Networking

Chapter 16

In the realm of networking, Linux stands as an unparalleled powerhouse. Its versatility spans across the construction of various networking infrastructures and devices, from firewalls and routers to name servers and NAS (Network Attached Storage) systems.

As the networking domain sprawls expansively, so does the repertoire of commands available for its configuration and management. In this discussion, we'll narrow our focus to key commands, prioritizing the most commonly utilized ones. These commands encompass network monitoring tools, file transfer functionalities, and a comprehensive exploration of the ssh program, instrumental for remote logins.

This chapter aims to delve into the following aspects:

  • ping - Send an ICMP ECHO_REQUEST to network hosts

  • traceroute - Print the route packets trace to a network host

  • netstat - Print network connections, routing tables, interface statistics, masquerade connections, and multicast memberships

  • ftp - Internet file transfer program

  • wget - Non-interactive network downloader

  • ssh - OpenSSH SSH client (remote login program)

We'll proceed under the assumption of some familiarity with networking concepts. In today's digital era, a foundational comprehension of networking is indispensable for anyone utilizing a computer. To maximize the benefits of this chapter, a level of familiarity with the following terms is recommended:

  • IP (Internet Protocol) address

  • Host and domain names

  • URI (Uniform Resource Identifier)

For additional insights into these terms, please refer to the "Further Reading" section below, featuring informative articles that delve into these concepts.

Note

Certain commands we'll discuss might necessitate the installation of extra packages from your distribution's repositories, contingent on your specific distribution. Additionally, executing some commands may call for superuser privileges.

Examining And Monitoring A Network

Even if you don't hold the system administrator role, exploring the network's performance and functionality can often be advantageous.

ping

At its core, the fundamental network command is ping. This command dispatches an ICMP ECHO_REQUEST packet to a designated host, prompting most network devices to respond. This interaction verifies the network connection's functionality.

Note

Most network devices, Linux hosts included, can be configured to disregard these packets. This action is typically taken for security purposes, obscuring a host from potential attackers to some extent. Additionally, it's common for firewalls to be set up to block ICMP traffic.

As an illustration, to test connectivity with linuxcommand.org, we can employ the ping command in this manner:

Once initiated, ping consistently dispatches packets at a set interval (typically one second by default) until halted:

Following interruption (in this scenario, after the sixth packet) by pressing Ctrl-c, ping displays performance statistics. An optimally functioning network should display zero percent packet loss. A successful "ping" indicates that the network components—interface cards, cabling, routing, and gateways—are functioning well overall.

traceroute

The traceroute program (on some systems, the tracepath program may be used instead) generates a list of all the "hops" that network traffic traverses from the local system to a designated host. To visualize the route taken to reach slashdot.org, one would execute the following command:

The output looks like this:

The generated output illustrates that reaching slashdot.org from our test system involves passing through sixteen routers. Routers that disclose identifying information reveal their hostnames, IP addresses, and performance data, presenting three samples of round-trip time between the local system and each router. In instances where routers don't divulge details (due to router settings, network congestion, firewalls, etc.), asterisks appear, as seen in the line representing hop number 2.

netstat

netstat serves as a tool to inspect diverse network configurations and statistics. By leveraging its numerous options, we gain access to a multitude of network setup features. Employing the -ie option allows us to scrutinize the network interfaces within our system:

In the example provided above, our test system showcases two network interfaces. The initial one, named enp0s31f6, functions as the WiFi interface, while the second, known as lo, represents the loopback interface—a virtual interface employed by the system for internal communication.

When conducting basic network diagnostics, it's essential to note the presence of the term "UP" at the outset of the fourth line for each interface, signaling that the network interface is activated. Additionally, verifying a valid IP address in the inet addr field on the second line is crucial. For systems utilizing DHCP (Dynamic Host Configuration Protocol), a valid IP address in this field confirms the DHCP functionality.

Utilizing the -r option enables the display of the kernel's network routing table. This table illustrates the configuration for routing packets between different networks:

In this straightforward example, we observe a standard routing table typical for a client machine situated within a LAN (Local Area Network) and shielded by a firewall/router. The initial line in the listing represents the destination 192.168.1.0. IP addresses ending in zero denote networks, encompassing multiple hosts rather than individual ones, signifying any host within the LAN. The subsequent field, Gateway, identifies the gateway (router) utilized to access the destination network from the current host. An asterisk in this field signifies that no gateway is required.

The concluding line features the destination default, applicable to any traffic directed towards a network not explicitly listed in the table. In our instance, the defined gateway is a router bearing the address 192.168.1.1, presumably responsible for managing the traffic destined for the specified destination.

While netstat offers numerous options, our exploration has covered only a few. Refer to the netstat manual page for a comprehensive list of available functionalities.

Transporting Files Over A Network

What purpose does a network serve if we can't efficiently transfer files across it? Numerous programs specialize in moving data over networks. Presently, we'll delve into two of these programs, with several more to be explored in subsequent sections.

ftp

One of the revered "classic" programs, ftp derives its name from the File Transfer Protocol it operates on. Widely used across the Internet for file downloads, fpt enjoys support from most—if not all—web browsers, often identified by URIs starting with the ftp:// protocol. Before the era of web browsers, the ftp program reigned supreme. It facilitates communication with fpt servers, machines housing files available for upload and download over a network.

The original version of ftp lacks security measures, transmitting account names and passwords in plain text, making them susceptible to interception by anyone monitoring the network. Consequently, the vast majority of Internet-based ftp operations occur through anonymous FTP servers. An anonymous server permits login using the username "anonymous" and a meaningless password.

In the example below, we demonstrate a standard session using the ftp program to download a readme.txt file stored at /pub/example/readme.txt in a demo FTP server named test.rebex.net:

Here is an explanation of the commands entered during this session:

Command
Meaning

ftp test.rebex.net

Invoke the ftp program and have it connect to the FTP server test.rebex.net.

demo

Login name. After the login prompt, a password prompt will appear. Some servers will accept a blank password, others will require a password in the form of an email address. In that case, try something like “[email protected]”.

cd pub

cd example

Change to the directory on the remote system containing the desired file. Note that on most anonymous FTP servers, the files for public downloading are found somewhere under the pub directory.

ls

List the directory on the remote system.

lcd Desktop

Change the directory on the local system to ~/Desktop. In the example, the ftp program was invoked when the working directory was ~. This command changes the working directory to ~/Desktop.

get readme.txt

Tell the remote system to transfer the file readme.txt to the local system. Since the working directory on the local system was changed to ~/Desktop, the file will be downloaded there.

bye

Log off the remote server and end the ftp program session. The commands quit and exit may also be used.

Entering "help" at the "ftp>" prompt reveals a roster of available commands supported by FTP. When using ftp on a server with adequate permissions granted, it becomes feasible to execute numerous standard file management actions. Although slightly cumbersome, it remains a functional method.

lftp – A Better ftp

ftp isn't the sole command-line ftp client available. In reality, there are numerous alternatives. Among the more renowned and widely used options is lftp, developed by Alexander Lukyanov. It operates similarly to the traditional ftp program but offers a multitude of added convenience features. These include support for multiple protocols (including HTTP), automatic retries for failed downloads, background processes, tab completion for path names, and an array of other functionalities.

wget

An additional widely-used command-line tool for downloading files is wget. It proves beneficial for retrieving content from web and ftp sites. It facilitates the download of individual files, multiple files, and even entire websites. For instance, to download the initial page of linuxcommand.org, one could execute the following command:

With its diverse array of options, wget can perform recursive downloads, download files in the background (permitting ongoing downloads even after logging off), and resume downloads of partially completed files. These capabilities are comprehensively detailed in its superior-quality manual page.

Secure Communication With Remote Hosts

For an extensive period, Unix-like operating systems have offered remote administration capabilities over networks. In earlier times, prior to the widespread usage of the Internet, a couple of prevalent programs were employed for remote host logins: rlogin and telnet. However, these programs share a critical vulnerability akin to the ftp program; they transmit all communications, including login credentials, in plain text. Consequently, in the Internet era, they are deemed entirely unsuitable for use.

ssh

To tackle this challenge, a new protocol known as SSH (Secure Shell) was developed. SSH effectively resolves two fundamental issues surrounding secure communication with a remote host. Firstly, it authenticates the legitimacy of the remote host, thwarting potential "man-in-the-middle" attacks. Secondly, it encrypts all communications exchanged between the local and remote hosts.

SSH comprises two components. An SSH server operates on the remote host, actively listening for incoming connections on port 22. Concurrently, an SSH client is employed on the local system to establish communication with the remote server.

Most Linux distributions include an implementation of SSH called OpenSSH, originating from the OpenBSD project. Certain distributions, like Red Hat, package both the client and server by default, whereas others, such as Ubuntu, solely provide the client. To enable a system to accept remote connections, the OpenSSH-server package must be installed, configured, and running. Additionally, if the system is active or protected by a firewall, it must permit incoming network connections on TCP port 22.

Tip

If you lack a remote system for connecting but wish to experiment with these examples, ensure the OpenSSH-server package is installed on your system. Then, employ "localhost" as the name of the remote host. This approach enables your machine to establish network connections internally, interacting with itself.

The SSH client program utilized for connecting to remote SSH servers is appropriately named ssh. To establish a connection with a remote host named remote-sys, the ssh client program is employed in the following manner:

During the initial connection attempt, a message appears, signaling that the authenticity of the remote host is yet to be established. This occurs as the client program encounters the remote host for the first time. To acknowledge and accept the remote host's credentials, simply input "yes" when prompted. Following this, the user is prompted to enter his/her password once the connection is established:

Upon successful entry of the password, the shell prompt from the remote system is displayed:

The remote shell session persists until the user enters the exit command at the remote shell prompt, thereby concluding the remote connection. Subsequently, the local shell session resumes, and the local shell prompt becomes visible once more.

Moreover, connecting to remote systems with an alternative username is feasible. For instance, if the local user me possesses an account named bob on a remote system, user me could access the bob account on the remote system using the following method:

As mentioned earlier, ssh authenticates the remote host's legitimacy. If the remote host fails to authenticate successfully, the following message emerges:

This message stems from one of two potential scenarios. Firstly, it could indicate an attempted "man-in-the-middle" attack, although this occurrence is rare considering ssh's notification system. The more probable cause is a change in the remote system, such as a reinstallation of its operating system or SSH server. However, for the sake of security, the possibility of a security breach should not be dismissed outright. It's advisable to consult the remote system's administrator whenever encountering this message.

Once it's established that the message is due to a benign cause, it becomes safe to rectify the issue on the client side. This can be accomplished by using a text editor (such as vim) to remove the outdated key from the ~/.ssh/known_hosts file. Referencing the example message above, this action appears as follows:

This indicates that the first line within the known_hosts file holds the problematic key. Removing this line from the file will enable the ssh program to accept fresh authentication credentials from the remote system.

Apart from establishing a shell session on a remote system, ssh also enables the execution of a singular command on a remote system. For instance, to execute the free command on a remote host named remote-sys and exhibit the results on the local system:

This technique can be employed in more intriguing ways, as demonstrated in this example where we execute an ls command on the remote system and redirect the output to a file on the local system:

Observe the utilization of single quotes in the aforementioned command. This is deliberate to prevent pathname expansion from occurring on the local machine; instead, it directs the operation to be performed on the remote system. Similarly, if our intention were to redirect the output to a file on the remote machine, we could enclose the redirection operator and filename within the single quotes:

Tunneling With SSH

When establishing an SSH connection with a remote host, an encrypted tunnel is established between the local and remote systems. Ordinarily, this tunnel securely transmits commands typed on the local system to the remote system and returns the results safely. Beyond this fundamental functionality, the SSH protocol permits various types of network traffic to traverse the encrypted tunnel, effectively creating a kind of VPN (Virtual Private Network) linking the local and remote systems.

One prevalent use of this feature is enabling the transmission of X Window system traffic. On a system operating an X server (i.e., a machine displaying a GUI), it's feasible to initiate and operate an X client program (a graphical application) on a remote system and have its graphical display visible on the local system. Executing this is straightforward; consider the following example: Suppose we're situated at a Linux system named linuxmachine, running an X server, and desiring to execute the xload program on a remote system named remote-sys while viewing its graphical output locally. This can be achieved as follows:

Upon executing the xload command on the remote system, its window materializes on the local system. On certain systems, you might need to employ the -Y option rather than the -X option for this functionality.

scp And sftp

The OpenSSH package incorporates two programs leveraging an SSH-encrypted tunnel to transfer files across the network. The first, scp (secure copy), functions akin to the well-known cp program for file copying. Notably, the source or destination pathnames can precede the name of a remote host, followed by a colon character. For instance, to copy a document named document.txt from our home directory on the remote system, remote-sys, to the present working directory on our local system, the procedure is as follows:

Similar to ssh, you can prepend a username to the remote host's name if the account name on the remote host differs from that on the local system:

The other SSH file-copying tool is sftp, serving as a secure substitute for the traditional ftp program. sftp operates similarly to the initial ftp program; however, instead of transmitting data in plaintext, it utilizes an SSH encrypted tunnel. Notably, sftp offers a significant advantage over standard ftp—it doesn't mandate an FTP server to be operational on the remote host. All it necessitates is the presence of an SSH server. Consequently, any remote machine capable of connecting via the SSH client can also serve as an FTP-like server. Here's an example session:

Tip

Numerous graphical file managers in Linux distributions offer support for the SFTP protocol. Utilizing file managers like Nautilus (GNOME) or Konqueror (KDE), users can input a URI starting with sftp:// in the location bar to manage files stored on a remote system running an SSH server.

Summary

Throughout this chapter, we've explored the array of networking tools commonly available on Linux systems. Given Linux's extensive use in servers and networking devices, numerous additional tools can be incorporated by installing extra software. However, even with the foundational set of tools, users can accomplish a multitude of valuable network-related tasks.

PreviousStorage MediaNextSearching For Files

Last updated