Assumptions for Network Setup

In the demonstrations in this chapter, we've used a network setup that may not be the same as your own environment. You can choose your own server, connectivity devices (switch, hub, and direct cable connection), IP address scheme, and hostname to set up networking in a way that suits your own requirements. In this section, we'll note the network settings that we've used. In order to facilitate, we'll begin with an explanation of the terms that we'll use.

The network topology refers to the shape or layout of the network. It defines how different systems in a network are connected and communicate with each other. The ideal topology of a network depends on the size of the network, the requirements of its users, and the policies of the company or organization.

The following diagram shows an ideal network topology:



OSWWxtaM XP ftofeuioral 0S: Racruit llou, g 0

This diagram shows a client machine (in the form of a Windows 2000 laptop), and a server (running Red Hat Linux 9, and supporting some of the services described in this chapter. They are connected via an internal network, and protected from the Internet by a hub and a firewall. Of course, your network topology might look different than this.

For example, if you have a smaller network, then it's possible that you don't have a hub, and that your firewall is not a dedicated machine but a piece of software hosted by your Linux server machine. If you have only a single machine, then it can double as both the server and client machines in this diagram (being a client of its own services!). We've used a topology like this one in this chapter.

Note that both the end-user's laptop and the server machine have been assigned an IP address - this is used to uniquely identify the computer (or other device) within a TCP/IP network. An IP address is a 32-bit binary address in the form of four decimal values. Each of these four decimals represents an 8-bit value (an octet), and hence is in the range 0 to 255. This format is known as dotted quad notation.

In fact, the TCP/IP network is divided into different class networks. The important classes are defined below:




Class A

l.x.x.x to 127.x.x.x

The first octet specifies the network number, and the remaining octets specify the host within the network. This class supports about 1.6 million hosts per network.

Class B

128.0.x.x to 191.255.x.x

The first two octets specify the network number, and the remaining octets specify the host within the network. This class supports 16,320 networks with 65,024 hosts each.

Class C

192.0.0.x to 223.255.255.x

The first three octets specify the network number, and the remaining octets specify the host within the network. This class supports nearly two million networks with up to 254 hosts each.

Class D to

These are reserved for multicast groups

Class E to

These are reserved for special purpose use.

The TCP/IP networks can also be divided by public and private IP addresses:

• Your public IP address is assigned to you by your Internet service provider (ISP). In fact, this IP address is allocated to the ISP by the Internet Assigned Numbers Authority (IANA - It's your public IP address that identifies you on the Internet.

• Private IP addresses are those (like the addresses used in the diagram above) that identify a device within a private or non-Internet connected network. There are three TCP/IP network address ranges reserved for use in private networks:,, and

A netmask is a 32-bit string that hides the network part of an IP address, so that only the host (computer) part of the address remains. For example, the netmask looks like an IP address, but in fact hides the first 24 bits of the IP address so that only the last 8 bits remain (recall that 255 is the decimal representation of the binary 11111111). Hence, the netmask is commonly used for Class C IP addresses to reveal the specific host computer address publicly.

There are a few more definitions that will be useful here:

• A gateway is a host that is connected to two or more physical networks, and hence allows messages to switch between them.

• The term hostname refers to the unique name of the machine, so that it can identify itself on the network.

• Finally, the domain name system (or DNS) is a distributed database that translates domain names into IP addresses and vice versa. For example, DNS translates the domain name to the IP address

With all that in mind, let's return to those two machines on our network - the Red Hat Linux 9 server machine and the Windows 2000 client laptop machine. Here are the network configuration settings we used for those two machines in this chapter.

The Red Hat Linux 9 server machine is set up as a firewall and router, and has Internet connectivity. Over the course of the chapter we'll show how it can be configured as a file server, printer server, DHCP server, FTP server, web server, or mail server:



Operating system

Red Hat Linux 9

IP address




The laptop is a simple client machine, designed for an end-user and (potentially) sharing resources with other similar machines on the network:




A Client laptop system to be used to share resources configured on Linux server

Operating system

Windows 2000 Professional

IP address


As we've said, your network topology, and the network configuration of your machine(s), may be different; but many of the principles in this chapter remain the same, and you should be able to get any of these services going regardless of your own environment.

We will assume that you have root access to the Red Hat Linux server. Throughout the chapter, we'll use the terms Linux server and client to refer to the Red Hat Linux 9 server machine and the Microsoft Windows 2000 client machine respectively.

Was this article helpful?

0 0

Post a comment