHCP — worth investigating not just as a connectivity fault, but as a possible sign of a DHCP outage, a rogue/exhausted DHCP scope, or (in rarer cases) DHCP starvation as part of an attack.

Summary

• An IPv4 address is a 32-bit number in dotted-decimal form; the subnet mask splits it into network and host portions.
• Special addresses include 0.0.0.0, the broadcast 255.255.255.255, the 127.0.0.0/8 loopback range (usually 127.0.0.1), and the 169.254.0.0/16 APIPA/link-local range.
• Subnetting divides an address space; netmasks/CIDR define subnet size — the network address (all host bits 0) and broadcast address (all host bits 1) in any subnet are always unusable for hosts.
• A router forwards between networks; a host's default gateway is the router it sends off-subnet traffic to. A LAN is the local, directly reachable segment; "the network" more broadly can include everything a gateway can forward toward.
• View IP config with ipconfig (Windows), ip addr show (modern Debian/Linux, with legacy ifconfig still around), or ifconfig (FreeBSD's standard tool).
• A host can have multiple interfaces (multihomed); IP aliasing assigns several addresses to one interface.
• Private addressing (RFC 1918: 10.0.0.0/8, 172.16.0.0/12, 192.168.0.0/16) plus NAT lets many private hosts share one public address.
• Troubleshoot IP in layers: local config -> local subnet -> gateway -> beyond the gateway, checking DNS separately.

Glossary

Review questions

1. What is the structure of an IPv4 address, and what determines which part is the network vs the host portion?
2. What is the loopback address, and what range is actually reserved for it?
3. What are the two unusable addresses in any subnet, and why are they unusable?
4. Convert /24 to its dotted-decimal netmask, and state how many usable host addresses it provides.
5. What is a default gateway, and what happens to off-subnet traffic without one configured correctly?
6. Distinguish a LAN from "the network" more broadly.
7. Give the command to view IP configuration on Windows, on modern Debian/ Linux, and on FreeBSD.
8. What is a multihomed host?
9. What is IP aliasing, and give a reason you might use it.
10. Name the three RFC 1918 private address ranges.
11. What does NAT do, and why is it needed for private addresses to reach the internet?
12. Describe the layered approach to troubleshooting an IP connectivity problem.
13. Scenario: a host shows an address starting 169.254. What does this indicate?
14. Scenario: a service is configured to listen only on 127.0.0.1. Can it be reached from another machine on the network? Explain.
15. Scenario: why is "the server is behind NAT" not, by itself, a security justification for skipping a firewall rule?

Answer key

1. A 32-bit address in four dotted-decimal octets; the subnet mask determines the network/host split, not the address alone. Mask defines the boundary.
2. 127.0.0.1 is the address in practice; the whole 127.0.0.0/8 range is reserved for loopback. A wider reserved block than just one address.
3. The network address (all host bits 0, identifies the subnet) and the broadcast address (all host bits 1, reaches every host on the subnet) — neither can be assigned to a single host. Structural reservations.
4. 255.255.255.0; 254 usable host addresses. Standard small-LAN size.
5. The router a host sends traffic to when the destination is outside its own subnet; without one configured, that traffic simply cannot leave the local subnet. No path beyond the LAN.
6. A LAN is the local segment reachable without routing; "the network" can also mean everything reachable via the gateway and routers beyond it. Local vs wider reachability.
7. ipconfig (Windows); ip addr show (modern Debian/Linux, ifconfig still exists as legacy); ifconfig (FreeBSD, still the standard tool there). Three different platforms, three commands.
8. A host with more than one network interface (and typically more than one IP address). Connected to multiple networks at once.
9. Assigning multiple IP addresses to a single interface; useful for hosting several services that each need their own address without extra physical interfaces. One interface, many addresses.
10. 10.0.0.0/8, 172.16.0.0/12, and 192.168.0.0/16. The RFC 1918 ranges.
11. NAT translates a private address to a public one (and back) so private hosts can communicate over the internet, which does not route private addresses directly. Address translation bridges the two.
12. Check local config, then local subnet reachability, then the default gateway, then beyond the gateway — checking DNS as a separate step. Narrow down layer by layer.
13. DHCP failed, so the host self-assigned an APIPA/link-local address — it can likely only reach other hosts on the same local segment, if that. A DHCP problem signal.
14. No — a loopback-only service is reachable only from the same machine, not from the network at all. Loopback binding is a strong local-only restriction.
15. NAT exists for address translation, not security; a port forwarded through it, or an outbound connection it permits, is just as reachable as a direct public address would be for that traffic. NAT isn't a firewall.