35. Network Devices And Configuration

Question 1

Learning Objectives

By the end of this chapter, you should be able to:

Answer 1

• Identify network devices and understand how the operating system names them and binds them to specific duties.
• Use the ip utility to display and control devices, routing, policy-based routing, and tunnelling.
• Use the older ifconfig to configure, control, and query network interface parameters from either the command line or from system configuration scripts.
• Understand the Predictable Network Interface Device Names scheme.
• Know the main network configuration files in /etc.
• Use Network Manager (nmtui and nmcli) to configure network interfaces in a distribution-independent manner.
• Know how to set default routes and static routes.​
• Configure name resolution as well as run diagnostic utilities.

Question 2

Network Devices

Unlike block and character devices, network devices are not associated with special device files, also known as device nodes. Rather than having associated entries in the /dev directory, they are known by their names.

So know this what type of network device do each of these represent?

1. eth0, eth1, eno1, eno2, etc.
2. wlan0, wlan1, wlan2, wlp3s0, wlp3s2, etc.
3. br0, br1, br2, etc.
4. vmnet0, vmnet1, vmnet2, etc.

Answer 2

1. Ethernet devices
2. wireless devices
3. bridge interfaces
4. virtual devices for communicating with virtual clients.

Question 3

ip

What wide variety of tasks can the ip command/utility be used for.

Answer 3

It can be used to configure, control and query interface parameters and control devices, routing, policy-based routing, and tunneling.

Question 4

ip

The ip utility is preferred to the venerable ___ utility because it is more versatile, as well as more efficient because it uses netlink sockets, rather than ioctl system calls.

Answer 4

ipconfig

Question 5

ip

What is the basic syntx of the ip utility? It has two different forms for different purposes what are they?

Answer 5

• ip [OPTIONS] OBJECT { COMMAND | help }
• ip [-force] -batch filename

where the second form can read commands from a designated file.

ip is a multiplex utility; the OBJECT argument describes what kind of action is going to be performed. The possible COMMANDS depend on which OBJECT is selected.

You can see below some of the main values of OBJECT.

OBJECT - FUNCTION

address - IPv4 or IPv6 protocol device address

link - Network Devices

maddress - Multicast Address

monitor - Watch for netlink messages

route - Routing table entry

rule - Rule in the routing policy database

tunnel - Tunnel over IP

Question 6

Using ip: Examples

How do you show information for all network interfaces?

Answer 6

$ ip link show

Question 7

Using ip: Examples

How do you show information for the eth0 network interface, including statistics?

Answer 7

$ ip -s link show eth0

Question 8

Using ip: Examples

How do you set the IP address for eth0 to the address 192.168.1.7?

Answer 8

$ sudo ip addr add 192.168.1.7 dev eth0

Question 9

Using ip: Examples

How do you bring interface eth0 down?

Answer 9

$ sudo ip link set eth0 down

Question 10

Using ip: Examples

How do you set MTU to 1480 bytes for interface eth0?

Answer 10

$ sudo ip link set eth0 mtu 1480

Question 11

Using ip: Examples

How do you set route from the local 192.168.1.5 address to network 172.16.1.0/24?

Answer 11

$ sudo ip route add 172.16.1.0/24 via 192.168.1.5

Question 12

ip

What is the OBJECT is used to perform actions on IPv4 or IPv6 protocol device address?

Command syntax:

ip [OPTIONS] OBJECT { COMMAND | help }

Answer 12

address

ip address …

Question 13

ip

What is the OBJECT is used to perform actions on network devices?

Command syntax:

ip [OPTIONS] OBJECT { COMMAND | help }

Answer 13

link

ip link …

Question 14

ip

What is the OBJECT is used to perform actions on multicast addresses?

Command syntax:

ip [OPTIONS] OBJECT { COMMAND | help }

Answer 14

maddress

ip maddress …

Question 15

ip

What is the OBJECT is used to perform actions to watch for netlink messages?

Command syntax:

ip [OPTIONS] OBJECT { COMMAND | help }

Answer 15

monitor

ip monitor …

Question 16

ip

What is the OBJECT is used to perform actions to mod

routing table entry?

Command syntax:

ip [OPTIONS] OBJECT { COMMAND | help }

Answer 16

route

ip route …

Question 17

ip

What is the OBJECT is used to perform actions to mod Rules in the routing policy database?

Command syntax:

ip [OPTIONS] OBJECT { COMMAND | help }

Answer 17

rule

ip rule …

Question 18

ip

What is the OBJECT is used to perform actions to tunnel over IP?

Command syntax:

ip [OPTIONS] OBJECT { COMMAND | help }

Answer 18

tunnel

ip tunnel …

Question 19

ifconfig

There was usage info but ifconfig is not installed by default on some distros in favor of ip which is more relevant as of today. This card is just a notice for this. Don’t use ifconfig use ip utility instead.

Question 20

Predictable Network Interface Device Names

Predictable Network Interface Device Names (PNIDN) is strongly correlated with the use of udev and integration with systemd. There are now 5 types of names that devices can be given:

These 5 types represent a single device but can provide different information of the device such as where it is on the motherboard or geographic location.

based on the 5 types what type device names do these examples represent?

1. eno1
2. ens1
3. enp2s0
4. enx7837d1ea46da
5. eth0

Answer 20

1. Incorporating Firmware or BIOS provided index numbers for on-board devices
2. Incorporating Firmware or BIOS provided PCI Express hotplug slot index numbers
3. Incorporating physical and/or geographical location of the hardware connection.
4. Incorporating the MAC address
5. Using the old classic method

__

EX: For a wireless device that previously would have been simply named wlan0:

$ ip link show | grep wl

3: wlp3s0: mtu 1500 qdisc mq state UP mode DORMANT qlen 1000

$ lspci | grep Centrino

03:00.0 Network controller: Intel Corporation Centrino Advanced-N 6205 [Taylor Peak] (rev 34)

It is easy to turn off the new scheme and go back to the classic names. We will leave that as a research project. In what follows, we will mostly follow the classic names for definiteness and simplicy.

Question 21

Predictable Network Interface Device Names

How do you see info on a machine with two onboard PCI network interfaces that would have been eth0 and eth1 in the old classic naming method?

Answer 21

$ ip link show | grep enp

• 2: enp4s2: mtu 1500 qdisc pfifo_fast state DOWN mode DEFAULT qlen 1000
  3: enp2s0: mtu 1500 qdisc pfifo_fast state UP mode DEFAULT qlen 1000*

__

These names are correlated with the physical locations of the hardware on the PCI system:

$ lspci | grep Ethernet

• *02:00.0** Ethernet controller: Marvell Technology Group Ltd. 88E8056 PCI-E Gigabit Ethernet Controller (rev 12)
• *04:02.0** Ethernet controller: Marvell Technology Group Ltd. 88E8001 Gigabit Ethernet Controller (rev 14)

The triplet of numbers at the beginning of each line from the lspci output is the bus, device (or slot), and function of the device; hence it reveals the physical location.

__

__

EX: We see the same pattern. For a wireless device that previously would have been simply named wlan0:

$ ip link show | grep wl

3: wlp3s0: mtu 1500 qdisc mq state UP mode DORMANT qlen 1000

$ lspci | grep Centrino

03:00.0 Network controller: Intel Corporation Centrino Advanced-N 6205 [Taylor Peak] (rev 34)

It is easy to turn off the new scheme and go back to the classic names. We will leave that as a research project. In what follows, we will mostly follow the classic names for definiteness and simplicy.

Question 22

NIC Configuration Files

When using systemd (systemd is getting more standardized), it is preferable to use Network Manager.

On newer Linux distributions these configuration files are either non-existent, empty, or much smaller.

1. Where are the Red Hat NIC config files?
2. Where are the Debian NIC config files?
3. Where are the SUSE NIC config files?

Answer 22

Red Hat NIC config files:

• /etc/sysconfig/network
• /etc/sysconfig/network-scripts/ifcfg-ethX
• /etc/sysconfig/network-scripts/ifcfg-ethX:Y
• /etc/sysconfig/network-scripts/route-ethX

Debian NIC config files:

• /etc/network/interfaces

SUDE NIC config files:

• /etc/sysconfig/network

Question 23

Network Manager

As a system was booted, it consulted the network configuration files in the ___ directory subtree in order to establish the interface properties such as static or dynamic (DCHP) address configuration, whether the device should be started at boot, etc.

Answer 23

/etc

Question 24

Network Manager

What is Network Manager utility good to use to manage a network vs directly modifying the network config files?

Answer 24

Once upon a time, network connections were almost all wired (Ethernet) and did not change unless there was a significant change to the system.

As a system was booted, it consulted the network configuration files in the /etc directory subtree in order to establish the interface properties such as static or dynamic (DCHP) address configuration, whether the device should be started at boot, etc.

If there were multiple network devices, policies had to be established as to what order they would be brought up, which networks they would connect to, what they would be called, etc.

As wireless connections became more common (as well as hotplug network devices such as on USB adapters), configuration became much more complicated, both because of the transient nature of the hardware and that of the specific networks being connected to.

However, modern systems often have dynamic configurations:

• Networks may change as a device is moved from place to place.
• Wireless devices may have a large choice of networks to hook into.
• Devices may change as hardware such as wireless devices, are plugged in or turned on and off.

The previously discussed configuration files were created to deal with more static situations and are very distribution-dependent. A step away from distribution-dependent interfaces and configuration files was a big advance.

While Network Manager still uses configuration files, it is usually best to rely on its various utilities for manipulating and updating them.

Question 25

# Network Manager Interfaces What are 3 different Network manager interfaces that accomplish the same task at different difficulty levels? What the are different situation you would want to use the different tools that accomplish the same thing?

Answer 25

1. if you are using your laptop in a hotel room or a coffee shop, you are probably going to use whatever graphical interface your Linux distribution's desktop offers. You can use this to select between different networks, configure security and passwords, turn devices off and on, etc. 2. If you are making a configuration change on your system that is likely to last for a while, you are likely to use **nmtui** as it has almost no learning curve and will edit the underlying configuration files for you. 3. If you need to run scripts that change the network configuration, you will want to use **nmcli**. Or, if you are a command line junkie, you may want to use this instead of **nmtui**. If the GUI is properly done, you should be able to accomplish any task using any of these three methods. However, we will focus on **nmtui** and **nmcli** because they are essentially **distribution-independent** and hide any differences in underlying configuration files.

Question 26

# Network Manager What is the distro agnostic utility used to manage a network via a command line ui?

Answer 26

**nmtui**

Question 27

# Network Manager What is the distro agnostic utility to manage a network via the command line?

Answer 27

**nmcli**

Question 28

# Network Manager What is the command to see examples of **nmcli** usages?

Answer 28

**man nmcli-examples**

Question 29

# Routing Network **routing** is the process of?

Answer 29

selecting paths in a network along which to send network traffic.

Question 30

# Routing The routing table is?

Answer 30

a list of routes to other networks managed by the system. It defines paths to all networks and hosts, sending remote traffic to routers.

Question 31

# Routing The ___ is a list of routes to other networks managed by the system. It defines paths to all networks and hosts, sending remote traffic to routers.

Answer 31

routing table

Question 32

To see the current routing table, you can use these 2 utilities ___ or \_\_\_. What **options** for the command/utilities would you use to view the current routing table?

Answer 32

Utilities: * route * ip ​Usage: * route -n * ip route

Question 33

# Default Route The default route is?

Answer 33

the way packets are sent when there is no other match in the routing table for reaching the specified network.

Question 34

# Default Route The default route can be setup dynamically via DHCP. But how do you configure it manually (static address setup) with the nmcli utility? Note: This isn't the only way to manually configure the default route. You can modify configuration files directly.

Answer 34

via nmcli: $ **sudo nmcli con mod virbr0 ipv4.routes 192.168.10.0/24 +ipv4.gateway 192.168.122.0** $ **sudo nmcli con up virbr0** via config file mod: On **Red Hat**-based systems, you can modify **/etc/sysconfig/network** putting in the line: **GATEWAY=x.x.x.x** or alternatively in **/etc/sysconfig/network-scripts/ifcfg-ethX** on a device-specific basis in the configuration file for the individual NIC. On **Debian**-based systems, the equivalent is putting: **gateway=x.x.x.x** in **/etc/network/interfaces**. On either system, you can set the default gateway at **runtime** with: $ **sudo route add default gw 192.168.1.10 enp2s0** $ **route** Kernel IP routing table Destination Gateway Genmask Flags Metric Ref Use Iface default 192.168.1.10 0.0.0.0 UG 0 0 0 enp2s0 default 192.168.1.1 0.0.0.0 UG 1024 0 0 enp2s0 172.16.132. 0 0.0.0.0 255.255.255.0 U 0 0 0 vmnet1 192.168.1.0 0.0.0.0 255.255.255.0 U 0 0 0 enp2s0 192.168.113.0 0.0.0.0 255.255.255.0 U 0 0 0 vmnet8 Note that this might wipe out your network connection! You can restore either by resetting the network, or in the above example by doing: $ **sudo route add default gw 192.168.1.1 enp2s0** These changes are not persistent and will not survive a system restart.

Question 35

# Static Routes Static routes are used to?

Answer 35

control packet flow when there is more than one **router** or **route**. They are defined for each interface and can be either persistent or non-persistent. When the system can access more than one router, or perhaps there are multiple interfaces, it is useful to selectively control which packets go to which router. Either the **route** or **ip** command can be used to set a non-persistent route. $ **sudo ip route add 10.5.0.0/16 via 192.168.1.100**

Question 36

# Static Routes What is the command to add a non-persistent route using the **ip** command? route **192.168.1.100** traffic to **10.5.0.0/16** How do you make it a persisten route?

Answer 36

$ **sudo ip route add 10.5.0.0/16 via 192.168.1.100** On a **Red Hat**-based system, a persistent route can be set by editing /etc/sysconfig/network-scripts/route-ethX as shown by: $ **cat /etc/sysconfig/network-scripts/route-eth0** **10.5.0.0/16 via 172.17.9.1** On a **Debian**-based system you need to add lines to **/etc/network/interfaces**, such as: **iface eth1 inet dhcp post-up route add -host 10.1.2.51 eth1 post-up route add -host 10.1.2.52 eth1** On a **SUSE**-based system you need to add to or create a file such as **/etc/sysconfig/network/ifroute-eth0** with lines like: * *# Destination Gateway Netmask Interface [Type] [Options] 192. 168.1.150 192.168.1.1 255.255.255.255 eth0 10. 1.1.150 192.168.233.1.1 eth0 10. 1.1.0/24 192.168.1.1 - eth0** where each field is separated by tabs.

Question 37

# Name Resolution Name resolution is?

Answer 37

the act of translating hostnames to the IP addresses of their hosts. For example, a browser or email client will take training.linuxfoundation.org and resolve the name to the IP address of the server (or servers) that serve training.linuxfoundation.org in order to transmit to and from that location.

Question 38

# Name Resolution There are two facilities for doing this name resolution translation. What are they?

Answer 38

* Static name resolution (using **/etc/hosts**). * Dynamic name resolution (using DNS servers).

Question 39

# Name Resolution There are 3 command line tools that can be used to resolve the IP address of a hostname. What are they and why are there 3?

Answer 39

$ **[dig | host | nslookup] linuxfoundation.org** * dig: generates the most information and has many options * host: more compact * nslookup: older. **dig** is the newest and the **others** are sometimes considered **deprecated**, but the output for **host** is the **easiest** to **read** and contains the basic information. **One sometimes also requires reverse resolution: converting an IP address to a host name.** Try feeding these three utilities a known IP address instead of a hostname, and examine the output.

Question 40

# /etc/hosts **/etc/hosts** holds/contains?

Answer 40

a local database of hostnames and IP addresses.

Question 41

# /etc/hosts /etc/hosts contains a set of records (each taking one line) which map?

Answer 41

IP addresses with corresponding hostnames and aliases.

Question 42

# /etc/hosts A typical /etc/hosts file looks like?

Answer 42

$ **cat /etc/host** *127.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdomain4 ::1 localhost localhost.localdomain localhost6 localhost6.localdomain6 192.168.1.100 hans hans7 hans64 192.168.1.150 bethe bethe7 bethe64 192.168.1.2 hp-printer 192.168.1.10 test32 test64 oldpc* Such static name resolution is primarily used for local, small, isolated networks. It is generally checked before DNS is attempted to resolve an address; however, this priority can be controlled by **/etc/nsswitch.conf** (not often used today).

Question 43

# /etc/hosts **/etc/hosts** usage for static name resolution is primarily used for?

Answer 43

local, small, isolated networks. It is generally checked before **DNS** is attempted to resolve an address; however, this priority can be controlled by **/etc/nsswitch.conf** (**not often used today**).

Question 44

# /etc/hosts student@ubuntu:/etc$ ls -l host\* ## Footnote * -rw-r--r-- 1 root root 92 Oct 22 2015 **host.conf** - rw-r--r-- 1 root root 7 Apr 21 08:46 **hostname** - rw-r--r-- 1 root root 221 Apr 21 08:46 **hosts** - rw-r--r-- 1 root root 411 Apr 20 17:14 **hosts.allow** - rw-r--r-- 1 root root 711 Apr 20 17:14 **hosts.deny*** Explain these other host-related files and their purposes.

Answer 44

The other host-related files in **/etc** are /etc/**hosts.deny** and /etc/hosts.allow. These are self-documenting and their purpose is obvious from their names. **The allow file is searched first and the deny file is only searched if the query is not found there**. **/etc/host.conf** contains general configuration information; it is **rarely used**.

Question 45

# DNS If name resolution cannot be done locally using **/etc/hosts**, then the system will query a ___ server.

Answer 45

**DNS** (Domain Name Server)

Question 46

# DNS DNS is dynamic and consists of a network of servers which a client uses to look up names. The service is \_\_\_; any one DNS server has only information about its ___ of authority; however, all of them together can cooperate to resolve any name

Answer 46

* distributed * zone

Question 47

# DNS The machine's usage of DNS is configured in \_\_\_.

Answer 47

* /etc/**resolv.conf** which historically has looked like: *search example.com aps.org nameserver 192.168.1.1 nameserver 8.8.8.8*

Question 48

# DNS What does the **/etc/resolv.conf** contain?

Answer 48

The machines DNS configuration.

Question 49

# DNS What does the **/etc/resolv.conf** contents look like?

Answer 49

*search example.com aps.org nameserver 192.168.1.1 nameserver 8.8.8.8* which: * **Can specify particular domains to search** * **Defines a strict order of nameservers to query** * **May be manually configured or updated from a service such as DHCP** (Dynamic Host Configuration Protocol). Most modern systems will have an **/etc/hosts.resolv** file generated automatically, such as: Generated by NetworkManager 192.168.1.1 which was generated by NetworkManager invoking DHCP on the primary network interface.

Question 50

# Network Diagnostics A number of basic network utilities are in every system administrator's toolbox name them?

Answer 50

* **ping** * **traceroute** * **mtr** * **dig** Examples: * $ ping -c 10 linuxfoundation.org * $ traceroute linuxfoundation.org * $ mtr linuxfoundation.org

Question 51

# Network Diagnostics What does the **ping** network utility do?

Answer 51

Sends 64-byte test packets to designated network hosts and (if it finds them) tries to report back on the time required to reach it (in milliseconds), any lost packets, and some other parameters. Note that the exact output will vary according to the host being targeted, but you can at least see that the network is working and the host is reachable.

Question 52

# Network Diagnostics What does the **traceroute** network utility do?

Answer 52

This utility is used to display a network path to a destination. It shows the routers packets flow through to get to a host, as well as the time it takes for each hop.

Question 53

# Network Diagnostics What does the **mtr** network utility do?

Answer 53

It combines the functionality of **ping** and **traceroute**, and creates a **continuously** **updated** **display**, like top.

Question 54

# Network Diagnostics What does the **dig** network utility do?

Answer 54

It is useful for testing DNS functionality. Note that one can also use **host** or **nslookup**, **older** programs that also try to return DNS information about a host.