Most Ethernet configuration is centralized in a single file, /etc/network/interfaces. If you have no Ethernet devices, only the loopback interface will appear in this file, and it will look something like this:

# This file describes the network interfaces available on your system
# and how to activate them. For more information, see interfaces(5).

# The loopback network interface
auto lo
iface lo inet loopback
address 127.0.0.1
netmask 255.0.0.0

If you have only one Ethernet device, eth0, and it gets its configuration from a DHCP server, and it should come up automatically at boot, only two additional lines are required:

auto eth0
iface eth0 inet dhcp

The first line specifies that the eth0 device should come up automatically when you boot. The second line means that interface (“iface”) eth0 should have an IPv4 address space (replace “inet” with “inet6” for an IPv6 device) and that it should get its configuration automatically from DHCP. Assuming your network and DHCP server are properly configured, this machine's network should need no further configuration to operate properly. The DHCP server will provide the default gateway (implemented via the route command), the device's IP address (implemented via the ifconfig command), and DNS servers used on the network (implemented in the /etc/resolv.conf file.)

To configure your Ethernet device with a static IP address and custom configuration, some more information will be required. Suppose you want to assign the IP address 192.168.0.2 to the device eth1, with the typical netmask of 255.255.255.0. Your default gateway's IP address is 192.168.0.1. You would enter something like this into /etc/network/interfaces:

iface eth1 inet static
	address 192.168.0.2
	netmask 255.255.255.0
	gateway 192.168.0.1

In this case, you will need to specify your DNS servers manually in /etc/resolv.conf, which should look something like this:

search mydomain.example
nameserver 192.168.0.1
nameserver 4.2.2.2

The search directive will append mydomain.example to hostname queries in an attempt to resolve names to your network. For example, if your network's domain is mydomain.example and you try to ping the host “mybox”, the DNS query will be modified to “mybox.mydomain.example” for resolution. The nameserver directives specify DNS servers to be used to resolve hostnames to IP addresses. If you use your own nameserver, enter it here. Otherwise, ask your Internet Service Provider for the primary and secondary DNS servers to use, and enter them into /etc/resolv.conf as shown above.

Many more configurations are possible, including dialup PPP interfaces, IPv6 networking, VPN devices, etc. Refer to man 5 interfaces for more information and supported options. Remember that /etc/network/interfaces is used by the ifup/ifdown scripts as a higher level configuration scheme than may be used in some other Linux distributions, and that the traditional, lower level utilities such as ifconfig, route, and dhclient are still available to you for ad hoc configurations.

This section explains how to configure which nameserver to use when resolving IP addresses to hostnames and vice versa. It does not explain how to configure the system as a name server.

To manage DNS entries, you can add, edit, or remove DNS names from the /etc/resolv.conf file. A sample file is given below:

search com
nameserver 204.11.126.131
nameserver 64.125.134.133
nameserver 64.125.134.132
nameserver 208.185.179.218

The search key specifies the string which will be appended to an incomplete hostname. Here, we have configured it to com. So, when we run: ping ubuntu it would be interpreted as ping ubuntu.com.

The nameserver key specifies the nameserver IP address. It will be used to resolve a given IP address or hostname. This file can have multiple nameserver entries. The nameservers will be used by the network query in the same order.

To manage hosts, you can add, edit, or remove hosts from /etc/hosts file. The file contains IP addresses and their corresponding hostnames. When your system tries to resolve a hostname to an IP address or determine the hostname for an IP address, it refers to the /etc/hosts file before using the name servers. If the IP address is listed in the /etc/hosts file, the name servers are not used. This behavior can be modified by editing /etc/nsswitch.conf at your peril.

If your network contains computers whose IP addresses are not listed in DNS, it is recommended that you add them to the /etc/hosts file.

Bridging multiple interfaces is a more advanced configuration, but is very useful in multiple scenarios. One scenario is setting up a bridge with multiple network interfaces, then using a firewall to filter traffic between two network segments. Another scenario is using bridge on a system with one interface to allow virtual machines direct access to the outside network. The following example covers the latter scenario.

Before configuring a bridge you will need to install the bridge-utils package. To install the package, in a terminal enter:

sudo apt-get install bridge-utils

Next, configure the bridge by editing /etc/network/interfaces:

auto lo
iface lo inet loopback

auto br0
iface br0 inet static
        address 192.168.0.10
        network 192.168.0.0
        netmask 255.255.255.0
        broadcast 192.168.0.255
        gateway 192.168.0.1
        bridge_ports eth0
        bridge_fd 9
        bridge_hello 2
        bridge_maxage 12
        bridge_stp off

Now restart networking to enable the bridge interface:

sudo /etc/init.d/networking restart

If setting up a bridge interface using Ubuntu Desktop Edition, or if dhcdbd is installed, the dhcdbd daemon will need to be stopped and disabled.

After configuring the bridge in /etc/network/interfaces, shutdown dhcdbd by:

sudo /etc/init.d/dhcdbd stop

Now to disable it from starting on boot enter:

sudo update-rc.d -f dhcdbd remove

The new bridge interface should now be up and running. The brctl provides useful information about the state of the bridge, controls which interfaces are part of the bridge, etc. See man brctl for more information.