T.

me/nettrain
Shell Samurai
Master the Linux Command Line

Stetson Blake
Cincinnati, Ohio

T.me/nettrain
Table of Contents

Preface 1
Introduction and Thank You 8
Acknowledgements 4
Who Am I and Why Am I Worth Listening to? 9
Why You Should Learn Linux 9
What is a Shell Anyway? 10
Why You Should Use the Command-Line 11
Who This Book is For 12
What You Should Know Already 12
What is a Shell Samurai? 14
How this book is structured 7
Syntax Used In This Book 6
Book Changelog 2
Need Help? 3
1. Setting Up a Lab Environment 15
Lab Setup Intro 15
Install Linux on an old Laptop or Desktop computer 15
Use Windows Subsystem for Linux or WSL 16
Use Virtualization Software 17
Dual Boot 17
Use a Server from A Cloud Host 17
Using Cloud Hosting Setup Tutorial 19
2. Linux Fundamentals 31
What is Linux? 31
Linux Distributions 33
What is The Kernel? 34
Kernel Space vs User Space 35
Interrupts 35
What is a System Call? 36
Signals 37

T.me/nettrain
 Everything is a file? Always has been 39
Linux Directory Structure 41
An Overview of Filesystems 42
Devices 43
What is a daemon? 45
Boot Process and Init, systemd 45
What's a TTY? 51
inodes 51
3. White Belt Command Line Skills - The basics, file manipulation, etc 55
Intro to the Shell - pwd, cd and basic ls 55
more ls, mkdir, cat, touch, copy and move 61
Users, Groups, Root and Sudo 68
File Permissions 74
Flying Around the Command Line with Shortcuts 78
file 80
rm — Proceed with extreme caution 80
Helping Yourself with help, man and tldr 83
Installing Packages and Programs 88
Compressing Files and Uncompressing Them 91
The Environment 95
whereis and which 96
Shutting the System Down 97
Aliases - Not just for Criminals, Spies and Detectives 97
4. Black Belt Command Line Skills - Output Redirection, Text and More 99
Black Belt Intro 99
Compiling Code from Source 99
Managing Processes 102
stdout, stdin and stderr 110
Symlinks and Hardlinks 115
Chaining Commands in the Shell with ; and && Operators 117
Text Processing 119
Spying on syscalls with strace 142

T.me/nettrain
 Finding Files with Find 143
Swap Space 145
Scheduling tasks with cron 146
Secure Shell or SSH 149
CURL - Not just for Olympians and Hair Stylists 153
jq - Processing JSON 158
Tmux - Secret Background Windows 161
lsof 165
The /proc directory 169
troubleshooting low disk 171
Logs - Not Just for Lumberjacks 174
Log Rotate - Don’t let the disk fill up! 175
5. The Network 179
DNS - The Yellow Pages of the Internet 188
Transferring files with rsync and scp 193
6. Real World Samurai Skills and Interview Questions 197
Outro 226

T.me/nettrain
Preface

1
T.me/nettrain
 Book Changelog

This book is a living document. Over time, I’ll add sections to it and make it
even better. This section serves as a log of the changes that were made over time.
If you bought a copy, you’ll have access to all future updates.

Version 1.0 - March 17, 2023

Release Day edition

2
T.me/nettrain
 Need Help?

As a purchaser of this e-book, I invite you to join our Discord Server where
you can connect with fellow Shell Samurai alumni. I’ll also hang out there from
time-to-time and try to provide help as I can:

https://discord.gg/vBv8VCS63Y

You can also reach out to me personally at [email protected] and

I’ll try my best to help you. If you have problems with the book or notice
something that doesn’t quite look right, let me know!

3
T.me/nettrain
 Acknowledgements

It seems that no book is complete without gratitude for those that helped to
shape its contents and the author’s life. With that said, I’d like to give some
gratitude of my own, without sounding too sappy. Each person mentioned below
I consider a friend and is someone who helped me achieve my goals or who has
impacted my career in one way or another.
I want to first thank YOU, the reader, for taking a chance on a e-book by a self-
published author who has never put out a massive book like this before. You
took a risk and I’m working my best to make sure it pays off for you! You the real
MVP.
I want to thank Ryan Kulp for inspiring me to ship a product, own my
direction in life and go after my dreams.
I want to thank Reilly Chase for his feedback on all the projects and business
ideas I’ve run by him over the years.
I’d like to thank James Keating for being an excellent manager and developer
of people.
I want to thank Aaron Shultz for giving me confidence in my skills and
abilities.
I want to thank Ryan Chewning for introducing me to VIM many years ago
and making me think about DNS as the Yellow Pages of the internet. Ryan is one
of the first Linux wizard’s I met.
Thank you to Dave Mezera for building an excellent business and local ISP
and taking a chance on hiring me as an engineer.
I want to thank JP Harvey for being a tough boss but forcing me to grow and
take ownership.
I want to thank Brad Kramer for his leadership and operational excellence in
eliminating toilsome processes for engineering teams.
I want to thank Edward Nevard for his sharp mind and detailed eye in
editing and helping grow our Facebook group, This is an IT Support Group.
Thank you to Jordan Andrews for his speedy graphic design work and keen

4
T.me/nettrain
eye.
I want to thank John Williams for his friendship and support.
I want to thank Andrew Baker for introducing me to new software projects
like Node.js before I even wrote code.
I want to thank my parents for pushing me to look into this computer thing as
a career.
I want to thank Chuntel Murawski for supporting me in my goals.

And thank you to all the people I’ve forgotten to mention, because I’m sure
there’s a few!

5
T.me/nettrain
 !

Syntax Used In This Book

Throughout the book, we’ll give “code” samples for you to read through and
try yourself. Code samples will be formatted in Code Span. Code samples will
be formatted in monospace font. Here’s an example:

stetson@linux_rocks:~$ echo "hello, Shell Samurais!"

hello, Shell Samurais!
stetson@linux_rocks:~$

Code samples are formatted to be as readable as possible and use spacing to

separate them from other content. Yes, I know that commands aren’t really code,
just roll with it.
When I specify a command in the book, I’ll also use monospace font to
emphasize that the command is something you should use in your own shell. For
example, commands like echo, ls, man, and others will all be formatted in
monospace.
When I introduce new phrases or concepts or want to emphasize a word, that
word will be italicized.

Hot Tips
Occasionally, you’ll see a Hot Tips section in the middle of a page. This
indicates a tip pertaining to the topic that I’m throwing in as an extra.

Suggestions
If you find that any portion of this book is not formatted well, please reach out
to me and I will fix it! I’ve done my best to make the book as readable and
accessible as possible in PDF and ePub formats.

6
T.me/nettrain
 How this book is structured

This book is structured into chapters which are meant to ease you into Linux
and then quickly ramp up with using Linux and learning commands to
administer systems.

Here’s a brief overview, be sure to also check out the Table of Contents to find
specific sections.

Preface
This chapter! We cover who I am, why you should learn Linux and what skills
you should have already to read this book.

Setting up a Lab Environment

You can’t be a Shell Samurai without a Shell. In this chapter, we get your
system up and running and ready to learn. Whether you choose to dive in and
install Linux on an old laptop, use a web host or use a virtual machine, this step
is essential to our journey. We can’t be a Shell Samurai without having a dojo to
train in, after all!

Linux Fundamentals
This chapter introduces the basics of how Linux operates. We’ll dive into the
kernel, directory structures, how the system boots and more! We recommend
coming back to this chapter later as a reference. This section is mostly theory and
doesn’t focus on too many commands.

White Belt Command Line Skills

Here’s where it gets exciting. We introduce you to the trusty shell and get you
up and running with basic Linux commands. You’ll learn how to navigate
around the filesystem, create files, copy and move them, install packages and
much more!

7
T.me/nettrain
Black Belt Command Line Skills
This chapter builds upon the previous one and goes more in-depth on
advanced commands and concepts you’ll use in your Shell Samurai journey.
You’ll learn about compiling code from source, standard streams, managing
processes, symlinks, regex and more.

The Network
We’ll briefly cover some networking concepts, then dive right into Linux
network commands including capturing packets, working with DNS and
transferring files over the network.

Real World Samurai Skills

This chapter is a compilation of knowledge from Systems Administration
interviews I’ve personally been through along with some real world scenarios
you can learn from. Ready yourself to tackle any interview and operational
problems with Linux.

Outro
This chapter provides some reference materials and links along with
suggestions on where to take your Linux journey next.

Introduction and Thank You

Firstly, thank you for purchasing Shell Samurai! I am honored that you
want to learn something new and that you want to learn it from me. I’ve been
wanting to teach the world about Linux for many years now, but just recently
worked up the courage to write this book! If you’ve ever been in the same
situation, I encourage you to launch your project, write the book, or take the trip
you’re thinking about. It could pay off in ways you might never expect!

8
T.me/nettrain
 Who Am I and Why Am I Worth Listening
to?

My name is Stetson Blake. I’m an Entrepreneur and Software Developer from

Cincinnati, Ohio. I’ve worked in IT for nearly 10 years now -- as a Help Desk
Technician, Web Developer, Network Engineer, Software Developer, Site
Reliability Engineer, Cloud Engineer and DevOps Engineer. I currently work as a
Software Engineer for Planet Labs, who designs, builds and operates a fleet of
satellites that image the earth every day from low-earth orbit. I’ve previously
worked for startups, an ISP, a utilities company and a university in various
technical roles.
I’ve been working with Linux my entire career. I want to teach you about
Linux because I think it’s one of the most important concepts to know as an IT
professional today. Every single industry is impacted by Linux. Most web servers
run Linux. As a Network Engineer, firewalls, routers and switches run Linux. As
an SRE, the production servers I’m on-call for require me to know Linux to
troubleshoot and bring them back up. As an entrepreneur, the servers I maintain
for side projects run Linux and code written in Ruby on Rails.
Especially today, as remote work is ubiquitous and cloud computing is
becoming standard, knowing the foundations of Linux can really help in a huge
way for an aspiring professional!
This book is a compilation of the knowledge I’ve built up over the years from
other books, blog posts, and my own learnings and experiences. Shell Samurai is
the knowledge I wish I could time travel with and give to myself. I can’t time
travel, so I’m doing the next best thing and giving you this knowledge.

Why You Should Learn Linux

Learning how Linux works and how to use it effectively is an essential skill
for anyone interested in an Information Technology career. Linux is the most
widely used Server operating system. The entire world runs on Linux! From cell

9
T.me/nettrain
phones to super-computers, fridges, cars and satellites, Linux is running
everywhere. In 2017, it was estimated that 90% of the public cloud was running
Linux. Every time you use the internet, you’re using Linux.
By mastering Linux, you will gain a deeper understanding of how computers
work in general and how to build career skills that will serve you. Learning
Linux is an investment in your career and your future.
To prove my point, here’s a photo I took of Linux running (or…not running
correctly) on my plane’s entertainment system. Pretty nifty, huh? Also kinda
scary to see ANYTHING not working right on a plane, even if it’s just the thing
that lets me watch movies.

What is a Shell Anyway?

You might be wondering what a Shell is. It’s in the book title, after all. We’re
not talking about seashells here. No, think of the Linux Shell as an outermost

10
T.me/nettrain
layer that wraps the operating system and lets us talk to it. The Linux Shell is a
program that provides a Command-Line Interface (CLI) to perform various tasks
like running programs, configuring the network and adding users. In contrast
with a GUI (Graphical User Interface), a CLI is text only, baby!

Why You Should Use the Command-Line

CLIs are shrouded with mystery and give even some seasoned IT
professionals pause. It can seem so intimidating, just sitting there with its cursor
blinking. All arrogant with its plain-text. Where do you even start? How can you
cram your head with all these commands? “This is too hard, I give up! Why do I
even need this stupid command-line?!” is a phrase we’re throwing away. Stick
with it, the Command-Line is worth learning.
Firstly, using a command-line is much quicker than GUI configuration! Once
you become a Shell Samurai, you can command a Linux system to do your
bidding with ease and speed. The GUI just doesn’t offer the same level of control.
Nearly every boring, repetitive task can be automated away with shell scripts
that run from the command-line. Think of all the saved time you’ll spend on the
beach.
Secondly, learning how to use the CLI will give you a deeper understanding
of how the Linux OS works. By interacting at a lower level, you’ll be able to see
how underlying processes and systems work. The GUI often hides features and
switches to appear more user-friendly, but ultimately handicap our learning.
Knowing how the operating system truly works will help you maintain and
troubleshoot Linux systems more effectively.
Another reason to learn the command-line is because it’s the standard way to
administer servers. I suspect many of you reading this book want to work with
Linux in your full-time job. As a sysadmin, DevOps Engineer, SRE or any other
numerous IT-roles, knowing the shell is essential. It’s the way things are done.
That’s not to say we should do things just because they’ve been done a long time.
However, on matters of maintaining production servers, this one is non-
negotiable. Any “good” Linux sysadmin will have at least a basic understanding
of using the Linux command-line.

11
T.me/nettrain
 Finally, it’s just plain fun! Learning how things work is fun. Typing away at a
shell in a dark room can make you feel like a super-hacker. Unlocking new
abilities and skills is fun too. I imagine you work with computers because you
think in a similar way. There’s much foundational knowledge we’ll begin to
build about Linux, so let’s get started!

Who This Book is For

This book is for anyone who is new to Linux systems administration. Maybe
you’ve tried to learn Linux and the command-line in the past, but just couldn’t
get over some obstacle or it never “clicked” for you. You might have heard how
cool Linux is previously, but you’re not really sure what all the rumors are about.
You could’ve used Linux a few times before but phrases like Vim or Kernel make
you uneasy. My goal with Shell Samurai is to give you the essential tools and
knowledge to be competent enough to understand how Linux operates. In
addition, we’ll learn how to use the Linux command-line.
You’ll learn to configure Linux, connect to other servers, install source code,
standup webservers and so much more. Don’t expect to touch a GUI in this book,
we’re focusing entirely on the Command Line.

What You Should Know Already

This book is aimed at folks who already know a tiny bit about computers. You
might know what a server and motherboard are. You might be working towards
a career in Computer Science or IT or already be working in the field. You might
even not work in IT and have no plans to do so! Perhaps you just have a
tenacious mind. No matter your path, this book aims to explain things simply,
without talking to you like you’re a baby. We’ll go deep enough that things aren’t
hidden “behind the veil”. You don’t need a PhD in Computer Science to read this
book.

12
T.me/nettrain
 !
This book will not cover every single little detail of Linux. You will have to do
some homework on your own. Things will go wrong sometimes. You will have to
Google and make decisions on your own. Commands might not give you the
expected output you thought they might. That’s how the real world is! Although,
if you bought this book, you have access to our Discord server where you can ask
for some help and you CAN feel free to reach out to me and I’ll do my best to
help you! (After you’ve tried Google )
Finally, hats off to you for taking the time and effort to better yourself and
understand something new! That attitude will take you far. I’m clapping for you
right now from my couch.

13
T.me/nettrain
 What is a Shell Samurai?

A Shell Samurai has an intuitive understanding of the CLI and Linux

fundamentals. They can navigate the Command Line and configure the Linux
OS to do their bidding. A Shell Samurai tells computers what to do, not the other
way around.
A Shell Samurai does not know everything and never claims to. They are
resourceful, creative and modest. When an issue pops up that they can’t solve,
they’re able to finding documentation or resources to push through to victory.
A Shell Samurai never stops learning. They’re always improving their skills
and learning more.

Thank you for joining me and taking the path to becoming a Shell Samurai!
Let’s learn Linux and kick ass at it together!

- Stetson Blake

14
T.me/nettrain
 1. Setting Up a Lab Environment

Lab Setup Intro

This book is one that I want you to be able to follow along with, reference and
use to learn Linux fundamentals more deeply. To assist with that goal, I strongly
recommend setting up a “Lab Server” to follow along with the commands we’ll
give. Without a real working Linux system, you’re just kinda reading words and
hallucinating what a system should behave like. You need to use a real shell and
feel a little bit of pain and struggle, but then success and mastery. Real lab
experience like this is your friend.
I also recommend not using the GUI. Yes, it is a little bit harder to only use a
CLI, but it is much quicker and most production Linux systems you touch will
not offer a GUI. If you decide to install Linux on an old computer or laptop that
has a Window manager, you’ll just need to use the terminal program to follow
along with the examples.
Depending on what computers you have access to and use, there’s a few
options for getting a Linux shell at your disposal. We won’t go deep on every one
of these options, but we’ll guide you towards victory. A Linux Samurai has to do
some homework of their own, after all.

Install Linux on an old Laptop or Desktop computer

Installing Linux on an old system is a great option to get started using Linux.
Installing Linux and using it as your daily driver is a great way to dive into the
deep end. Just be cautious as many newbies may install Linux on their “main
computer” and then break things beyond repair. This isn’t a big deal because you
can simply reinstall and the only cost is your time. If you go this route, be sure to
install Linux on a machine that isn’t your primary computer.
If you decide to take this path, use Ubuntu Desktop available at Ubuntu.com.
Grab the latest version, which is currently 22.02.2 LTS. The LTS references
Long Time Support and means that this edition will be supported for a few years.

15
T.me/nettrain
 You’ll need a USB drive to install Ubuntu from a bootable disk or if you’re a
little more old school, you can burn the Ubuntu image to a disc and boot from
that. Ubuntu has setup instructions for installing on a Bootable USB stick here.

Use Windows Subsystem for Linux or WSL

If you’re using Windows 10 or newer, you’re in luck. Windows Subsystem for

Linux (WSL) is a feature of Windows that allows you to run a Linux environment
without running a separate virtual machine.
Windows Subsystem for Linux is probably the easiest way to get a Linux shell
up quickly, but I don’t totally recommend it because it isn’t a “true” Linux
system. There are a couple of drawbacks and performance disadvantages. Some
of the tutorials in this book won’t work 100% correctly if you use WSL and you
might have to google your way through a few error messages. I think it’s more
valuable as a beginner to use one of the other options. That said, if you really,
really want to use WSL…
Here’s how to get a Linux Shell using WSL:
• Enable the Windows Subsystem for Linux feature:
⁃ Open the Start Menu and search for “Turn windows Features on or
off”
⁃ Check the box next to “Windows Subsystem for Linux”
⁃ Click OK and wait for it to install
⁃ Click Restart Now to restart your computer
• Open the Start Menu and search for Microsoft Store and Open It
• Search for Ubuntu
• Click on Ubuntu and make sure you are using the Ubuntu program put
out by Canonical Group Limited
• Click Get
• Wait for it to Install
• Click Open
• WSL will prompt you for a username, go ahead and set one.
• WSL will ask for a password, you can type one and hit enter. Note that the

16
T.me/nettrain
 password will not show up on the terminal! This is called blind typing and
is normal. Enter your password again to confirm.
• Congrats you should now have your very own Ubuntu terminal up and
running!

You can access WSL in a few ways. First, either through the start menu and
searching for Ubuntu. Then, you can pin it to start or the task bar. The other way
you can access your shell is by opening Command Prompt or Powershell and
simply typing ubuntu and hitting enter.

Check out the Microsoft Documentation for more info on Windows Subsystem
for Linux.

Use Virtualization Software

Another great option is to use software like VirtualBox or VMWare to install a

Linux system on your laptop or desktop. This option is flexible but requires a
little bit of configuration and setup. It’s free and lets you customize everything
you need to and run any ISO image and flavor of Linux. The Virtual Machine is
segregated from your main computer so you’re free to mess things up and get it
into a broken state without pain.
For more info on using VirtualBox, reference the installation docs Ubuntu has
provided here.

Dual Boot
Dual Booting is a way of installing Linux alongside another operating system
on the same machine. This is a similar route to just installing Linux on a
computer like I mentioned above. You’ll have to do some partitioning of your
hard drive and do some other configuration. Due to that, it’s an option, but not a
very good one in my opinion due to the complexity involved for a newbie.

Use a Server from A Cloud Host

17
T.me/nettrain
 This option is my favorite. Today, the “Cloud” is ubiquitous and cheap. For
many years, you’ve been able to rent someone else’s computer in the form of a
“VPS” or Virtual Private Server. There are tons of platforms like Digital Ocean,
Vultr and AWS. Be careful with AWS, because you can really run your bill up if
you don’t know what you’re doing!
Point is, any basic Linux server shouldn’t cost more than $5/month. For some
cloud providers, you can even shut down your server when you aren’t using it
and just pay for the amount of time it’s turned on! Billing changes constantly, so
be sure to read your provider’s FAQ and Billing Documentation to be sure.
If you use this option, be sure to check out the section on SSH in this book as
that is the primary way you’ll connect to a cloud server! When you install your
server, be sure to use the latest version of Ubuntu, which is the Linux
Distribution we’ll use in this book.
The path you take to get access to a Linux Shell doesn’t matter too much. All
that matters is that you start and get a basic bash shell up and running.

18
T.me/nettrain
 Using Cloud Hosting Setup Tutorial

Setting up Digital Ocean VPS

Digital Ocean is one of my favorite providers for basic Linux servers. In this
section, we’ll walk through setting up a Ubuntu Linux host with Digital Ocean.
Some of these instructions may change slightly as Digital Ocean updates their
website and UI, but the general process should be the same.
You can sign up by navigating to DigitalOcean.com or using my affiliate link
here. With my affiliate link, you should be able to get some free credit for 60
days. If not, Digital Ocean is currently only $4/month or so for a bare-bones
shell. By the way, I usually set up subscription services with Privacy.com, so I can
make a digital, one-time credit card and limit spending when it comes to renew.
Once you’re signed up and logged into Digital Ocean you can create a new
droplet. A droplet is Digital Ocean’s term for a simple virtual machine running in
the cloud. They offer a ton of other services, but we’ll just use droplets for now.
From the top panel, click Create > Droplets. Or, navigate to the droplets
section along the left-hand pane and click Create Droplet.

19
T.me/nettrain
 20
T.me/nettrain
Creating a Droplet
Next, you’ll be taken to a wizard that allows you to select various settings
related to your virtual machine. For the region, select the one geographically
closest to you so that you have lower latency connecting to it. I chose New York.

Scroll down to Choose an Image and select Ubuntu and pick the latest
(highest version number) that is currently offered. That’s 22.10 x64 as of this
writing. The other options under marketplace, snapshots and backups have
images that others have created with additional software pre-installed. You can
ignore those for now.

21
T.me/nettrain
 Under size, choose the smallest and cheapest plan you can find. That’s
currently the $4/month tier which has 512 mb / 1 CPU, 10 GB disk and 500 GB
transfer. I like Digital Ocean because their pricing is simple! Digital Ocean
currently bills hourly, so technically you could spin your VPS up for an hour,
play with it and delete it when you’re done and pay almost half a cent. Not too
bad.

22
T.me/nettrain
 Next, we’ll setup authentication for your VPS. Scroll down to Choose
Authentication Method and choose SSH Key or Password. My account shows a
few keys that already exist for other accounts. If you’re using Password
authentication, just keep in mind that this is a less secure option. Probably fine
for your Linux lab, but certainly a no-no in production environments.

23
T.me/nettrain
 Click New SSH Key and you’ll see a window pop up asking for a Public SSH
Key along with some instructions on how to create one. We cover SSH more in-
depth in Black Belt Command Line Skills here, but we can go ahead and make a
public/private key-pair now and introduce the basics.
If you’re on Windows, you can open Command Prompt or Powershell (Start
button > search for Command Prompt or Powershell). If you’re on a Mac, open
the terminal from Applications and use that. Once you have a prompt open, type
ssh-keygen and walk through the instructions to create a new key. You’ll be
asked for an optional passphrase (you can skip this by just hitting enter) and a
location to save your key-pairs.
Once you’re done, you can use the cat command in Powershell or Mac
Terminal and type command in Command Prompt on the path of the PUBLIC
key file, likely named id_rsa.pub. Copy the output from that command into
your Public SSH Key prompt.
If this is confusing right now, read up in the SSH section more, check out the
Digital Ocean documents or as a worst-case scenario, use Password based
authentication. The tl;dr is that we just made a public and private key for
authenticating to your server. This is a cryptographically secure way to connect
to a server through Public Key Cryptography or asymmetric cryptography.
Finally, give the key a memorable name and click Add SSH Key.

24
T.me/nettrain
 We’re almost there! In the finalize details section, give your VPS a hostname
(no underscores allowed!) and any optional tags. Tags could be things like
“database” or “myWebApp”. We don’t need them for now, but you can add
some if you’d like. There should also be a default project where you can place
your droplet for organization purposes.
Once you’re satisfied, go ahead and hit Create Droplet!

25
T.me/nettrain
Connecting To Your VPS
Digital Ocean will provision your VPS and it usually doesn’t take long. Like,
under a minute. You can monitor the status from your Digital Ocean dashboard:

If you click your VPS’ name you’ll be taken to another dashboard that has all
kinds of data on the health of your instance and other details. Your VPS should
be up once you see an ipv4 address assigned. Go ahead and click the ipv4
address to copy it to your clipboard. Mine is 137.184.198.176 in this screenshot:

26
T.me/nettrain
 Back in your Command Prompt, Powershell or Mac terminal, we can connect
to our VPS with the ssh command.
The syntax and format you’ll use is ssh root@<IP_Address>. For my example,
I would perform ssh [email protected]
We use the root user to connect to our server. Later, we can set it up so that we
can only connect as the Ubuntu user, which is more secure.
You’ll then get a message that seems scary if you’ve never seen it before:

➜ ~ ssh [email protected]
The authenticity of host '137.184.198.176 (137.184.198.176)' can't
be established.
ED25519 key fingerprint is SHA256:JPW506TBNx9PIZoxtug+AVhsO/
eHG+0Cr/UfP0UheoA.
This key is not known by any other names
Are you sure you want to continue connecting (yes/no/
[fingerprint])?

Go ahead and type yes and hit enter. Our computer is telling us that we’ve
never connected to this VPS or machine before and it can’t verify it knows who it
is. It also gives us a fingerprint that identifies this server uniquely. This can be
used to verify the server is who it says it is.

27
T.me/nettrain
 If all has gone well, you’ll be dumped into your new Ubuntu system and see
some system statistics and various links to support:

Troubleshooting Key Errors

You may get a message when trying to connect that looks like this:

➜ ~ ssh [email protected]
[email protected]: Permission denied (publickey).

28
T.me/nettrain
 This typically indicates that you are connecting as the wrong user (I’m not,
I’m using root here) or that your id_rsa private key has the wrong permissions or
you may not be using the correct key! You can get some more info on the error by
providing -vvv flags to the ssh command to indicate you’d like verbose output:
ssh [email protected] -vvv
This will generate a ton of output that you can skim through and maybe find
something obvious that’s gone wrong.

Other Methods of Connecting to your VPS

You don’t only have to use the command-line to connect to your terminal. You
can also use an SSH client like PuTTY to connect to your server, which allows
you to configure and save your server settings so you don’t have to remember
them every time. PuTTy will save your server name and IP address, but you’ll
have to configure it to point at your private key, if you didn’t chose password
authentication:

29
T.me/nettrain
 Congratulations! You’ve spun up your first Ubuntu server on Digital Ocean!
You now have a light-weight, instantly accessible Linux server that you can
connect to from anywhere in the world. You could run a web server on it, start a
side project or in our case…Learn Linux more deeply and become a Shell
Samurai!

30
T.me/nettrain
 2. Linux Fundamentals

What is Linux?
The word Linux is a bit of a tricky to fully explain. Linux actually refers to the
Linux Kernel, which is the core brains of Linux operating systems. We’ll get into
the kernel more in later sections. Mostly, you’ll hear people say Linux and they
broadly mean a Linux operating system, like Ubuntu.
Linux was first released in September, 1971 by Linus Torvalds from Finland as
a free and open-source alternative to commercial operating systems. Since then,
it’s grown to massive popularity with over 90% of the world’s super computers
running Linux. Linus, of course, was standing on the shoulders of giants. Prior to
Linux’s development, Ken Thompson and Dennis Ritchie developed the Unix
operating system at Bell Labs along with tons of other utilities which would
make it into Linux. Linus also took inspiration from and was initially developed
on an Operating System called Minix. Minix was developed by a professor,
Andrew Tanenbaum, for teaching his students how operating systems work.

What is Open-Source?
In the context of Linux, open source refers to the concept of making the
underlying source code of the operating system freely available to anyone to use,
modify and distribute. This has some awesome implications. One being that the
code can be modified and tuned to specific use-cases. Another being that it can be
audited for security. In addition, anyone can read Linux’s source-code. This
means that Linux is developed quickly and for a wide-range of use-cases.
Linux’s open source philosophy has indeed led to a massive community of
thousands of developers working on improving Linux, fixing bugs and
developing new features. Due to this, Linux has become a stable, reliable and
secure operating system running all over the world.

Unix vs Linux
It’s important to call out that Unix and Linux are not the same. This can be
confusing for beginners. Unix was developed in at Bell Labs in the 60’s, while

31
T.me/nettrain
Linus Torvalds developed Linux in the 90’s, 20-something years later. Much of
Linux’s design was inspired by Unix. Linux is even called Unix-like. Linus picked
up parts of Unix that he thought were helpful and dropped other pieces. Unix
systems are still developed today and running many production systems.

32
T.me/nettrain
 Linux Distributions

Since Linux is open source, lots of different developers have worked on it and
forked their work into separate distributions or “distros” based on the opinions
they have on how the operating system should work. A fork is what happens
when software is cloned from other software and takes a different direction.
Linux distributions each have their own set of unique features, tools and pre-
installed applications.
It’s important to keep in mind that Linux actually refers to the Linux kernel,
while distributions using the kernel are known as Linux operating systems.
For this book, all of our examples will use Debian or Ubuntu tooling. Let’s
briefly cover some of the most popular Linux distributions.
Debian
Debian is one of the oldest Distributions based on the Linux kernel. It was
first released in 1993 and has a lot of momentum to this day! Debian has different
“branches”, which are just different versions of the same operating system. Stable,
Testing and Unstable are the three main Debian branches. Stable is the branch
that most people should use, while testing and unstable are more frequently
updated with changes that eventually make it over to Stable. Debian is the basis
for a lot of other distributions, with the most popular being Ubuntu. Debian, like
many distributions, utilizes a lot of free and open-source software.

Ubuntu
Ubuntu was initially released in 2004 and is developed by a British company
called Canonical. It is very popular among beginners and professionals alike. It’s
the distribution we’ll use in this book’s examples. Ubuntu is based on Debian
and even uses Debian’s package management system for installing software.
Ubuntu has many versions of its OS for desktops, servers and even a version
used in Internet of Things devices and robots called core.
Ubuntu was my first introduction to Linux and I think you’ll enjoy using it
too.

33
T.me/nettrain
RHEL
RHEL, or Red Hat Enterprise Linux is developed by none other than…a
company called Red Hat. Red Hat, as of this writing, is owned by IBM. Red Hat
Enterprise Linux has more restrictions on its distribution and is more commercial
than other distributions. As well, RHEL uses a different packaging system for
installing software than Debian and other distributions called RPM or Red Hat
Package Manager.

Fedora
Fedora is in some ways the pre-cursor and basis for Red Hat Enterprise Linux.
Red Hat is a fork of Fedora. Fedora shares many similarities with Red Hat, but
comes with a lot fewer restrictions and commercial agreements. Fedora also uses
RPM or Red Hat Package Manager since it’s based on Red Hat.

Arch
Arch is worth a mention in this book, though it isn’t commonly used on
servers. Arch is usually installed on Desktops and Laptops for users that want
deep control over their system. Arch is very opinionated about many of its design
decisions. This just means that the developers have strong convictions about the
way to do things. I don’t recommend Arch for beginners, but learning it could
give you a deeper understanding of the Linux operating system if you’re up for
the challenge.

What is The Kernel?

The kernel is the most important part of the Linux operating system. It is the
brains and core of the operating system. Linux’s kernel is responsible for starting
the OS up, managing processes, memory, devices, processes and so much more.
For now, just understand that the Linux Kernel is a crucial piece of the operating
system.
If you’d like to see which kernel version is running on your system, you can
run uname -r:

34
T.me/nettrain
 "
stetson@linux_rocks:~$ uname -r
5.15.0-56-generic

There are many kernel versions and because the Linux kernel is open-source,
we can inspect the kernel source code, modify it and run it on our own systems.
We won’t dive that deep in this course, but know that it’s something you could
do. For more info on the Linux kernel, check out Kernel.org

Kernel Space vs User Space

The Linux system is abstracted into 2 distinct layers: User Mode and Kernel
Mode. These are sometimes also called User Space and Kernel Space.
When we refer to user space, we’re talking about parts of the operating
system dedicated to running user-level programs and applications. This area is
separate from kernel-space, which handles running system-level operations that
interact directly with the hardware. This separation of concerns keeps Linux
modular and secure. If your machine were infected with malware, you wouldn’t
want it to have all of the permissions that Kernel mode has. Direct access to
hardware, memory and CPU would be a nightmare if it fell into the wrong
hands!
User Space and Kernel Space are also sometimes called privilege levels or
protection rings. There are 4 “rings” where the Kernel, Device Drivers occupy
rings 0, 1 and 2. Applications running in User Space run in the least privileged,
outter-most layer, ring 3. These keeps Linux secure so that programs can’t
directly interact with the innerworkings of the system.
However, we know that some programs definitely interact with our hardware
and inner workings of the system. So how do they do it? They use a system call!
We’ll cover system calls shortly.
Hot Tip : You’ll sometimes also hear user space referred to as userland.

Interrupts
An interrupt is a signal sent by hardware or software to the kernel indicating

35
T.me/nettrain
that immediate attention is needed. Interrupts, like in real life, call attention to
something and STOP the current execution of whatever else is going on.
A simple example of a hardware interrupt is input from a device like your
keyboard or mouse. Any time you type a key, the kernel suspends operation,
saves its state and executes an interrupt handler or code for the device driver
associated with your keyboard. Once the signal is handled, typically by
outputting a character to your screen, control is returned back to the kernel.

What is a System Call?

System calls are a type of software interrupt.
We mentioned earlier that user space programs are able to use system calls to
elevate their privileges or access system resources. For example, if a program
needs to access your webcam, it needs to go through the Kernel first. To do so, it
must use a system call to ask the kernel to do its bidding.
A system call is an interface between software and the Linux kernel. If you
open a file, the open syscall is invoked, then the read and write syscalls. If you’re
running a web server, the connect, send and recv system calls are used for
connecting to clients, sending data and receiving data. System calls are
constantly used by programs to connect to other servers, backup files, write logs,
interact with the network and much more. Programs can’t do these tasks on their
own, so they use system calls to ask the kernel to do these processes for them.

How does a system call work?

When a system call is made by a program, the program asks the kernel to
perform that task. The process pauses execution and passes the task to the kernel.
In kernel mode, the system call is performed. Once the call is finished, the kernel
passes control back to the user level process that made the system call.

A few more common syscalls are:

• fork() - Duplicates an existing process and creates a new one

• exec() - Replace the current process with a new one
• read() - reads data from a file descriptor

36
T.me/nettrain
 • write() - writes data to a file descriptor

If system calls don’t quite make sense right now, don’t worry. Just know that
they are a process that happens behind the scenes to help programs do their job.
This is a pretty advanced concept for this early in the book, but I feel I’d be doing
you a disservice if I didn’t introduce system calls!

Signals
Signals are a king of software interrupt. A signal is a way that processes can
communicate with each other (sometimes called interprocess-communication)
and the operating system. We can even send our own signals to processes from
the command-line! A signals might be sent for a lots of different reasons, like the
completion of a piece of code or asking a process to shutdown and terminate
itself.

How do Signals Work?

Signals are first delivered to processes and the process decides what to do
with it. The process has a few different ways it could deal with the signal. It
could ignore it. It could “catch” the signal with a handler or function and then
perform some other code. The process could also be forced to terminate without
it ever being able to make a decision on how to handle the signal, in the case of a
SIGKILL signal.
Signals are represented by a number along with an all-caps name. Below are a
few of the most common Signal types:

• SIGHUP or 1: Hangup - Used to notify a process that the controlling

terminal has disconnected.
• SIGINT or 2: Interrupt — Occurs when a user presses Control+C together or
is called from another process or event
• SIGKILL or KILL or 9: Used to immediately terminate the process with no
chance to cleanup and release resources, like memory and cpu.
• SIGSEGV or 11: Segmentation Fault - An Error caused by a program

37
T.me/nettrain
 "
accessing memory it isn’t allowed to.
• SIGTERM or 15: Software Termination. Gracefully terminate a process and
give it a chance to cleanup.
• SIGSTOP: Suspend execution until a SIGCONT signal is received
• SIGCONT: Tells a process to continue execution

These are just a few examples of Signals and their names, Linux currently
implements about 30 total.
I mentioned that we can also send our own signals to processes. We do this
with the kill command. By default, you can use the kill command and pass it a
process ID to send a SIGKILL signal. This is a very useful command for
“runaway processes” or processes that you’ve tried to shut down and won’t stop
when we’ve asked nicely.
If you’d like to send a signal to a process of a certain type, you can specify it
in the kill command as it’s Signal number. Eg:

$ kill -9 <process_id>

The above command sends the SIGKILL signal, so the command is the same
as issuing kill <process_id>. You could instead specify signal ids 1, 2, 11 or
any of the others as a flag.

Hot Tip : If you ever want some process that’s running on your terminal to
immediately stop in Linux, you can hit Control+C on your keyboard to send a
SIGINT or Signal Interrupt. This requests that the current process stop. If you
have a script that is taking too long to run or you entered a command you didn’t
mean to run, try hitting Control+C to stop execution and return control to the
shell.

38
T.me/nettrain
 Everything is a file? Always has been

A phrase you’ll hear a lot about Linux is that “Everything is a File”. What?
How could that be? Is my CPU a file? Yes. My Keyboard? Yes, also. The
Monitor?! Yes! While the components of your system aren’t literally a file, that’s
how Linux sees them and interacts with them. This is a design philosophy of
Linux. Almost every single aspect of the system, including devices like monitors
and keyboards, processes and more are represented as a file.
For example: A printer is represented as a file and printing can be performed
by writing to that file. Similarly, processes are represented by files in the /proc
directory on the file system. Detailed information about a process can be
obtained by reading the contents of that file.
This abstraction of system resources as files makes Linux highly flexible and
powerful. It enables the system and kernel to treat a wide range of resources in a
consistent way. It also makes it easier for users and developers like us to interact
with these resources, as they can use familiar file operations, such as reading,
writing, and copying, to interact with them.
In short, the "Everything is a file" philosophy in Linux provides a consistent
way of interacting with and manipulating system resources, making the system
both user-friendly and highly flexible.

39
T.me/nettrain
 40
T.me/nettrain
 Linux Directory Structure

Linux Directory Structure

Every Linux install has a directory structure with folders and files that help
the system run properly. If you have shell access, you can list these directories by
issuing ls -l /

Most every Linux system has the following directories under its root
filesystem:

• / - Represents the top-level or ROOT index of the filesystem. This is

unrelated to the root user.
• /bin - Contains binaries and files like cat, ls, etc.
• /boot - Files related to the boot process, like the kernel and bootloader
• /dev - Device files that represent any devices that may be connected to the
system
• /etc - Contains configuration for software and daemons running on the
system
• /home - Contains individual user’s home directories
• /lib - Library files that are used by other programs or binaries
• /media - Mount point for detachable media like USB drives
• /mnt - Mount point for temporary filesystems
• /opt - Optional software packages
• /proc - Information about running processes
• /root - The root user’s home directory, separate from the /home directory
• /run - Contains info about the system since it was started
• /sbin - Essential system binaries used for systems administration
• /srv - Stores data and services provided by the system.
• /tmp - Temporary files created by users or programs. Cleared on reboot!

41
T.me/nettrain
 • /usr - used for user installed programs and utilities.
• /var - Variable data. Usually logs are output somewhere in this directory

We’ll go deeper into what’s kept in some of these directories later, for now just
keep the general structure in mind.
An Overview of Filesystems

As we saw in the previous section, a filesystem is a method used by the

operating system to store and organize data like files and directories on disk in a
hierarchal structure. The filesystem provides a standardized way for applications
and users to interact with stored data.

Below are some of the most common types of filesystems used today:

• ext4 - The most recent version of native Linux filesystems. This filesystem
succeeds the ex2 and ext3 versions. Supports disk volumes up to 1
EXABYTEand file sizes up to 16 terabytes.
• btrfs or B-tree filesystem - Sometimes called Butter FS or Better FS. New
filesystem type that is looking to expand on features from ext4.
• XFS - A high-performance journaling filesystem. Typically used on systems
like media-servers which have large files.
• NTFS and FAT - Windows filesystems
• HFS+ and APFS - Apple standard used on most Mac systems

You can also use df -T to see info about storage on your system and what
types of filesystems are in use:

root@linux_rocks:/tmp# df -T
Filesystem Type 1K-blocks Used Available Use% Mounted on
overlay overlay 61255492 3491404 54622764 7% /
tmpfs tmpfs 65536 0 65536 0% /dev
tmpfs tmpfs 1018268 0 1018268 0% /sys/fs/
cgroup
shm tmpfs 65536 0 65536 0% /dev/shm
/dev/vda1 ext4 61255492 3491404 54622764 7% /etc/hosts

42
T.me/nettrain
 tmpfs tmpfs 1018268 0 1018268 0% /sys/
firmware

• Filesystem: This column displays the name of the filesystem.

• Type: This column displays the type of filesystem.
• 1K-blocks: This column displays the total size of the filesystem in 1-
kilobyte blocks.
• Used: This column displays the amount of space used on the filesystem in
1-kilobyte blocks.
• Available: This column displays the amount of free space on the
filesystem in 1-kilobyte blocks.
• Use%: This column displays the percentage of the filesystem that is
currently being used.
• Mounted on: This column displays the directory where the filesystem is
mounted.

Keeping an eye on storage space in your Linux shell is essential! You should
also be familiar with the disks and filesystem types they’re using.

Devices
Devices like keyboards, mice and disks all need drivers to function on your
system correctly. Since everything is a file on Linux, we can interact with devices
through device files. Device files are stored in the /dev directory on your system.
Go ahead and perform an ls -l command on the /dev directory on your system.
(This command simply lists in long format -l every file and directory in /dev):

stetson@linux_rocks:/$ ls -l /dev

You’ll get back a ton of devices on each line. Some of them represent disks,
others represent virtual terminals. One file is called /dev/null which is a special
type of device on our system that takes input and throws it all away. You can
echo some text to /dev/null now and then cat the file and observe that it

43
T.me/nettrain
contains nothing:

stetson@linux_rocks:/$ echo "Hello Shell Samurais" > /dev/null

stetson@linux_rocks:/$ cat /dev/null
stetson@linux_rocks:/$

The Linux Kernel knows that /dev/null should return nothing since it acts as
a kind of “black hole” for anything we throw into it. Above, we echo a simple
string of text and use the > redirect operator to send our output to /dev/null. We
cover command redirection more in Black Belt Command Line Skills.
Looking back to our ls -l /dev output, we’ll see a few columns of data. The
first character on each line has a special meaning:

stetson@linux_rocks:/$ ls -l /dev
crw--w---- 1 root tty 4, 0 Dec 30 00:38 tty0
brw-rw---- 1 root disk 252, 1 Dec 30 00:38 vda1
prw-r--r-- 1 root root 0 Dec 30 20:13 fdata
srw-rw-rw- 1 root root 0 Dec 30 20:13 log

See each letter to the far left of the column? The c, b, p and s?

These stand for:

• c - character
• b - block
• p - pipe
• s - socket

Character devices transfer data one character at a time and work with data
streams. /dev/null is a character device. Printers are also represented by a
character device.
Block Devices are accessed by programs in fixed chunks. Hard drives and file
systems are both a type of block device.
Pipe Devices or named pipes are similar to character devices. They allow two or

44
T.me/nettrain
more processes to communicate.
Socket Devices also allow communication between processes like pipe devices
do. They typically allow communication with many processes or programs at
once. These files are also often found outside of the /dev directory we’ve been
looking at.

Device Naming
If you explore the device names in /dev, you’ll see some funny looking
names. Some common device names for storage are /dev/sda, /dev/sdb and /
dev/sdba2. These names mean SCSI device a, SCSI device b and SCSI device a
partition 2. These names come from the SCSI (pronounced like “scuzzy“)
protocol or Small Computer System Interface. Although SCSI interfaces are
largely not used anymore, this naming scheme still sticks around.

What is a daemon?
A daemon (pronounced DAY-MAN, DAY-MON or sometimes DEE-MON) is a
background process that runs on the Linux OS continuously. When we say
background, we mean that the program runs without any user interaction. It
runs “behind the scenes“ so to speak.
An example of a Daemon would be processes like Apache (web server),
Nginx (also a web server), PostgreSQL (a database server) and SSHD (a secure
shell daemon). If you’re a business or individual who maintains a web service of
any kind, it’s likely you want it to stay running all the time, and without much
drama or management. You want it to start when your server starts and keep
running forever. This process would act as a daemon.

Boot Process and Init, systemd

45
T.me/nettrain
How does Linux Boot?
Every operating system has a boot process. Linux is no different. In this
section, we’ll go over the basics of the boot process and explain each component.
There’s a ton of processes happening between powering up a Linux system
and when it finally gets to a blank prompt. The boot process can be explained
very simply, then we’ll go deeper.

Here’s the summary:

• A BIOS or UEFI boot firmware runs a bootloader

• The bootloader finds the kernel image from the disk, loads it into memory
or RAM and starts it up with kernel parameters
• The kernel initializes hardware devices and drivers needed to talk to its
devices and mounts the root filesystem.
• The kernel next starts a process called init and gives it a process ID of 1. On
newer systems, this is now replaced by systemd
• Init brings up the rest of the system processes needed like networking
services, web daemons and logging services.
• Init finally allows you to login near the end of the boot sequence

Linux Boot Process - BIOS and UEFI

In order to understand the Linux boot process further, we should familiarize
ourselves with BIOS and UEFI and the differences between the two systems.

BIOS
BIOS stands for Basic Input/Output System. BIOS is a firmware used in the
boot procedure to find the bootloader among some other secondary system
functions like changing boot disk order and network settings. BIOS firmware is
stored on an EPROM chip, which allows for firmware updates from the
manufacturer.
Once BIOS boots up, its goal is to search for the Master Boot Record (MBR)
and execute it. The MBR contains the primary boot-loader in the first 512-bytes of

46
T.me/nettrain
the disk.

UEFI
UEFI or Unified Extensible Firmware Interface is newer and has the same goal
of starting the system as BIOS does. However, UEFI stores the initialization data
in a .efi file, rather than storing it on the firmware. This .efi file is stored on the
disk of the system on a partition called the EFI System Partition, alongside the
bootloader. The GUID Partition Table (GPT) format is used in UEFI systems.
UEFI was developed to improve on some of the shortcomings of BIOS, such
as:

• Faster Boot Time

• Larger Disk Size support: BIOS supports 2.2 TB, while UEFI supports 9
zettabytes (huge! As of today…)
• Secure Boot - UEFI can utilize the Secure Boot feature to require that boot
loaders are digitally signed, preventing malware from being loaded on the
system.

Today, UEFI is the standard and BIOS has become archaic and dated.

The Bootloader
Once BIOS or UEFI has found the bootloader and executed it, the bootloader
takes over. The bootloader’s main goal is to start the Linux kernel. It does this by
finding the kernel, loading it into memory and passing in any kernel parameters
that may be needed.
During this process, the kernel has still not been loaded, so accessing the disk
must be done via slower device drivers than the one the system uses once it’s
booted.
The most common and ubiquitous Bootloader is GNU GRUB or GRand
Unified Bootloader. There are also other Linux bootloaders like LILO (Linux
Loader), BURG, Syslinux and core boot. These bootloaders were developed to be
faster, simpler or have different features than GRUB.

47
T.me/nettrain
Init and User Space Start
After the bootloader loads the kernel into memory, it starts its first user-space
process, init. This kicks off the rest of the process to bring the system completely
up by starting essential services. The init process will always be assigned a
process id of 1. Once init starts, its goal is to start up the rest of the programs that
are necessary to run the system.
Systemd is the newest and current standard for init on modern Linux
distributions. System V init (SysV) is found on older Red Hat Enterprise Linux
distributions. Upstart is the init that was used on older Ubuntu distributions (pre
15.04). There are a few other init systems used for embedded applications,
cellphones and other devices. The three I mentioned are the ones you should be
most familiar with: Systemd, SysV and Upstart. We’ll be working with and
covering systemd.

systemd
systemd works with targets which can be thought of as goals for systemd to
reach. Targets are a type of unit used to group other units and define the system
state that should be reached. A unit is a file that has instructions for how to start a
daemon or system service and the dependencies for starting it. The dependencies
for a service are also units.
For example, consider a Web Server like Nginx. It doesn’t make sense to start
up a web server if the system doesn’t already have its network online and
functional. A Web Server or Nginx unit file will have a dependency on the network
unit file.

Let’s take a look at how systemd boots the system:

• Systemd loads configuration files or units, which are typically located in /

etc/systemd/system or /usr/lib/systemd/system
• Based on those configuration files, systemd determines its boot goal or
target, this is usually default.target. systemd then figures out
dependencies of the boot target and activates those.
• units are configuration files used to describe a resource managed by

48
T.me/nettrain
 systemd. Units might be services, sockets, devices and more. Systemd
units are started in parallel based on the dependencies they define.
• Services are the most common unit, these define daemons or background
processes running on the system. Systemd starts these up as well if they
are a dependency of the unit files

An Example Unit File

Let’s take a look at an example systemd unit file for the popular web server,
Nginx. This file would typically be located at /lib/systemd/system/
nginx.service

[Unit]
Description=The NGINX HTTP and reverse proxy server
After=syslog.target network-online.target remote-fs.target nss-
lookup.target
Wants=network-online.target

[Service]
Type=forking
PIDFile=/run/nginx.pid
ExecStartPre=/usr/sbin/nginx -t
ExecStart=/usr/sbin/nginx
ExecReload=/usr/sbin/nginx -s reload
ExecStop=/bin/kill -s QUIT $MAINPID
PrivateTmp=true

[Install]
WantedBy=multi-user.targetManaging systemd Services

This config file looks very similar to .INI file formats you might have seen in
the past. The sections are denoted by square brackets for Unit, Service and
Install. Each of the keys and values underneath the sections are known as
directives.
The Unit section contains a description of the service along with an After and
Wants directive. The After directive specifies units that should be started before

49
T.me/nettrain
this unit file. The Wants directive is similar to a directive called Requires, but less
strict. It tells systemd to attempt to start any units listed here when this unit file is
activated.
Under the Service section there are instructions for how systemd should start
the daemon. The nginx binary lives at /usr/sbin/nginx. The ExecStartPre,
ExecStart, ExecReload and ExecStop directives tell systemd what program it
should run and with what options before the unit starts, when the unit starts,
when it reloads and finally when the unit is stopped.
The last section is Install. It’s an optional section. The WantedBy directive
sets up a dependency to specify target units that should be started when the
current unit is enabled.
This is just the beginning of Unit Files. Hopefully, this gives you a real-world
example of a Unit file you might find running in production. Next, we’ll look at
interacting with systemd unit files via systemctl.

Systemctl
Systemctl is a command-line tool used to control systemd. We most often use
it to start, stop, restart, enable or disable a service. Starting and stopping are
pretty clear, but what about restarting and enabling or disabling?
Restarting a service stops it and starts it again. You might restart a service
when it isn’t working properly or when you’ve changed a configuration file that
you want to be applied.
Enabling or disabling a service will mark the service to start at boot or to not
start at boot. Pretty simple.

Let’s look at a few commands:

systemctl list-units - List all active units

systemctl status <service> - View the status of a unit
systemctl start <service> - Start the service
systemctl start <service> - Stop the service
systemctl restart <service> - Restart the service

50
T.me/nettrain
 systemctl enable <service> - Enable a service to start on boot
systemctl disable <service> - Disable a service from starting on boot

Systemd and the systemctl command are essential pieces of the Linux
operating system today. Understand them well and you will thrive in your Linux
journey!

What's a TTY?
You’ll soon run into the term TTY on your Linux journey, which stands for
TeleTYpwriter. It refers to a physical or virtual terminal you use to interact with
Linux. Originally, TTYs were physical terminals with a keyboard and screen that
sent your commands to a remote server.
On modern systems, TTYs are usually virtual terminals that you access
through the command line. Each TTY gets an associated device and file located in
the /dev directory. An example device might be named /dev/tty1.
Any time you login to Linux via command-line, the system will create a login
shell process. Multiple TTYs mean that multiple users can be logged in!
You’ll also see pseudo-TTYs (or PTYs) used for remote logins and terminal
emulators. PTYs provide a virtual terminal that appears to be connected to a
physical TTY device.

inodes
Behind the scenes, Linux uses a concept known as inodes or index nodes to
store metadata about files and directories. In fact, every single file and directory
on the system is represented by an inode. Inodes contain info on the file’s
permissions, owner, size, timestamps and location on the disk. When your
filesystem is created, space for inodes is also allocated. Inodes are managed by a
database known as the inode table.
Inodes are assigned unique numbers when they are created. If a file is deleted,
the inode id gets recycled and used again. You can view inode numbers by
running ls -li in any directory.

51
T.me/nettrain
 stetson@linux_rocks:~$ ls -li
total 16
397530 -rw-rw-r-- 1 stetson stetson 10240 Feb 27 00:28
important_file
397528 -rw-rw-r-- 1 stetson stetson 0 Feb 27 00:27
moon_landing_photos.png
397522 -rw-rw-r-- 1 stetson stetson 209 Feb 27 00:44
myArchive.tar.gz
397525 -rw-rw-r-- 1 stetson stetson 0 Feb 27 00:27
secret_documents

Above, the numbers below the first column are inode ids.

If you want to see detailed information about a file and its inode, you can use
the stat command to do so:

stetson@linux_rocks:~$ stat important_file

File: important_file
Size: 10240 Blocks: 24 IO Block: 4096 regular file
Device: a2h/162d Inode: 397530 Links: 1
Access: (0664/-rw-rw-r--) Uid: ( 1000/ stetson) Gid: ( 1000/
stetson)
Access: 2023-02-27 00:44:53.688165013 +0000
Modify: 2023-02-27 00:28:02.000000000 +0000
Change: 2023-02-27 00:44:53.688165013 +0000

Inodes contain important metadata about files, but Linux also uses inodes to
locate files on disk and manage them. If you read or write to a file, Linux uses the
inode id to look the file up and retrieve its contents for you in the background.
Something surprising is that the number of inodes available are limited. It’s
possible to run out of inodes even if your system still has tons of space left on the
disk. This isn’t a very common scenario, but one you should be aware of! You can
check how many inodes are available with the df -i command:

stetson@linux_rocks:/var/log$ df -i
Filesystem Inodes IUsed IFree IUse% Mounted on
tmpfs 59672 744 58928 2% /run
/dev/vda1 1290240 239867 1050373 19% /
tmpfs 59672 3 59669 1% /dev/shm

52
T.me/nettrain
 tmpfs 59672 3 59669 1% /run/lock
/dev/vda15 0 0 0 - /boot/efi
tmpfs 11934 26 11908 1% /run/user/0
tmpfs 11934 28 11906 1% /run/user/1000

Above, my system has 1050373 inodes free on its man /dev/vda1 disk with
239867 free.

In summary, inodes are a fundamental piece to how Linux operates behind

the scenes that allows it to store and manage metadata about files and
directories!

53
T.me/nettrain
 54
T.me/nettrain
 3. White Belt Command Line Skills - The
basics, file manipulation, etc

Intro to the Shell - pwd, cd and basic ls

We begin our journey into learning the Linux shell with White Belt command-
line skills. The pages ahead describe the most fundamental concepts any Shell
Samurai should master to operate the command-line effectively. After you’ve
mastered these concepts, move on to the Black Belt Command Line Skills in the
next chapter.

The Ominous Blank Prompt

At last, you’ve finally got the Linux Operating System installed. You’re
running a Virtual Machine in the cloud, Windows Subsystem for Linux, or
maybe you’ve even decided to take the leap and install Linux on real physical
computer. Now you’re blankly staring at that ominous root $ prompt,
wondering what to do next…
Let’s proceed onward to victory!
I’ve already explained that the shell is a program that interprets commands
and forwards them to the Linux operating system. The shell is also known as a
terminal, prompt or console.

Your shell probably looks something like this:

username@hostname:current_folder $

For example:

stetson@linux_rocks:~/$

The prompt is composed of: the current user you are logged in as, the
hostname of the machine you’re working on and the current working directory.
In the example above, stetson is the user, linux_rocks is the hostname and ~/ is
the current working directory. ~ is a shortcut that represents your home directory

55
T.me/nettrain
which is located at /home/user_name\

Anatomy of a Linux Command

Linux commands can be pretty intimidating if you’ve never used them before
or never worked at a command-line! Don’t worry. You got this.
Commands are simple words or phrases that we enter into the shell. Many of
them are verbs, describing to the operating system what we’d like it to do for us.
For instance, cp for copy and mkdir for make directory.
Sometimes, we won’t just enter a command or verb, but we’ll also provide
additional data to that same command. These are known as options, switches or
flags. A couple common flags you might use are --help and —version. Other
times, we’ll use flags and provide additional data after them. For example,
consider a command formatted like this:

stetson@linux_rocks:~/$ launch_server --memory 4096 --cpus 2

This command is not real but serves to show how flags could be used to
provide additional parameters like 4096 for the memory value and 2 for the CPU
value. The command we’re running, launch_server will then parse the
arguments we’ve given it and act in a different way!

One Ocean, Many Shells

There are many shells that can be used on Linux and Unix. Bourne shell
(located at /bin/sh on the system) was created at Bell Labs along with Unix and
is one of the first created shells. Almost every version of Linux today uses BASH
or Bourne Again Shell as the default shell. It’s located at /usr/bin/bash. You
might also find DASH, or Debian Almquist Shell on your system. DASH usually
replaces /bin/sh and is used by the system for administrative operations.

Our First Command

What is the $ sign for at the end of our prompt? Are we going to get paid to
use Linux? Maybe some day. For now, though, the symbol right at the end of the
prompt indicates what type of user you’re logged in as. $ indicates you’re logged
in as a normal user and # indicates the system administrator (also known as root,

56
T.me/nettrain
which has permissions to do anything).
The prompt sits there, waiting for us to take control and command Linux to
do our bidding. Let’s try our first command, hostname

Go ahead and type hostname at your prompt and hit enter:

stetson@linux_rocks:/home/stetson$ hostname
linux_rocks

Whoa! We got some text back. Sweet. The bash interpreter took our command,
sent it to operating system and replied with the output that the hostname
command returned. Now check this out, it’s going to blow your mind. Hit the up
arrow on your keyboard. The same hostname command should be back in your
prompt, waiting for you to execute it again. Hit enter again, if you want. Or hit
the down arrow key, and the prompt will be empty again. You can also hit the
left and right arrow keys and delete letters with backspace like you could do on a
GUI (Graphic User Interface). Maybe that wasn’t that mind-blowing, but let’s
venture forward anyway.

The Directory Tree

Linux organizes all its files in a hierarchal directory structure. This is a fancy
way to say that the file structure is shaped like a tree. Here’s an example of what
the Ubuntu directory structure looks like (refer back to the Linux fundamentals
chapter to see an overview of what each directory is used for):

/
|-- bin -> usr/bin
|-- boot
|-- dev
|-- etc
|-- home
|-- lib -> usr/lib
|-- media
|-- mnt
|-- opt
|-- proc
|-- root

57
T.me/nettrain
 |-- run
|-- sbin -> usr/sbin
|-- srv
|-- sys
|-- tmp
|-- usr
|— var

The / or “root” directory is up top and each line below it represents a

directory or folder inside of the root directory. Each of those directories also
contain files and directories themselves.
Don’t be too worried with what every folder does, just keep the general tree
structure in mind for now.

The Working Directory

Let’s try another command right now. pwd or Print Working Directory will tell
you where you are in the Linux filesystem. This is also known as the path.

Give pwd a try right now.

stetson@linux_rocks:~$ pwd
/home/stetson

You’ll get back a path to indicate where your prompt or shell is currently
focused in the Linux operating system.
Earlier we mentioned that your prompt shows the current working directory
after the colon and before the $ sign. You might instead see a tilde or ~. The tilde
indicates that the current path is your home directory. Your user might be ubuntu,
so the home path would be /home/ubuntu instead of /home/stetson. The tilde is
simply a shortcut to refer to the current user’s home directory.

List Contents of a Directory

If you want to list the files and directories inside your current working
directory, you can simply enter ls or the list command.

Enter ls now:

58
T.me/nettrain
 #$

stetson@linux_rocks:~$ ls
Desktop Documents Music Pictures Public Templates Videos

Depending on where you ran ls from, you’ll see nothing or a ton of output.
In the example above, every name we see is a directory. We’ll go deeper into the
ls command later.

How do I get to a different directory?

Linux isn’t very useful if we can only hang out in our home directory. We saw
a bunch of directories earlier, so we know that more exist. Let’s try “going”
somewhere else in the filesystem.
We’ll use cd or the change directory command to do that. In order to use cd, we
should first know the difference between Absolute Paths and Relative Paths.

Absolute Paths
Absolute Paths are the “full” directory path from the root directory. Just earlier,
we saw that a user’s home folder might reside somewhere like /home/stetson.
This is an example of an Absolute Path. Whenever I say absolute path to myself, I
slam my first on the table to emphasize absolute. I’m probably weird for this but it
helps me to remember the difference vs Relative Paths.

Relative Paths
Relative Paths are the path referenced from your current directory. For
example, if we’re in /home/stetson and we’d instead like to change our
directory to /home/stetson/Documents, we don’t need to tell cd to go to /home/
stetson/Documents. We can instead tell it to take us to Documents with cd
Documents. The path is relative from where we currently are. It also saves us
time from typing!

With that out of the way, let’s do some filesystem navigation and folder
surfing!

Go ahead and use cd to navigate to the /usr/bin directory

59
T.me/nettrain
 "

stetson@linux_rocks:~$ cd /usr/bin
stetson@linux_rocks:/usr/bin$

We won’t get any text returned back other than a new empty prompt with the
path now changed to /usr/bin.
Let’s cd back to our home directory, this time using either /home/user_name or
just the ~ tilde. Remember that you should see your username all the way left
of the prompt.
cd ~ or cd /home/ubuntu will work the same (assuming that your current
user is named ubuntu).

Once there, issue pwd again to verify your path was changed:

stetson@linux_rocks:/usr/bin$ cd ~
stetson@linux_rocks:~$ pwd
/home/stetson

Let’s learn a few more cd tricks. Just like we can specify ~ to refer to our home
directory, we have a few other shortcuts available to us:

cd ..
.. Refers to the directory “above” our current directory. Give it a try from
your user’s home directory to “go back” or “up” a directory. The pwd command
should return /home. Note that the double periods .. refers to the directory
above and a single period . refers to the current directory.

cd -
Performing cd - will take you back to the directory you were just in. So if
you were in /home/stetson, performed cd .. and then a cd -, you should land
back in /home/stetson again. Try it out now!

Hot Tip : The text you enter after cd is known as an argument. When we
write an argument after a command, we would say that we’re passing an
argument to a command.

60
T.me/nettrain
Your Homework:
• Try cd-ing your way around the Linux filesystem. Don’t worry about
breaking things. You can’t do any harm with only cd. What’s the “longest”
path you can find? What about the shortest?
• Try using cd without any arguments. What happens? Verify with pwd.

more ls, mkdir, cat, touch, copy and move

Learning More ls
We previously used the ls command to list directories and files. However,
there’s one thing I didn’t tell you. Some files and directories are hidden! Any file
in Linux which has a period in front of it is hidden. Luckily for us, we can reveal
them with ls. We’re going to use a flag called a which stands for all.

From your home directory prompt, run ls -a:

stetson@linux_rocks:~$ ls -a
. .. .bash_history .bash_logout .bashrc .profile

We’ll see the . and .., which again reference the current directory and
directory above us. We also should see a few other files that are prefixed with a
period. Like .bashrc, .profile. If you perform ls without the -a flag, you’ll
no longer see the hidden files.

Let’s try adding another flag to ls. We’re using two flags here, a for all and l
for long.

Try entering ls -l -a:

stetson@linux_rocks:~$ ls -la
total 24
drwxr-xr-x 2 stetson stetson 4096 Feb 14 23:36 .

61
T.me/nettrain
 "
drwxr-xr-x 1 root root 4096 Feb 14 23:33 ..
-rw------- 1 stetson stetson 170 Feb 14 23:37 .bash_history
-rw-r--r-- 1 stetson stetson 220 Feb 14 23:33 .bash_logout
-rw-r--r-- 1 stetson stetson 3771 Feb 14 23:33 .bashrc
-rw-r--r-- 1 stetson stetson 807 Feb 14 23:33 .profile

There’s a lot here! The long listing output from ls gives us the file’s
permissions, links to the file, owner name, owner group, size of the file, last
modified time stamp and finally, the name of the file or directory.
The first character of the permissions lets us know if the item is a file or a
directory, indicated by d for a directory and - for a file. Eg: drwxr-xr-x is a
directory, while -rw-r--r-- is a file.
We’ll break down the weird r, w and x flags we see in a later chapter, just keep
in mind they related to the file’s permissions.

Hot Tip : We don’t need to write ls -l -a, we can instead use ls -la.
Give it a try.

One more ls tip. Imagine you have a directory filled with other directories
and files. How can you see the entire output? You can use the -R or recursive
flag to do that.

Observe:

stetson@linux_rocks:~/more_files$ ls -R cloud_files/
cloud_files/:
secret_files

cloud_files/secret_files:
aws

cloud_files/secret_files/aws:
s3

cloud_files/secret_files/aws/s3:
cloud_spend container_audit security_audits

cloud_files/secret_files/aws/s3/cloud_spend:

62
T.me/nettrain
 cloud_files/secret_files/aws/s3/container_audit:

cloud_files/secret_files/aws/s3/security_audits:

Reading Files
Remember that .bash_history file from earlier? Let’s read it. We’ll use the
cat (short for concatenate) command and pass in .bash_history as an argument
to read the hidden file:

stetson@linux_rocks:~$ cat .bash_history

cd /home
cd /home/stetson
cd ..
ls
pwd

Whoa! We get back the commands that we entered earlier. Bash keeps a
history of the commands we entered earlier in .bash_history. Cool!

Let’s Make Our Own File From Scratch

We’re really cooking with gas now. Let’s learn to make our own files with the
touch command. Enter in your prompt the touch command and then any ol’
name you want. Then, use ls to confirm it was created:

stetson@linux_rocks:~$ touch my_brand_spanking_new_file

stetson@linux_rocks:~$ ls
my_brand_spanking_new_file

Above, I created my_brand_spanking_new_file, and then verified it was

created with ls

That’s just one way to use touch. We can also update modified timestamps on
files and directories. Try using touch again, this time specifying a different file.
Use ls -l again and verify the modified timestamp was changed:

stetson@linux_rocks:~$ ls -l
total 4

63
T.me/nettrain
 "
-rw-rw-r-- 1 stetson stetson 0 Mar 14 19:59 file.txt
drwxrwxr-x 3 stetson stetson 4096 Mar 9 19:44 projects
stetson@linux_rocks:~$ touch file.txt
stetson@linux_rocks:~$ ls -l
total 4
-rw-rw-r-- 1 stetson stetson 0 Mar 15 21:22 file.txt
drwxrwxr-x 3 stetson stetson 4096 Mar 9 19:44 projects

Above, I used touch to change the modified time stamp on file.txt from
Mar 14 19:59 to Mar 15 21:22.

Creating Directories
Just as we can create a new file with touch, we can make our own directories
with the mkdir command and an argument. Give it a try now! Then, cd into that
directory, make more files with touch and more directories with mkdir. The
world is your oyster! Err — shell? The terminal is your shell? Never mind.

stetson@linux_rocks:~$ mkdir my_super_awesome_directory

stetson@linux_rocks:~$ cd my_super_awesome_directory/
stetson@linux_rocks:~/my_super_awesome_directory$ touch
another_file
stetson@linux_rocks:~/my_super_awesome_directory$ mkdir
ANOTHER_DIRECTORY

Hot Tip(s) : You can use mkdir with the -p flag to create subdirectories
automatically. For example, mkdir -p my_directory/another_one/secret_files
will make the directory another_one without you having to create it manually.
Just another way we can save ourselves some typing!
You can also give mkdir multiple directories as arguments separated by space
and it’ll make them all next to each other.
For example:

mkdir directory_one directory_two secret_files moon_landing_photos

Copy and Move

Imagine we created a file somewhere on Linux but we wanted it to be
somewhere else. Perhaps we’d like a copy of it to backup.

64
T.me/nettrain
 Simple enough. For this we’ll use the cp and mv commands, which stand for
copy and move, respectively.
Each command takes two arguments:

• The file or directory to be moved

• The path it should be moved to

Here’s an example of using mv:

stetson@linux_rocks:~/my_awesome_directory$ ls
file1.png
stetson@linux_rocks:~/my_awesome_directory$ mv file1.png /tmp
stetson@linux_rocks:~/my_awesome_directory$ ls /tmp
file1.png
stetson@linux_rocks:~/my_awesome_directory$ ls
stetson@linux_rocks:~/my_awesome_directory$

Above, we moved the file file1.png to the /tmp directory.

Using mv will entirely move the file! If we use cp instead, then…you guessed
it, we’ll copy the file. This retains the original file and makes a file where we tell
it to.

Try out the cp command now:

stetson@linux_rocks:~/my_awesome_directory$ ls
file1.png
stetson@linux_rocks:~/my_awesome_directory$ cp file1.png /tmp
stetson@linux_rocks:~/my_awesome_directory$ ls /tmp
file1.png
stetson@linux_rocks:~/my_awesome_directory$ ls
file1.png

Move can also be used to rename a file. Simply give a name as the second
argument. You can move a file and rename it at the same time, too!

65
T.me/nettrain
 Observe:

stetson@linux_rocks:~/my_awesome_directory$ ls
file1.png
stetson@linux_rocks:~/my_awesome_directory$ mv file1.png
my_awesome_file.png
stetson@linux_rocks:~/my_awesome_directory$ ls
my_awesome_file.png
stetson@linux_rocks:~/my_awesome_directory$ mv
my_awesome_file.png /tmp/my_renamed_awesome_file.png
stetson@linux_rocks:~/my_awesome_directory$ ls /tmp
my_renamed_awesome_file.png

We took file1.png and renamed it to my_awesome_file.png. Then, we moved

and renamed it once again, putting it in the /tmp directory and calling it
my_renamed_awesome_file.png.

Copying Directories
Let’s say we want to move an entire directory that contains some files:

stetson@linux_rocks:~$ cp my_awesome_directory/ another_directory

cp: -r not specified; omitting directory 'my_awesome_directory/'
stetson@linux_rocks:~$ ls -l
total 4
drwxrwxr-x 2 stetson stetson 4096 Feb 16 00:46 my_awesome_directory

We get this strange error message about an -r flag. What gives? Linux wants
us to use the recursive flag to copy any files and directories inside the directory.

Let’s try again:

stetson@linux_rocks:~$ cp -r my_awesome_directory/
another_directory
stetson@linux_rocks:~$ ls -l
total 8
drwxrwxr-x 2 stetson stetson 4096 Feb 16 00:48 another_directory
drwxrwxr-x 2 stetson stetson 4096 Feb 16 00:46 my_awesome_directory

66
T.me/nettrain
 "
Nice! That time it worked!

Hot Tip : Wildcards can be used in many places in the bash shell. Let’s say
we have a directory with a bunch of different file extensions, but we only want to
copy the files with .png extensions. We can use a wildcard to do so:

stetson@linux_rocks:~/important_files$ ls -l
total 0
-rw-rw-r-- 1 stetson stetson 0 Feb 16 00:53 AWS_Users.pdf
-rw-rw-r-- 1 stetson stetson 0 Feb 16 00:53 TPS_Reports.pdf
-rw-rw-r-- 1 stetson stetson 0 Feb 16 00:53 file1.png
-rw-rw-r-- 1 stetson stetson 0 Feb 16 00:53 file2.jpeg
-rw-rw-r-- 1 stetson stetson 0 Feb 16 00:53 house_inspection.jpeg
-rw-rw-r-- 1 stetson stetson 0 Feb 16 00:53 jazz.mp3
-rw-rw-r-- 1 stetson stetson 0 Feb 16 00:53 my_dog.png
-rw-rw-r-- 1 stetson stetson 0 Feb 16 00:53 rock_n_roll.mp3
-rw-rw-r-- 1 stetson stetson 0 Feb 16 00:53 vacation.png
stetson@linux_rocks:~/important_files$
stetson@linux_rocks:~/important_files$ mv *.png /tmp
stetson@linux_rocks:~/important_files$ ls /tmp
file1.png my_dog.png my_renamed_awesome_file.png vacation.png

We simply used mv *.png /tmp to move ALL of the png files to /tmp.

Well done on learning ls, mkdir, cat, touch, copy and move. You’re well on
your way to honing your Shell Samurai katana.

Hot Tip : You should know about about tab completion. Almost any time
while using the Linux shell, you can hit your tab key and your shell will attempt
to auto-complete whatever it is you’re typing. For example, if you’re in a
directory and trying to cat a file called
super_long_file_name_that_you_dont_want_to_type.txt, you can type super and
whack the tab key and bash will auto-complete the file name for you. Note that if
you have other files in that directory also starting with super, bash will cycle
through them. If there are two other files, superman and super_awesome.txt,

67
T.me/nettrain
bash will think you want one of those. Just keep hitting tab until you cycle back
to the file you intended.

Users, Groups, Root and Sudo

Like most modern operating systems, Linux is a multi-user operating system.
It was designed this way from the beginning and that functionality has some
impact on the way it works. In Linux, files, directories and even processes are all
owned by a user and group. Many background processes or daemons run in the
background and are assigned to various users and groups. Linux uses User IDs
and Group IDs to keep track of all of these permissions.

To see your assigned user id and group id, simply issue the id command:

stetson@linux_rocks:~$ id
uid=1000(stetson) gid=1000(stetson) groups=1000(stetson),27(sudo)

You’ll see that my user has id 1000, group id 1000 and belongs to two groups:
stetson and sudo.

We as users and administrators work with usernames, but the Kernel works
with user ids.

Introducing Root
You may have come across the term root access before. The root user has
permissions to do anything. Root can read any file or terminate any process. The
root user, has root access. For this reason, root access should be heavily guarded
and secured! You wouldn’t want anything bad to happen to your system, would

68
T.me/nettrain
you? (Not a threat).

As your regular non-root user, try now to cat the /etc/shadow file:

stetson@linux_rocks:~$ cat /etc/shadow

cat: /etc/shadow: Permission denied
stetson@linux_rocks:~$
stetson@linux_rocks:~$ ls -l /etc/shadow
-rw-r----- 1 root shadow 799 Feb 17 18:49 /etc/shadow
stetson@linux_rocks:~$

/etc/shadow is a protected file containing encrypted user passwords. We are

unable to read the file because it’s only accessible to the root user!

Let’s use the sudo (or superuser do) to change that:

stetson@linux_rocks:~$ sudo cat /etc/shadow

[sudo] password for stetson:
root:*:18891:0:99999:7:::
daemon:*:18891:0:99999:7:::
bin:*:18891:0:99999:7:::

Depending on how you installed Linux, you may not have the sudo command
available! If you’re on Ubuntu and get a command not found error, you’ll want
to install sudo with apt-get install sudo. You’ll also want to ensure that your
user has sudo permissions! Let’s see an example.

Sudo
Sudo or Super User Do is an effective tool that lets non-root users run programs
as super user or root. It’s best practice to not manage Linux systems while
logged in as root. As root, you have unlimited power, so what if you
accidentally deleted a bunch of directories you didn’t mean to? Sudo can act as a
temporary elevation of privileges that allows us to install software, delete files or
run other programs.
Many of the commands in this book may require sudo to be used right before

69
T.me/nettrain
 "
the command in order for it to run successfully. For example, you might run sudo
apt-get install nginx to install the Nginx web server. The same command
would fail if we only used apt-get install nginx and ran it as a non-root user.
Hot Tip : Many times while writing commands in my own terminal, I’ll
forget to pre-pend them with sudo, and I get a permission denied error. Rather
than scroll up and put sudo before them or rewriting the entire command, I’ll
instead just use sudo !!. The double-exclamation !! characters tell bash to re-
run the previous command with sudo. Performing sudo !! replaces the !! with
my original command on the previous line. This allows me to be speedier on the
command line.

Once you have sudo installed, you’ll need to add your normal user to the
sudo group from the root account. Use adduser <user_name> sudo to do so:

root@linux_rocks:/# adduser stetson sudo

Adding user `stetson' to group `sudo' ...
Adding user stetson to group sudo
Done.

Note that I am logged in as root while performing this command.

Running as Root…Cautiously
I mentioned that you shouldn’t run most commands as root, but sometimes
you just don’t have a choice. Another way to perform tasks that require
permissions to be elevated is to simply become root. You can do so by running
su, short for switch user. You’ll then need to enter the root password to become
root:

stetson@linux_rocks:~$ su
Password:
root@linux_rocks:/home/stetson# whoami
root

70
T.me/nettrain
 With great powers, come great responsibilities. You want to be extra careful
while running commands as root because you could make a mistake that may
alter or break the system.
As root, you can also use su to switch back to another user:

root@linux_rocks:/# su stetson
$ whoami
stetson

Adding Users
Adding users is straightforward. The primary command we can use is
useradd. This command allows us to specify any attributes we’d like along with
the user’s name.

Let’s go ahead and add a user now (you’ll need root or superuser permissions
to do so):

stetson@linux_rocks:~$ useradd bob

useradd: Permission denied.
useradd: cannot lock /etc/passwd; try again later.
stetson@linux_rocks:~$
stetson@linux_rocks:~$ sudo useradd bob
[sudo] password for stetson:
stetson@linux_rocks:~$
stetson@linux_rocks:~$ cat /etc/passwd | grep bob
bob:x:1001:1001::/home/bob:/bin/sh

Above, I added bob, but without sudo permissions, so I was denied access. I
used sudo to then add the user again and then checked the /etc/passwd directory
to verify that bob shows up. What is /the etc/passwd file?

/etc/passwd
The /etc/passwd file is a special type of file that serves as the central area for
storing user account information. Each line in the file represents a user account
and has a series of fields separated by colons, which provide additional details

71
T.me/nettrain
about the user.
The primary use of /etc/passwd is to allow the system to match usernames
with corresponding user ids during authentication. When a user logs in, Linux
checks that the credentials match what’s stored in /etc/passwd. The file also does
a few other administrative tasks like determining a user’s default groups and
primary group membership.
/etc/passwd used to be used to store encrypted passwords, but modern
systems store encrypted passwords in a file called /etc/shadow we worked with
earlier. /etc/shadow has a few more security restrictions in place.

Changing Your Password

When we change a user’s password, we interact with the /etc/passwd file!

Let’s try it out with our new user, bob, now:

stetson@linux_rocks:~$ sudo passwd bob

New password:
Retype new password:
passwd: password updated successfully
stetson@linux_rocks:~$ su bob
Password:
$ whoami
bob

Above, I tell Linux I want to give bob a new password. After confirming the
password, I’m able to use su or switch user to switch to bob with the new
password. Finally, I use the whoami command to verify I’m logged in as bob.

Changing a user’s default shell

You might have noticed that when su’ing to user, they have just a $ sign as the
prompt, rather than the full path name and username we are used to. This is
likely because the user’s default shell is sh, rather than bash.
You can check which shell a user is currently using with the command ps -p
$$. ps is a command that shows us process ids. The -p flag specifies a process
id we want to see. Finally, $$ is a special variable that gives us the process ID of

72
T.me/nettrain
the current shell instance. We’ll get deeper into processes later, so don’t worry if
this doesn’t make sense right now!

$ ps -p $$
PID TTY TIME CMD
1139 pts/0 00:00:00 sh

You can then use usermod --shell /bin/bash <user_name> as either root or
a sudo user to change a user’s default shell. Below, I change my default shell as
stetson, exit out of the shell and then become stetson again with su:

$ whoami
stetson
$ sudo usermod --shell /bin/bash stetson
$ exit
root@linux_rocks:/# su stetson
stetson@linux_rocks:/$

What’s a group anyway?

In Linux, a group is a set of users. Groups makes managing permissions easier
as we can add a user to multiple groups or multiple users to a single group. Each
user belongs to at least one primary group, which is automatically created when
the user is added. That group is also named the same as the user! If my user is
named bob, his group will also be named bob. This behavior aims to simplify
group management. Users can also be members of multiple secondary groups as
well.
We can check out the groups that any user belongs in with the groups
command. Let’s check in on bob and then use groups without any arguments to
check our own groups. Finally, we’ll also check out root:

stetson@linux_rocks:~$ groups bob

bob : bob
stetson@linux_rocks:~$ groups
stetson sudo
stetson@linux_rocks:~$ groups root

73
T.me/nettrain
 root : root

File Permissions
If you recall earlier, we performed an ls with the -l or long flag and got back
a bunch of output. The first column returned had a combination of r, w, and x
characters to indicate file permissions. Let’s take a closer look and demystify file
permissions.
On your command line, perform ls -l /etc or just follow along with the
example here:

stetson@linux_rocks:~/important_files$ ls -larth
total 36K
drwxr-xr-x 5 stetson stetson 4.0K Feb 16 00:52 ..
-rw-rw-r-- 1 stetson stetson 12 Feb 16 22:46 jazz.mp3
-rwxrwxrwx 1 stetson stetson 12 Feb 16 22:46 rock_n_roll.mp3
-rw-rw-r-- 1 stetson stetson 13 Feb 16 22:47 file2.jpeg
-rw-rw-r-- 1 news news 13 Feb 16 22:47
house_inspection.jpeg
-rw-rw-r-- 1 root root 13 Feb 16 22:47 AWS_Users.pdf
---x--x--x 1 stetson stetson 13 Feb 16 22:47 TPS_Reports.pdf
drwxrwxr-x 2 stetson stetson 4.0K Feb 16 22:58 folder_one
drwxrwxr-x 3 stetson stetson 4.0K Feb 16 22:58 .

Take a look at the first column: drwxrwxr-x

The first character denotes if we’re looking at a directory or a file. A d
indicates a directory and a dash - indicates a file.
The next characters determine access rights for users, specifically, the file
owner, the group owner and everyone else. Each set of 3 characters grouped
together lets us know if the user, group or everyone else can read, write or execute
the file.
Take a look at the diagram below breaking down the string of characters:
drwxrwxr-x:

74
T.me/nettrain
 Finally, the third and fourth columns tell us the file owner and group owner.
Most of these files are owned by the user stetson and the group stetson. The
AWS_Users.pdf file is owned by the root group and the root user. The
house_inspection.jpeg file is owned by the news group and news user.

Introducing chmod
If we’d like to change file or directory permissions, we can do so with the
chmod command in two ways. The first way is to use chmod ugo+rwx <file>.
Here, the u, g and o characters stand for user, group, other and rwx for read,
write and execute. Each of these characters are optional, depending on the
permissions we desire. We then use the + to add the permissions. If instead, we
wanted to remove permissions, we would replace the + with a - to do so. Here’s
an example:

75
T.me/nettrain
 stetson@linux_rocks:~/more_files$ ls -l
total 276
-rw-rw-r-- 1 stetson stetson 280245 Feb 17 18:19 jazz.mp3
stetson@linux_rocks:~/more_files$ chmod ugo+rwx jazz.mp3
stetson@linux_rocks:~/more_files$ ls -l
total 276
-rwxrwxrwx 1 stetson stetson 280245 Feb 17 18:19 jazz.mp3

We added all permissions to the file jazz.mp3. This is usually a very bad idea
because it gives any user on the system the ability to read and modify our file.
Let’s remove our permissions from the file so no one can modify it, and then add
back some permissions just for the user:

stetson@linux_rocks:~/more_files$ chmod ugo-rwx jazz.mp3

stetson@linux_rocks:~/more_files$ ls -l
total 276
---------- 1 stetson stetson 280245 Feb 17 18:19 jazz.mp3
stetson@linux_rocks:~/more_files$ chmod u+rx jazz.mp3
Try 'ls --help' for more information.
stetson@linux_rocks:~/more_files$ ls -l
total 276
-r-x------ 1 stetson stetson 280245 Feb 17 18:19 jazz.mp3
stetson@linux_rocks:~/more_files$

Observe how, after we removed every permission, the permission string is

just a series of dashes: ----------

I also mentioned that there’s another way to change file and directory
permissions. The second way is to use chmod with a binary format to change
permissions. Take a look at the following command:

chmod 777 jazz.mp3

This will give jazz.mp3 FULL permissions for anyone. Each number
corresponds to user, group or other. Each number tells Linux what each of those
groups of users can do. They are in binary format:

• 4: read bit

76
T.me/nettrain
 • 2: write bit
• 1: execute bit

If we add 4, 2, and 1, we get 7. That means full permissions. If instead we

wanted only execute permissions for every user, we could do chmod 111
jazz.mp3. If we wanted: read and write for the user, read, write and execute for
the group and finally, only read and execute for other, we would do: chmod 675.
This style of changing file permissions takes some getting used to, but is
commonly used! Give it a try on your own system, but keep in mind that
changing permissions incorrectly on a real system can be a security flaw!

Introducing chown
We’ve just learned how we might change access permissions on a file. How
could we change the group and owner of a file? Recall that ls -l gives us back
the user and group in the third and fourth columns:

stetson@linux_rocks:~/important_files$ ls -l
total 28
-rw-rw-r-- 1 root root 13 Feb 16 22:47 AWS_Users.pdf
---x--x--x 1 stetson stetson 13 Feb 16 22:47 TPS_Reports.pdf
-rw-rw-r-- 1 stetson stetson 13 Feb 16 22:47 file2.jpeg
drwxrwxr-x 2 stetson stetson 4096 Feb 16 22:58 folder_one
-rw-rw-r-- 1 news news 13 Feb 16 22:47
house_inspection.jpeg
-rw-rw-r-x 1 stetson stetson 12 Feb 16 22:46 jazz.mp3
-rwxrwxrwx 1 stetson stetson 12 Feb 16 22:46 rock_n_roll.mp3

Let’s use chown (change ownership) to modify that!

stetson@linux_rocks:~/important_files$ ls -l
total 28
-rwxrwxrwx 1 stetson stetson 12 Feb 16 22:46 rock_n_roll.mp3
stetson@linux_rocks:~/important_files$
stetson@linux_rocks:~/important_files$
stetson@linux_rocks:~/important_files$ sudo chown alice
rock_n_roll.mp3
[sudo] password for stetson:

77
T.me/nettrain
 stetson@linux_rocks:~/important_files$ ls -l
total 28
-rwxrwxrwx 1 alice stetson 12 Feb 16 22:46 rock_n_roll.mp3

Above, we used chown to set Alice as the user owner of the file
rock_n_roll.mp3.
Observe that Alice is now the owner of the file.
How can we change the group?
You can use the chgrp command (change group):

stetson@linux_rocks:~/important_files$ sudo chgrp root

rock_n_roll.mp3
stetson@linux_rocks:~/important_files$ ls -l
total 28
-rwxrwxrwx 1 alice root 12 Feb 16 22:46 rock_n_roll.mp3

Alice is now the owner of the file and the group root owns the file.

One more trick and we’ll move on. If we want to set the group and the user at
the same time, we can use chown.

Check it out, just give the user and then the group separated by a colon:

stetson@linux_rocks:~/important_files$ sudo chown bob:stetson

rock_n_roll.mp3
stetson@linux_rocks:~/important_files$ ls -l
…
-rwxrwxrwx 1 bob stetson 12 Feb 16 22:46 rock_n_roll.mp3

bob is now the user owner and stetson is the group owner.

Flying Around the Command Line with Shortcuts

78
T.me/nettrain
 If you’ve ever watched a seasoned systems administrator at the command-
line at work, you’ll notice that they seem to fly around the command-line with
ease. They’re able to delete words quickly, move their cursor to the front and
back of commands and quickly replace text. How do they do it? They’re most
likely using shortcuts. We can’t usually use the mouse in the Linux CLI, so
shortcuts help us to get around quickly.
Let’s try a few shortcuts out. It’s helpful if you type some text on your shell
first, so that we can try these examples. Anything will work, maybe a sentence or
two about your day or even a real world command like sudo apt-get install
nginx. Go ahead and type your sentence, but don’t hit enter yet:

stetson@linux_rocks:~$ sudo apt-get install nginx

Your cursor will now be at the end of the line.

Now, hit Control+A on your keyboard and your cursor will fly back to the
beginning of the text. Hit Control+E and your cursor goes back to the end. If
you’d like to go one word back, don’t use the left arrow, just use Alt+B. Similarly,
use Alt+F to move forward one word.
To delete the entire line of text to the left of your cursor, use Control+U. To
delete the entire line of text to the right of your cursor, use Control+K. Finally, if
you want to delete just one word, use Control+W.

Here’s a summary cheat sheet:

• Control+E - Move to End of Line

• Control+A - Move to Beginning of Line
• Alt+B - Move one word Back
• Alt+F - Move one word Forward
• Control+W - Delete One Word
• Control+U - Delete everything to the left of the Cursor
• Control+K - Delete everything to the right of the Cursor

There are more Linux shortcuts and even more that you can custom code

79
T.me/nettrain
yourself, but knowing these few will have you flying around the shell like a
Linux Samurai in no time.

file
The file command can quickly tell us what type of file a given file is. Here’s a
few examples:

stetson@linux_rocks:~/home/Desktop$ file python.csv

python.csv: ASCII text
stetson@linux_rocks:~/home/Desktop$ file ticket1.pdf
ticket1.pdf: PDF document, version 1.3, 1 pages
stetson@linux_rocks:~/home/Desktop$ file axe_throwing.jpg
axe_throwing.jpg: JPEG image data, JFIF standard 1.01, aspect
ratio, density 72x72, segment length 16, Exif Standard: [TIFF image
data, big-endian, direntries=12, manufacturer=Apple, model=iPhone 12
Pro Max, orientation=upper-left, xresolution=182, yresolution=190,
resolutionunit=2, software=16.0.2, datetime=2022:12:11 13:32:59,
hostcomputer=iPhone 12 Pro Max], baseline, precision 8, 4032x3024,
components 3
stetson@linux_rocks:~/home/Desktop$ file farnam_street.html
farnam_street.html: HTML document text, Unicode text, UTF-8 text,
with very long lines (38914)

File determines the file type and sometimes even has even more details. The
jpg photo above I took on my iPhone gives back model of phone, resolution and
more.

File can also look inside zipped files:

file -z foo.zip

rm — Proceed with extreme caution

We’ve so far learned all sorts of ways to create files, copy them and move
them around. How can we remove a file? Simple! Just use rm (or remove):

80
T.me/nettrain
 stetson@linux_rocks:~/more_files$ touch jazz.mp3
stetson@linux_rocks:~/more_files$ ls -l
total 0
-rw-rw-r-- 1 stetson stetson 0 Feb 18 09:18 jazz.mp3
stetson@linux_rocks:~/more_files$ rm jazz.mp3
stetson@linux_rocks:~/more_files$ ls -l
total 0
stetson@linux_rocks:~/more_files$

Above, I create jazz.mp3 with the touch command. Then, I list it to show
that it exists. Once I’m tired of listening to jazz, I decide to delete the file with rm
jazz.mp3. Try creating and deleting some random files on your system now.
A word of caution! Linux isn’t like other operating systems. There is no trash
can or recycle bin for deleted files. When you delete them, they’re gone for good.
However, some files and directories are write protected to guard against
accidental deletion. They’ll ask you to confirm before deleting:

stetson@linux_rocks:~/more_files$ rm jazz.mp3
rm: remove write-protected regular empty file 'jazz.mp3'? y

If you’d just like to delete it instead without being asked, supply the -f or
force flag. It should work if you have the right permissions:

stetson@linux_rocks:~/more_files$ rm -f jazz.mp3
stetson@linux_rocks:~/more_files$

If you’d like to write-protect your own files, we can use our chmod tricks we
picked up in the previous section:

Just use chmod go-w file.txt to remove the write permission from the
group and owner.

To delete many files at once, we can use wildcards or the asterisk to specify
what to delete. This is very dangerous, but hey, you can still do it!

Check it out, let’s blow away all the files inside the secret_files directory
now:

81
T.me/nettrain
 "

stetson@linux_rocks:~/secret_files$ ls
AWS_Report.pdf Classified_Documents.pdf Home_Inspection.pdf
My_dogs.png jazz.mp3 rock_n_roll.mp3
stetson@linux_rocks:~/secret_files$ rm *
stetson@linux_rocks:~/secret_files$ ls
stetson@linux_rocks:~/secret_files$

If instead we wanted to delete all .pdf files, we can use rm *.df. This feature is
called shell globbing, which is used to match and expand on patterns.

Hot Tip : Just another word of caution. rm is a very powerful command! If

you aren’t careful, you could end up breaking your Linux install or taking down
a production system. I really enjoyed this Server Fault answer I read recently:

when you have root, think of yourself as driving a bus full of

school children. Sometimes you can rock out to the song on the
radio, but other times you need to look both ways, slow things
down, and double check all your mirrors.

Be very careful and don’t use root unless absolutely necessary. Most
important system files you won’t be able to delete as a regular user, only as root.

Deleting Directories
Deleting directories isn’t hard, but we’ll need to use the -r flag (recursive) if
there’s any files or other directories inside the directory:

stetson@linux_rocks:~/more_files$ ls
secret_files
stetson@linux_rocks:~/more_files$ rm secret_files/
rm: cannot remove 'secret_files/': Is a directory
stetson@linux_rocks:~/more_files$ rm -r secret_files/
stetson@linux_rocks:~/more_files$ ls
stetson@linux_rocks:~/more_files$

Lastly, if a directory is empty, we can simply remove it with the rmdir

command:

82
T.me/nettrain
 stetson@linux_rocks:~/more_files$ ls
stetson@linux_rocks:~/more_files$ touch empty_directory
stetson@linux_rocks:~/more_files$ ls empty_directory
empty_directory
stetson@linux_rocks:~/more_files$ ls
empty_directory
stetson@linux_rocks:~/more_files$ rmdir empty_directory
stetson@linux_rocks:~/more_files$ ls
stetson@linux_rocks:~/more_files$

That’s the basics of rm! Powerful, dangerous, awesome. Respect it.

Helping Yourself with help, man and tldr

Linux sure has a disorienting number of commands. It can be hard to keep
them all straight and know exactly how to use them. Do Shell Samurais really
know all of this by memory? From my experience working with tons of smart
engineers…no. We all frequently google things and reference documentation.
RTFM (Read the F*cking Manual) is a super power. Knowledge is a spectrum
and it’s impossible to know it all. Don’t beat yourself up. I’ve found that once
you get over the hump of learning and know how to search and read
documentation, you’ll be moving smoothly.
The first command to help ourselves is simply help. Use help and then the
command you need help with or use help as a flag on other commands. For
example, ls —help
The help command is a good starter, but it really doesn’t cover all that much.
There are maybe 60 or so commands that help has documentation on. Luckily, we
have another tool we can use.

In the Beginning, There Was Man

Our second documentation command is man, short for manual. Man pages are
built into nearly every Linux operating system and contain a ton of in-depth
documentation on how to use commands. Give man a try with ls, pwd, echo

83
T.me/nettrain
or any other commands we’ve learned to view the man page. For example: man
ls
You’ll notice that when you use the man command, your entire screen may be
taken up with text! Man is using the less command to read text. It’s a little
intimidating, but you can navigate it easily with some practice. Simply use the
up and down arrow keys (or j and k keys) to browse up and down. If you want
to leave the man page, press q and you’ll return back to your friendly and
welcoming shell.

Understanding Man Sections

Something that always confused me when I started using man pages is that
man pages have sections. To see what I mean, take a look at the output of man
man in your terminal, which is the man page for man:

DESCRIPTION
man is the system's manual pager. Each page argument
given to man is normally the name of a program, utility or function.
The manual page associated with each of these arguments is then
found and displayed. A section, if provided, will direct man to
look only in that section of the manual. The default action is to
search in all of the available sections following a pre-defined order
(see DEFAULTS), and to show only the first page found, even if page
exists in several sections.

The table below shows the section numbers of the manual followed by
the types of pages they contain.

1 Executable programs or shell commands

2 System calls (functions provided by the kernel)
3 Library calls (functions within program libraries)
4 Special files (usually found in /dev)
5 File formats and conventions, e.g. /etc/passwd
6 Games
7 Miscellaneous (including macro packages and conventions), e.g.
man(7), groff(7)
8 System administration commands (usually only for root)
9 Kernel routines [Non standard]

84
T.me/nettrain
 What the heck? Why are there so many categories? We didn’t provide a
section number for our previous man commands like man ls or man man. What’s
going on here?
Not all man pages have an entry in one of these sections. Man pages generally
only have sections that are appropriate to that command. When you type man
and a command, man will search through the sections until it finds a match and
then return that page. If you’d like to see what sections a specific command has
an entry in, you can provide the f flag, like so:

stetson@linux_rocks:~$ man -f intro

intro (2) - introduction to system calls
intro (8) - introduction to administration and
privileged commands
intro (4) - introduction to special files
intro (3) - introduction to library functions
intro (6) - introduction to games
intro (5) - introduction to file formats and filesystems
intro (7) - introduction to overview and miscellany
section
intro (1) - introduction to user commands

To read a specific section, use man <section> <command>:

stetson@linux_rocks:~$ man 6 intro

If you’d like to go a bit deeper, you can explore the /usr/share/man

directory, which is generally where man pages are stored. Be sure to also check
the $MANPATH variable to see if this has bee overwritten (echo $MANPATH)
If you wanted to list all section 2 man pages, you could ls on /usr/share/
man/man2:

ls /usr/share/man/man2/

You can also work backwards and look up the files for each man page with
man -wa <command>:

stetson@linux_rocks:~$ man -wa open

85
T.me/nettrain
 /usr/share/man/man2/open.2.gz
stetson@linux_rocks:~$ man -wa cat
/usr/share/man/man1/cat.1.gz

Ok, we’ll leave it here! To be quite honest, I never use man page sections
myself. I generally just use man with no arguments and read through whatever
documentation I need that way. I did want to explain manual sections because it
was always a very confusing topic and I didn’t want to leave you missing out!

TLDR
One more option for reading docs I love is a package called tldr. Sometimes,
man pages are super long. Sometimes we just don’t have time to read all of that!
(Ironically, if we did, we probably wouldn’t need them as frequently).
Introducing, TLDR. TLDR is short for Too Long, Didn’t Read. It’s a community-
maintained help page that gives just the bare essentials to use a command. If
you’re on ubuntu, you can install tldr with apt-get install tldr

Check out how simple it is:

root@linux_rocks:/# tldr echo

echo
Print given arguments.More information: https://www.gnu.org/
software/coreutils/echo.

- Print a text message. Note: quotes are optional:

echo "{{Hello World}}"

- Print a message with environment variables:

echo "{{My path is $PATH}}"

- Print a message without the trailing newline:

echo -n "{{Hello World}}"

- Append a message to the file:

echo "{{Hello World}}" >> {{file.txt}}

- Enable interpretation of backslash escapes (special characters):

echo -e "{{Column 1\tColumn 2}}"

86
T.me/nettrain
Official Documentation
Two other great resources are the Official Ubuntu Documentation and the
Debian Handbook. These online resources really go deep and into more than just
commands, but they’re worth knowing about and perusing.

Together with help, man, tldr and online documentation, you should be
able to learn the inner-workings of any Linux command! Helping yourself is the
Shell Samurai way!

87
T.me/nettrain
 Installing Packages and Programs

Introducing…Package Managers
Linux and any other operating system would be pretty useless if we couldn’t
install or update software. Luckily, there’s a solution for this built right-in. We’ll
explore how to use a package manager to install system software. Although
software is usually installed via a package manager, it can also be compiled from
source, too. We’ll cover that in the black belt section of the book.
Depending on the distribution of Linux you’re using, there will be a different
package manager for installing software. Debian (.deb) packages and Red Hat
(.rpm) are the two most popular packages. Others exist, too, like Gentoo and
Slackware. Debian packages are used for distributions like Debian, Ubuntu, and
Linux Mint. Red Hat packages are used in Fedora, CentOS, RHEL, Suse, and
others.

What’s a package anyway?

Packages are all of the files required to install a piece of software, compressed
into a single file. For example, Google Chrome, man pages, ping, media players
like VLC, text editors and any other piece of software can be installed via a
package.
The person (or group of people) who create packages are known as package
maintainers. These are not the same people who wrote the software. Upstream
providers are the people who write the software and provide it to the package
maintainers to be bundled in a package file and distributed.

Where Can You Get a Package?

We now know what a package is…but how do we go about actually getting
one and installing it? Do we go to the store and buy it? No! Our package
management system allows us to download and install packages right from our
system. No need to download and move files around manually (or go to the

88
T.me/nettrain
store, for that matter).
Package Managers are configured to download and install their packages
from a repository. Linux distributions are already pointing at software repositories
that the distributors have configured. These are web servers that are well-known
and contain multiple versions of the software. If you’re on a Debian system like
Ubuntu, you can take a look at those repositories at /etc/apt/sources.list

Let’s Get Ready to Install a Package

We’re almost ready to install our first package. First, though, some routine
maintenance.
We’ll first run the command sudo apt-get update. This will tell our system
to reach out to the configured repositories and update our local package database
with info on package names, versions, dependencies and other meta-data. We
aren’t installing anything just yet.
Next, we’ll run sudo apt-get upgrade. This command will actually upgrade
packages we already have installed to the latest versions available.
It’s a good idea to issue both of these commands to maintain your system.

Install a Package!
Ok, now we are all set to actually install our first package. We’ll use the
command apt-get install <package-name>. Let’s try installing nmap, a
network scanning package. Remember to use sudo! Our package manger will
show us dependencies it needs to install and the size to install the new package.
Go ahead and hit Y and then enter to accept the install.

stetson@linux_rocks:/home/stetson/important_files# apt-get install

nmap
Reading package lists... Done
Building dependency tree
Reading state information... Done
The following additional packages will be installed:
libblas3 liblinear4 liblua5.3-0 libpcap0.8 lua-lpeg nmap-common
Suggested packages:
liblinear-tools liblinear-dev ncat ndiff zenmap
The following NEW packages will be installed:
libblas3 liblinear4 liblua5.3-0 libpcap0.8 lua-lpeg nmap nmap-
common

89
T.me/nettrain
 "
0 upgraded, 7 newly installed, 0 to remove and 0 not upgraded.
Need to get 5668 kB of archives.
After this operation, 26.9 MB of additional disk space will be
used.
Do you want to continue? [Y/n] Y

I’ve omitted the full output here, but you’ll see APT doing its thing and
unpacking and setting up the package. If you don’t see any error messages, the
install should’ve worked fine. Confirm by running which nmap or running the
nmap command. If you’re feeling up for a real challenge, read the nmap
documentation and use it to scan your local network!

Hot Tip : apt can be used in place of many apt-get commands. Instead of
running apt-get install nmap above, we could performed apt install nmap.
So what’s the difference? apt was introduced to be a bit more user-friendly than
apt-get. apt has less configuration options and is considered a bit easier to use.
apt-get has more configuration options and has a some lower-level control. For
basic commands like installing packages, it really doesn’t matter which you use!

Searching for Packages

There are a ton of packages available. How can we search through the haystack
for packages to install on our sytsems? Easy, just use apt search
<search_query>. You’ll notice that this searches through both the description and
the name of the package. If you’d instead like to just search for package names,
use apt search --names-only <search_query>

What in the world is a Dependency?

You’ll note in our example above that we asked APT to install nmap, but then it
went and installed all sorts of other things like lua-lpeg, libpcap, liblinear and
other names we don’t recognize. What’s going on here?
Software packages depend on other pieces of software to do their job.
Software Developers may write code but they have a dirty little secret…they
didn’t write it all by themselves. Developers are putting together lego blocks of
code and pulling in other libraries to help their code do its job.

90
T.me/nettrain
 "
Libraries are collections of pre-written code that perform specific tasks.
Dependencies are just software that is needed by another piece of software to
work. Dependencies could also be device drivers or other software packages.

DPKG - The True Hero

There’s another thing I didn’t tell you about the Debian package management
system. Above, we used APT to install our software. dpkg, or Debian Package, is
the real workhorse here. Dpkg is a tool that APT is using behind the scenes to
install packages. APT builds on dpkg and adds some special features like
managing dependencies, upgrades and searching for packages. Dpkg doesn’t
install dependencies, but it can still be very useful. If we have a .deb package on
our system, we can install it easily with dpkg:

root@linux_rocks:~#dpkg -i my_package_to_be_installed.deb

Removing the package is just as easy:

root@linux_rocks:~# dpkg -r my_package_to_be_removed.deb

What about if we’d like to see all the packages currently installed? Easy.

root@linux_rocks:~# dpkg -l

Use q to close the screen when you’re done reading packages.

Hot Tip : This book has been covering Debian-based Linux distributions like
Ubuntu. Red Hat or .rpm based distributions have a similar package
management process as .deb systems. However, instead of dpkg doing the work
and APT enhancing it, rpm (Red Hat Package Management) does the bulk of the
work and yum (Yellowdog Updater Modified) enhances it.

Compressing Files and Uncompressing Them

You’ve probably worked with compressed files at some point if you’ve ever
used a computer. Extensions like .rar, .zip and others are the types of compressed
files that most “regular people” who don’t work with computers for money have

91
T.me/nettrain
used to unzip songs, movies and office documents for years. Linux can unzip
those types of files, but it also has it’s own built-in utilities and ways of
compressing and archiving files.
We should note an important distinction between compression and archival.
Compression is the act of reducing a file or a bunch of files and directories into
smaller files. Archival is the process of creating a single file that contains one or
more files and directories. Yes, compression and archival are often used together,
but they serve two different purposes.
When working with compressed files in Linux, we’ll use the gzip and tar
tools for our compression and archival needs. There are some other utilities like
zip, unzip, compress and bzip2, but tar and gz are the most common tools for
compression and archival!

Compression with gzip

On Linux operating systems, we generally use gzip to compress files and
create a file with a .gz extension. Let’s give compressing a try now.

Run the following commands:

stetson@linux_rocks:~$ touch important_document.PDF

stetson@linux_rocks:~$
stetson@linux_rocks:~$ gzip important_document.PDF
stetson@linux_rocks:~$ ls
important_document.PDF.gz

If we’d like to unzip the file and remove the .gz extension, simply use gunzip:

stetson@linux_rocks:~$ gunzip important_document.PDF.gz

stetson@linux_rocks:~$ ls
important_document.PDF

Archival with tar

Gzip can compress files down, but it can’t add multiple files to an archive. For
that, we’ll have to use tar. The name comes from tape archive.
Let’s use tar with the c, v and f flags to create an archive, turn on verbose

92
T.me/nettrain
 "
output (this means we get output) and specify a filename. Be sure to use the .tar
extension for your file name!

stetson@linux_rocks:~$ tar czf tar_file.tar important_file

secret_documents moon_landing_photos.png
stetson@linux_rocks:~$ ls
important_file moon_landing_photos.png secret_documents
tar_file.tar
stetson@linux_rocks:~$ rm important_file secret_documents
moon_landing_photos.png
stetson@linux_rocks:~$ ls
tar_file.tar

Notice that tar keeps the original files intact and creates a new archive. Above,
I’ve deleted the original files to prove that our files are not gone.
Now, we can extract the contents of the tar file with the x, v and f flags to
extract, be verbose and specify the file name:

stetson@linux_rocks:~$ tar xvf tar_file.tar

important_file
secret_documents
moon_landing_photos.png
stetson@linux_rocks:~$ ls -l
total 16
-rw-rw-r-- 1 stetson stetson 10240 Feb 27 00:28 important_file
-rw-rw-r-- 1 stetson stetson 0 Feb 27 00:27
moon_landing_photos.png
-rw-rw-r-- 1 stetson stetson 0 Feb 27 00:27 secret_documents
-rw-rw-r-- 1 stetson stetson 212 Feb 27 00:29 tar_file.tar

Above, we see that our tar command shows each file that was extracted right
after we run the command.

Hot Tip : You will sometimes hear .tar files referred to as a tarball. This is an
informal term coming from the name of the tar utility together with ball, which
refers to bundling or balling up multiple files together.

93
T.me/nettrain
All Together Now — Compression and Archival
Tar files will have the extension .tar.gz, implying that we’ll need to first
gunzip to remove the .gz extension and then use tar to un-archive the files. We
work from the outside in.

Indeed, we can use the tools in order, first gunzip and then tar:

stetson@linux_rocks:~$ ls
tar_file.tar.gz
stetson@linux_rocks:~$ gunzip tar_file.tar.gz
stetson@linux_rocks:~$ ls
tar_file.tar
stetson@linux_rocks:~$ tar xvf tar_file.tar
important_file
secret_documents
moon_landing_photos.png
stetson@linux_rocks:~$ ls
important_file moon_landing_photos.png secret_documents
tar_file.tar

An easier option is to use the z flag with tar, which tells tar to use gzip/
gunzip to compress and un-compress, saving us a step:

stetson@linux_rocks:~$ tar czf myArchive.tar.gz secret_documents

moon_landing_photos.png important_file

And to uncompress and unarchive:

stetson@linux_rocks:~$ tar xzf myArchive.tar.gz

stetson@linux_rocks:~$ ls
important_file moon_landing_photos.png myArchive.tar.gz
secret_documents

Tar can be a bit confusing to remember all the options for! Try this trick, to
compress files: Compress Zee Files. For uncompressing: eXtract Zee Files. tar
czf and tar xzf. Simple. Should your life depend on extracting a file with tar,

94
T.me/nettrain
you’ll want to remember this.

The Environment
Let’s jump right into trying a few exercises and learn about the environment.

Type the following commands in your shell and observe the output:

echo $USER
echo $HOME

stetson@linux_rocks:/$ echo $USER

stetson
stetson@linux_rocks:/$ echo $HOME
/home/stetson

These values are called environment variables. To see them all, go ahead and
type env in your shell now. You’ll see a bunch of values come back!
Environment variables are used for programs and scripts running on the
system to learn about the environment in which they’re running. Environment
variables might also be used to set the location of a remote file server, set API
keys, or locations of other files and programs.
We’ll now look at another important environment variable, called PATH.

Please echo $PATH in your shell now:

stetson@linux_rocks:/$ echo $PATH

/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/
games:/usr/local/games:/snap/bin

This environment variable, PATH, is used to specify where Linux should look

95
T.me/nettrain
for executable files any time a shell command is entered. The order of directories
is the order that bash will look for the command that you want to run. Each path
is separated by a colon. You might update your PATH if you download a
program and it’s installed in a non-standard location on your system. An
indicator that you might need to update PATH is if bash returns a command not
found error.

whereis and which

Path lets us specify directories we’d like bash to search for executable
commands. However, what if we want to know what path a command or
executable is being called from? We might update our path to add the location of
a new program, but how will we know that our shell called the proper
executable and not an old one in another location?
I’m glad you asked! We can use the which command to show which
executable is associated with a given command! It searches through the
directories listed in the PATH environment variable we just looked at and finds
the first occurrence of the command.
We can also use the whereis command to determine the path of a binary or
program. The difference is that which shows us just the path of the command
we’re looking for, while whereis shows us all possible paths that are evaluated
and searched, along with the man pages and source files.
Go ahead and try the which and whereis commands now for a few common
commands like ls, cat, and others you’ve learned:

stetson@linux_rocks:/$ which python3

/usr/bin/python3
stetson@linux_rocks:/$ which ls
/usr/bin/ls
stetson@linux_rocks:/$ which cat
/usr/bin/cat

stetson@linux_rocks:/$ whereis python3

python3: /usr/bin/python3.8 /usr/bin/python3 /usr/lib/python3.8 /
usr/lib/python3 /etc/python3.8 /etc/python3 /usr/local/lib/python3.8 /

96
T.me/nettrain
usr/share/python3
stetson@linux_rocks:/$ whereis ls
ls: /usr/bin/ls
stetson@linux_rocks:/$ whereis cat
cat: /usr/bin/cat

In the example above, whereis returns all of the paths that it searches for the
python3 executable in. which outputs the executable path that is actually used.
which and whereis are a great help for troubleshooting commands and
executables!

Shutting the System Down

Sometimes we just get sick of Linux and need a break. I’m starting to feel it
while editing this book. Shutting down or rebooting Linux is simple.
We can shut the Linux system down with the command shutdown -h now.
The -h flag specifies a time. We can tell Linux we’d like to shutdown now or
specify a time in minutes: shutdown -h +5 will tell Linux to shutdown in 5
minutes.
If we’d like to instead reboot the system, we can tell Linux to reboot with
shutdown -r now
Here, the -r flag specifies a time like the -h flag does. If that’s too much to
remember, just use the easy to remember reboot command.

Aliases - Not just for Criminals, Spies and Detectives

Along your shell journey, you’ll find that there are certain commands you
being to type over and over again that are long and tedious. The Linux Shell
Samurai mindset does not conform to the will of such tedious commands!
Wouldn’t it be nice if there was some way to instead use a shortcut and type way
less?
You’re in great luck, friend, aliases do just that!

Let’s write our first alias, enter the following commands at your shell:

97
T.me/nettrain
 stetson@linux_rocks:/tmp$ alias sayhello="echo Hello Shell
Samurais!"
stetson@linux_rocks:/tmp$
stetson@linux_rocks:/tmp$ sayhello
Hello Shell Samurais!

An alias is a shortcut to a short snippet of shell code. Above, we set our alias
sayhello to run the echo command along with some text. Then, we can call
sayhello any time we’d like and get the same echo output. Usually aliases do a
lot more than just echo some text to the command-line. They might search and
parse through history or run a backup job.
One problem with our alias: If we reboot, we’ll lose it!
If we decide our alias is pretty decent and worth keeping around, we’ll need
to make sure it’s loaded again. To do this, we can use the .bashrc file in our
home directory. .bashrc is a hidden file that’s executed every time we login to
our shell. It’s commonly used for setting up shell configurations like colors,
completion and aliases!
Go to your home directory if you aren’t already there. You can use the tilde ~
shortcut to cd to your home directory with cd ~. If you list all files and use the
-a flag to show hidden files, you will see a .bashrc file.
Use your favorite text editor to edit the .bashrc file and add your own alias
now. You can use the same format above and put any linux command inside the
quotes “” to the right:

alias sayhello="echo Hello Shell Samurais!"

Save the file. Next, you’ll need to re-load the .bashrc file for it to take effect
in your current shell. You can do so with source .bashrc
Remember that bash loads the .bashrc on every new login, we are just forcing
it to load the .bashrc after we made our changes.. Now, try running your new
alias, sayhello!

Excellent work! After a while, you’ll begin to identify commands and

processes that you are performing over and over again. Customize your .bashrc
with new aliases to save yourself time, young Shell Samurai!

98
T.me/nettrain
 4. Black Belt Command Line Skills -
Output Redirection, Text and More

Black Belt Intro

In this chapter, we will dive deeper into the world of Linux and explore
advanced topics that will help you become a Shell Samurai. From managing
processes and services on your system to working with system directories like /
proc, you will learn the ins and outs of how Linux works behind the scenes. We
will also cover useful tools like strace and jq, which can help you debug and
process API information more effectively. In addition, we’ll learn how to
schedule tasks with cron, access remote systems with SSH, and work with APIs
using CURL. So buckle up and get ready to become a Linux black belt!

Compiling Code from Source

Previously, we looked at how to use our package manager to install software.

Life didn’t always used to be this easy. Software used to be always compiled
from source. Compiling from source is the process of building software from its
original source code. While that may seem daunting, there are many reasons why
you’d want to compile code from source.
One of the primary reasons is that it allows for greater control and
customization over the software being installed. When using a package manager,
the software comes pre-compiled with default options and settings, but
compiling from source allows us to tailor the software to our specific needs and
preferences.
Additionally, compiling from source can be necessary when using software
that is not available through a package manager, or when a newer version of the
software is needed. By understanding how to compile code from source, we’ll
gain greater flexibility and control over our system.

99
T.me/nettrain
 Let’s compile our first piece of Linux software. We’re going to download and
compile GNU Hello. Hello is a simple piece of software written in C built exactly
for the purpose of teaching beginners to compile code.
Before we get started, we’ll need the build tools that will allow our system to
compile the code. Go ahead and run sudo apt install build-essential to
install these now. Build-essential includes a C compiler, make and a few other
utilities.
Next, ake sure you have wget installed with apt-get install wget. Wget is
a downloader tool we’ll use to download the Hello source.
Now navigate to the GNU Hello home page. Scroll down to download and
click the HTTPS link. You’ll see a ton of pairs of files named similarly:
hello-2.12.tar.gz and hello-2.12.tar.gz.sig. 2.12 is the latest version as of this book’s
writing. The file ending in tar.gz.sig is a signature file used to verify that the
source code hasn’t been tampered with. We want the file ending in tar.gz. Scroll
down to the latest version of the Hello source code and copy the link.
From our terminal, run wget to download the compressed source code:

stetson@linux_rocks:~$ wget
https://ftp.gnu.org/gnu/hello/hello-2.12.tar.gz
Resolving ftp.gnu.org (ftp.gnu.org)... 209.51.188.20,
2001:470:142:3::b
Connecting to ftp.gnu.org (ftp.gnu.org)|209.51.188.20|:443...
connected.
HTTP request sent, awaiting response... 200 OK
Length: 1017723 (994K) [application/x-gzip]
Saving to: 'hello-2.12.tar.gz'

hello-2.12.tar.gz 100%[================>] 993.87K 3.91MB/s in

0.2s

2023-02-19 10:13:47 (3.91 MB/s) - 'hello-2.12.tar.gz' saved

[1017723/1017723]

The source code is compressed as a tar.gz formatted file. Recall that we can
uncompress tar files with tar -zxvf hello-2.12.tar.gz. Go ahead and
uncompress the file, then cd into the directory that was created with cd
hello-2.12 or whatever the version is that you downloaded.

100
T.me/nettrain
 There’s a few README files inside this directory, we can check them out with
the less command. The file called INSTALL tells us how to compile the code. It’s
always a good idea to read the documentation from source code.
The first thing we’ll run inside this directory is the configure shell script. Feel
free to read any of the files we’re running with less or cat. We’ll run configure
with ./configure. Configure will check for dependencies on our system and
throw an error if we’re missing anything. Configure also generates a Makefile, ,
which is a script that defines how the software should be compiled and installed.
After running configure, we run make. The make command uses Makefiles to
understand how to build the software. We installed make with the build tools
earlier. Go ahead and run make now. You’ll see a flurry of text fly by your screen.
If you don’t see any scary looking errors, it’s a good sign that make ran correctly.
Verify by running ls -ltr. This command lists files in reverse order of time
created. You should see a hello file at the bottom of the list. Go ahead and test it
works by running ./hello:

stetson@linux_rocks:~/hello-2.12$ ./hello
Hello, world!

We compiled the code and ran it, but to complete the process we’ll run sudo
make install. This command installs the compiled software onto your system
so that you don’t have to run it from the directory you downloaded the code to
every time. Run sudo make install and then hello without the ./ at the
beginning:

stetson@linux_rocks:~/hello-2.12$ hello
Hello, world!

Hello was installed to our system! You can verify this by running which
hello. This shows us the path that the command hello is running from:

stetson@linux_rocks:~/hello-2.12$ which hello

/usr/local/bin/hello

Congratulations! You’ve compiled your first piece of Linux software from

101
T.me/nettrain
 & "
source!

Hot Tip : Be cautious about installing with make install. Compiling code
from source may install other packages we don’t know about. Instead of using
make install, you can apt-get install checkinstall and instead run
checkinstall to keep track of all the files created by the install script. It’ll even
generate a .deb package file. This makes uninstalling the compiled source easier.
Take a look at the docs for checkinstall.

Managing Processes

What is a process?
A process is simply an instance of a running program. Multiple processes run
on Linux at the same time and the kernel manages and oversees them all!
Each process has a unique ID, called the PID (Process ID) that identifies it. You
can list running processes with the ps command. The process id is assigned in
sequential order as processes are created.
At your shell, run the ps command to see the currently running processes on
your system:

stetson@linux_rocks:~$ ps
PID TTY TIME CMD
5030 pts/0 00:00:00 sh
5031 pts/0 00:00:00 bash
5034 pts/0 00:00:00 ps

ps, or process status gives a very plain and simple overview of currently
running processes. In order from left to right:

• PID: Process ID
• TTY: The controlling terminal of the process
• TIME: Total CPU time used
• CMD: Name of the command or executable that spawned the process

102
T.me/nettrain
 Our ps output is only giving us a few processes back, how come? By default,
we need to provide additional flags to see all of the processes that are running. A
very popular command is ps aux, which will show us processes belonging to
every user. Note that we don’t provide ps aux with a dash - before the aux flags.
The reasoning for this goes back to an older style command flag. For more info
on this, check out this StackExchange post.
Back to ps aux. The a flag tells ps that we’d like to see all processes running,
including other user’s processes. The u flag gives us additional details. The x
flag shows processes that don’t have a TTY associated, which are usually
daemons that are running in the background or part of system boot.
Try out ps aux now in your terminal:

stetson@linux_rocks:/var/log$ ps aux
USER PID %CPU %MEM VSZ RSS TTY STAT START TIME
COMMAND
root 1 0.0 1.7 167772 8132 ? Ss 2022
6:29 /lib/systemd/systemd --system --deserialize 39
root 2 0.0 0.0 0 0 ? S 2022 0:00
[kthreadd]
root 3 0.0 0.0 0 0 ? I< 2022 0:00
[rcu_gp]
root 4 0.0 0.0 0 0 ? I< 2022 0:00
[rcu_par_gp]
root 5 0.0 0.0 0 0 ? I< 2022 0:00
[netns]
root 7 0.0 0.0 0 0 ? I< 2022 0:00
[kworker/0:0H-events_highpri]
root 9 0.2 0.0 0 0 ? I< 2022 238:40
[kworker/0:1H-events_highpri]

We now see a ton of output! No need to memorize everything, but it’s good to
be familiar with some of these columns:

• USER: User that owns the process

• PID: process ID
• %CPU: CPU time used
• %MEM: Ratio of process’ resident set size to the physical memory on the

103
T.me/nettrain
 system
• VSZ: - Virtual Memory Usage of the Process in KiB
• RSS: Resident Set Size - non-swapped physical memory a process is using
in KiB
• TTY: Controlling Terminal
• STAT: Process’ state. For example: Z - Zombie, S - Sleeping, R - Running
• START: Starting time of the process
• TIME: cumulative used CPU Time
• COMMAND: The command that launched the process

ps is a great command to be familiar with, however, it’ll only give us a snapshot

of our processes at the moment we run it. If instead, we wanted to see live,
updating processes, we can use top. Top updates every 3 seconds by default
and lists the “top” processes by activity. Top will show us a summary up top
with a table of processes sorted by most CPU activity used:

stetson@linux_rocks:/home$ top
top - 07:36:40 up 8 days, 7:55, 1 users, load average: 0.00, 0.00,
0.00
Tasks: 12 total, 1 running, 11 sleeping, 0 stopped, 0 zombie
%Cpu(s): 0.4 us, 0.2 sy, 0.0 ni, 99.2 id, 0.0 wa, 0.0 hi, 0.1
si, 0.0 st
MiB Mem : 1988.8 total, 111.1 free, 399.4 used, 1478.2 buff/
cache
MiB Swap: 1024.0 total, 961.5 free, 62.5 used. 1164.6 avail
Mem

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+

COMMAND
1 root 20 0 3872 2860 2584 S 0.0 0.1 0:03.38
bash
4412 root 20 0 50536 212 28 S 0.0 0.0 0:00.00
nginx
4413 www-data 20 0 50864 2056 1892 S 0.0 0.1 0:00.04
nginx
4414 www-data 20 0 50864 1488 1372 S 0.0 0.1 0:00.00
nginx

104
T.me/nettrain
 4415 www-data 20 0 50864 1488 1372 S 0.0 0.1 0:00.00
nginx
4416 www-data 20 0 50864 1492 1372 S 0.0 0.1 0:00.00
nginx
18522 root 20 0 3884 3064 2568 S 0.0 0.2 0:00.74
bash
20611 root 20 0 12220 3468 3184 S 0.0 0.2 0:00.02
sshd
29968 root 20 0 6640 2860 2440 S 0.0 0.1 0:00.07
su
29969 stetson 20 0 2060 452 384 S 0.0 0.0 0:00.00
sh
29970 stetson 20 0 3884 3100 2588 S 0.0 0.2 0:00.05
bash
29979 stetson 20 0 5724 2648 2224 R 0.0 0.1 0:00.04
top

Here’s a breakdown:

The first line provides general system information:

top - 07:36:40: The current time is 07:36:40.

up 8 days, 7:55: The system has been running for 8 days and 7 hours, 55
minutes.
1 users: There is 1 logged-in user currently.
load average: 0.00, 0.00, 0.00: The system load averages for the past 1, 5,
and 15 minutes, respectively.
The second line shows the number of tasks and their states:

12 total: There are 12 total tasks.

1 running: One task is running.
11 sleeping: Eleven tasks are sleeping.
0 stopped: No tasks are stopped.
0 zombie: No tasks are in a zombie state.
The third line displays CPU usage information:

%Cpu(s): The percentage of CPU usage.

0.4 us: 0.4% user processes.
0.2 sy: 0.2% system processes.
0.0 ni: 0.0% nice processes.
99.2 id: 99.2% idle.

105
T.me/nettrain
0.0 wa: 0.0% wait.
0.0 hi: 0.0% hardware interrupts.
0.1 si: 0.1% software interrupts.
0.0 st: 0.0% steal time.
The fourth and fifth lines show memory usage in MiB:

1988.8 total: Total memory is 1988.8 MiB.

111.1 free: Free memory is 111.1 MiB.
399.4 used: Used memory is 399.4 MiB.
1478.2 buff/cache: Memory used by buffers/cache is 1478.2 MiB.
1024.0 total: Total swap memory is 1024.0 MiB.
961.5 free: Free swap memory is 961.5 MiB.
62.5 used: Used swap memory is 62.5 MiB.
1164.6 avail Mem: Available memory is 1164.6 MiB.
From the 7th line onwards, the output lists individual processes with their
details:

PID: Process ID.

USER: The user who owns the process.
PR: Process priority.
NI: Nice value.
VIRT: Virtual memory size.
RES: Resident memory size.
SHR: Shared memory size.
S: Process state (S: sleeping, R: running, etc.).
%CPU: Percentage of CPU usage.
%MEM: Percentage of memory usage.
TIME+: Cumulative CPU time used by the process.
COMMAND: The command that started the process.

Top supports a few keyboard shortcuts for sorting each process by pressing
Shift and one of the following keys:

M - Sort by %MEM column

106
T.me/nettrain
 N - Sort by PID column
T - Sort by the TIME+ column
P - Sort by %CPU column

Htop is another popular package for viewing process usage. The name comes
from the author’s name, Hisham H. Muhammad or Hisham’s top. htop
provides a bit more readable interface for viewing process utilization.

Controlling and Managing Processes

Let’s experiment with controlling processes. For this tutorial, we’ll use the
nano text editor. In one terminal window, launch nano by entering the nano
command. In another terminal window, run ps aux | grep nano. This
command will list all running processes and then use grep to search for the name
we specified:

root@linux_rocks:/# ps aux | grep nano

stetson 31337 0.0 0.1 3108 2356 pts/2 T 07:55 0:00
nano

107
T.me/nettrain
 root 31333 0.0 0.0 2844 652 pts/3 S+ 07:56 0:00
grep --color=auto nano

We get back our process output with the nano process and another line that is
the process for grep we just ran!
In your same terminal window, run kill <process-id>. My nano process id
is 31337. Verify that the process has been killed by grepping for it again with ps
aux | grep nano. We just sent a kill signal to the process and told it to
terminate!

Running a process in the background

Now, let’s experiment with putting nano in the background. Run nano again,
but this time add an ampersand (&) on the end:

stetson@linux_rocks:/$ nano &

[1] 31340

This time, you won’t get a text editor, but instead two numbers. The first is a
job id: 1. The second is our process id: 31340. We’ve just sent nano to be run in
the background, hidden away from our command-line interface. If you run ps
aux | grep nano you can find nano still running as a process:

stetson@linux_rocks:/$ ps aux | grep nano

stetson 31340 0.0 0.1 3356 2568 pts/2 T 08:02 0:00
nano
stetson 31349 0.0 0.0 2844 692 pts/2 S+ 08:04 0:00
grep --color=auto nano

Run the command jobs and we’ll see a list returned with nano:

stetson@linux_rocks:/$ jobs
[1]+ Stopped nano

Let’s get back control of nano. Run fg %1 or simply fg for foreground and
you’ll have the text editor window back. We run fg with the %1 argument to tell
fg we want control of job id 1. Since nano is the only job that is running, we can

108
T.me/nettrain
 !
also just run fg without arguments.

Hot Tip : The ampersand & isn’t the only way to send a process to the
background. You can also use Control+Z on your keyboard, which will send a
SIGTSTP signal to the process and put it in the background. You can still run fg to
get back control of the process. Give it a try now!

Sending a process to run in the background can be helpful when you don’t
have the process managed by systemd and want to get some log output from it
or observe its behavior.

Processes vs Threads
When talking about processes, you’ve no doubt heard of the term threads or
multi-threading. In Linux, processes and threads are both abstractions used by the
system to execute code and allocate resources. There’s a couple differences
between the two concepts in how they are structured, share resources and are
isolated:

Process:

• A process is an independent execution unit with its own address space,

code, data, file descriptors, and environment variables.
• Each process has a unique process ID (PID) that distinguishes it from other
processes in the system.
• Processes are isolated from each other, and communication between
processes requires inter-process communication (IPC) mechanisms, such as
pipes, sockets, or shared memory.
• Creating a new process involves significant overhead, as the operating
system needs to allocate memory, create a new address space, and set up
other resources.
• A process can have one or more threads.

109
T.me/nettrain
 Thread:

• A thread is a lightweight execution unit that runs within the context of a

process.
• Threads within the same process share the process's address space, code,
data, and file descriptors. This means that communication between threads
can be more efficient, as they can directly access each other's memory.
• Each thread has a unique thread ID (TID) within the context of the process
it belongs to, but from the Linux kernel's perspective, threads are treated as
processes with shared resources.
• Creating a new thread requires less overhead compared to creating a new
process, as threads share most of the process's resources.
• Threads are useful for parallelizing tasks within a single process or for
executing tasks concurrently that need access to shared data.

In summary, processes and threads share resources differently and are

isolated differently. Processes are more isolated and have separate address space.
Threads within the same process share resources and communicate more
efficiently.

Understanding processes and how to manage them is an essential Shell

Samurai skill! I hope you found this section on processes interesting.

stdout, stdin and stderr

Ahh, stdout, stdin and stderr. Classic hits. Wait, what are these? Standard Out
(stdout), Standard In (stdin) and Standard Error are standard streams. These are
essential concepts in Linux for input and output communication between
programs.
What is a standard stream? This concept is a little abstract, so stay with me for
a moment. Standard streams are input and output channels that programs use to
communicate. You already know that computers are Input/Output devices. For

110
T.me/nettrain
example, you type a key and it takes your input and outputs it to computer.
Standard streams abstract this idea of physically connected cables and applies
the same design to programs. Data goes from process to process via standard
streams.
With these concepts in mind, let’s take a look at output redirection.

Output Redirection with stdout

From your trusty shell, follow along:

stetson@linux_rocks:~$ echo "I am a Shell Samurai" > quote.txt

Now, cat that file:

stetson@linux_rocks:~$ cat quote.txt

I am a Shell Samurai

Wow! The text was placed in the quote.txt file. How? Output Redirection!
We used the > redirection operator to take the Standard Output of the echo
command and redirected it to the quote.txt file. Instead of echo outputting to
our shell, it sent stdout to the file.
The redirect operator is used to redirect output to a file.
By the way, if you were wondering if you can redirect the output of cat to a
file, you totally can. Check it out:

stetson@linux_rocks:~$ cat quote.txt > another_quote.txt

stetson@linux_rocks:~$ cat another_quote.txt
I am a Shell Samurai

One thing to keep in mind is that using the > output redirection command
will overwrite a file if it already exists. If the file doesn’t exist, it’ll be created. Try
overwriting the quote.txt file now:

stetson@linux_rocks:~$ echo "this is my shell" > quote.txt

stetson@linux_rocks:~$ cat quote.txt
this is my shell

111
T.me/nettrain
 If instead, you’d like to append to the file so that it doesn’t get overwritten, you
can simply use 2 of the output redirection operators (>>). Try it now and append
to the quote.txt file:

stetson@linux_rocks:~$ echo "there are many like it" >> quote.txt

stetson@linux_rocks:~$
stetson@linux_rocks:~$ cat quote.txt
this is my shell
there are many like it

Well done! On to stdin.

Standard In - stdin
Just as we have stdout streams, we also have stdin streams at our disposal.
This is the standard stream that carries input data into programs. Interactive
programs like text editors read from stdin. Stdin is connected to the keyboard
by default.
From our quote.txt file, let’s change it up a bit and flip our output redirection
operator the other way and use <. The < operator indicates stdin redirection.
Here’s an example:

stetson@linux_rocks:~$ grep "shell" < quote.txt

this is my shell

Here, we redirect the contents of our quote.txt file to grep instead of grep
waiting for keyboard input. Grep highlights shell, since it found it in the file.
Stdin redirection can be used with tons of commands! Onwards to stderr.

Standard Error - stderr

Standard Error is our last standard stream. It’s used for error messages mostly.
Let’s try an experiment. We’ll generate an error on purpose by using a flag that
doesn’t exist, samurai. Then, we’ll redirect its output to stderr:

stetson@linux_rocks:~$ ls --samurai > stderr

ls: unrecognized option '--samurai'
Try 'ls --help' for more information.
stetson@linux_rocks:~$ cat stderr

112
T.me/nettrain
 "
stetson@linux_rocks:~$

Where is our output? Our redirect operator > only redirects stdout. Stderr
outputs to the screen, but we can’t redirect it like this. We’ll have to do something
else.
Instead, we’ll use file descriptors. We’ll use the 2> command instead of > and
catch the standard error.

Hot Tip : File descriptors are unique identifiers that can represent an open
file or an input/output stream. When we use > to redirect output, it’s the
equivalent of using 1>. Below, we instead use 2> for stderr. 0> is stdin.

stetson@linux_rocks:~$ ls --samurai 2> stderr

stetson@linux_rocks:~$
stetson@linux_rocks:~$ cat stderr
ls: unrecognized option '--samurai'
Try 'ls --help' for more information.

Nice! Now we’ve got the error message output to our file. This gave us the
error, but if there was any standard output, we wouldn’t see that.
To remedy this, we can use:

2>&1

Let’s break it down. 2> will redirect the stderr stream somewhere. The new
part, &1 will specify that our stderr stream should be the same as the destination
of stdout. We use 1 because it is the file descriptor for stdout.
Here’s an example:

stetson@linux_rocks:~$ ls /tmp/samurai > output.txt 2>&1

stetson@linux_rocks:~$
stetson@linux_rocks:~$ cat output.txt
ls: cannot access '/tmp/samurai': No such file or directory
stetson@linux_rocks:~$
stetson@linux_rocks:~$
stetson@linux_rocks:~$ touch /tmp/samurai

113
T.me/nettrain
 stetson@linux_rocks:~$
stetson@linux_rocks:~$ ls /tmp/samurai > output.txt 2>&1
stetson@linux_rocks:~$
stetson@linux_rocks:~$ cat output.txt
/tmp/samurai

Above, I tried to list a directory that didn’t exist, /tmp/samurai. ls threw an

error and we caught it in the output.txt file. Then, I created that directory and
ran the same command again. That time, we caught the standard output.
Look at this example again:

command > output.txt 2>&1

The order is important here. We run a command and redirect its output to
output.txt. Then with 2>&1 we tell stderr that it should output to the same place
that stdout is outputting. That place is output.txt here.
In conclusion, understanding and utilizing stdin, stdout, and stderr are crucial
for effective Linux usage. Although grasping these concepts might be challenging
and abstract initially, experimenting with various examples and redirection
techniques will help you become proficient in handling standard streams.

114
T.me/nettrain
 Symlinks and Hardlinks

If you’ve used Windows before, you’re probably familiar with the concept of a
shortcut. They’re essentially links to other files on the system. Linux has a similar
concept called symlinks or symbolic links. A symlink acts as a shortcut or
reference to another file or directory on the system. There is also the concept of a
hard link, which is a link to a inode. You might remember that an inode is a data
structure used by the filesystem to to store metadata about a file or directory.
When a program or user attempts to access a symlink, they are redirected to
the original file or directory’s “real” location on disk. Symlinks are extremely
handy for creating shortcuts to frequently used files and directories.
Let’s try a quick example. From your shell, create a new file with echo “hello
Shell Samurais!” > /tmp/shell_samurais.txt. Next, create a symlink with ln
-s /tmp/shell_samurais.txt ~/my_first_symlink. The first argument is the
file we’re linking to, and the second argument is the symlink name.

stetson@linux_rocks:~$ echo "hello shell Samurais!" > /tmp/

shell_samurais.txt
stetson@linux_rocks:~$
stetson@linux_rocks:~$ ln -s /tmp/shell_samurais.txt ~/
my_first_symlink
stetson@linux_rocks:~$
stetson@linux_rocks:~$ ls -l
total 0
lrwxrwxrwx 1 stetson stetson 23 Mar 4 01:34 my_first_symlink -> /
tmp/shell_samurais.txt

If you list the directory contents of your home directory at ~, you’ll see an
arrow -> indicating that my_first_symlink points to /tmp/shell_samurais.txt.
To drive this concept home, open up the symlink in your home directory with
your favorite text editor and delete the contents. Then, add some new text. I’ll
add Hello from VIM!!! - Stetson.
Cat the file in /tmp/shell_samurais.txt and you’ll see that the text has been

115
T.me/nettrain
updated! Whenever a symlink is modified, the file that it points to is modified as
well. The reverse is also true. Modifying a file that a symlink points to will also
update the symlink.
For example, below I update the symlink in the /tmp directory and then view
the symlink at ~/my_first_symlink:

stetson@linux_rocks:~$ cat /tmp/shell_samurais.txt

Hello from VIM!!! - Stetson
stetson@linux_rocks:~$
stetson@linux_rocks:~$ vim /tmp/shell_samurais.txt
… Note: I edited the file in /tmp! …
stetson@linux_rocks:~$
stetson@linux_rocks:~$ cat my_first_symlink
Hello from VIM!!! - Stetson
this was edited from the tmp directory!

Hard links
Hard links are similar to symlinks, but not the same thing. Whereas a symlink
points to the path name of a file, a hard link points to the same physical data on
disk as the original file. It uses an inode to do this. Let’s create a hard link now.
We still use the ln command, but we don’t provide the -s flag.
Below, I create a hard link at /tmp/shell_samurais.txt pointing to ~/
my_first_hard_link:

stetson@linux_rocks:~$ ln /tmp/shell_samurais.txt
my_first_hard_link

You should also know the difference between how a symlink or hardlink
behaves when the file being pointed to is deleted. A symlink will not work any
longer. If you try to cat the contents, Linux will complain that the file doesn’t
exist. A hard link will still show the data, since it’s pointing to the physical data
on disk. The only way to delete the data completely is to remove the original file
and all links to it. Observe what happens when we delete /tmp/
shell_samurais.txt:

stetson@linux_rocks:~$ rm /tmp/shell_samurais.txt

116
T.me/nettrain
 stetson@linux_rocks:~$ cat my_first_hard_link
Hello from VIM!!! - Stetson
this was edited from tmp!
stetson@linux_rocks:~$ cat my_first_symlink
cat: my_first_symlink: No such file or directory
stetson@linux_rocks:~$ ls -l
total 4
-rw-rw-r-- 1 stetson stetson 56 Mar 4 01:38 my_first_hard_link
lrwxrwxrwx 1 stetson stetson 23 Mar 4 01:34 my_first_symlink -> /
tmp/shell_samurais.txt

Another tip to keep in mind is that hard links cannot point to different file
systems since inodes are unique to a file system.
This is all great to know, but when might we actually use a symlink or hard
link? Links are useful for instances where the same file might have to be accessed
from different directories. Symlinks are useful when we need to create a shortcut
to a file or directory in a different location or when we want to create an alias for
a frequently accessed file or directory, while hard links are useful when we need
to make a copy of a file in a different location, but don't want to duplicate the
data or when we need to create multiple names for the same file within the same
file system.
Understanding the difference between these two types of links is important
for efficient file management and optimizing disk space usage. With this
knowledge, you can confidently create and manage links in your Linux system.
Chaining Commands in the Shell with ; and && Operators

Two very handy operators in the shell are the semi-colon (;) and the double-
ampersand && operators. These operators make it easier to execute multiple
commands in sequence or control their execution based on the success of
previous commands.
First, the semi-colon operator can be used to chain commands together that
would normally be all separate commands. Here's an example of how it works:

stetson@linux_rocks:/var/log$ echo 'hello Shell Samurais!'; ls

hello Shell Samurais!
alternatives.log apt bootstrap.log btmp dpkg.log faillog

117
T.me/nettrain
lastlog wtmp

Next, the double ampersand (&&) operator allows you to execute the second
command only if the first command exits successfully. To see it in action, let's try
a command that produces an error:

stetson@linux_rocks:/var/log$ ls -shellSamurais && echo 'Hi again!'

ls: invalid option -- 'e'
Try 'ls --help' for more information.

These two methods of chaining commands will be helpful in your journey,

use them wisely!

118
T.me/nettrain
 Text Processing

Editing Text with Nano

At some point, you’ll need to edit text files on your Linux system. Nano is one
of the easiest text editors for Linux beginners to get started with and boasts a
simple editor. First, install Nano with apt-get install nano. Fire it up with
nano <file_name> where file_name is any file name you’d like. You’ll see a
screen pop up that looks sort of like a regular GUI text editor. You can use the
arrow keys to get around, enter to start a new line and delete to erase characters.
The only difference is that you can’t use the mouse.
Type a few lines of text, move around, and when you're satisfied, save the file.
In the bottom-left corner of your terminal, you'll see ^X Exit. This indicates the
shortcut for saving and exiting: Ctrl + X. Press this key combination, and you'll
be prompted with "Save modified buffer?" Type "Y" to confirm saving, and Nano
will display the file name (pre-filled with the name entered earlier). You can
change it to create a new file or keep the existing name. Press Enter to save and
exit Nano:

Confirm your file was written by cat-ing it from your terminal or using less
to read it.
Nice Work! Nano is a great way to quickly edit text files, but it isn’t the text
editor we’d recommend for long-term use.
For that…we’ll talk about VIM.

119
T.me/nettrain
 Text Editing Like a Boss with Vim

The age-old debate between Vim and Emacs continues to divide text editor
enthusiasts. Both have been around for years, offering rapid editing speed and
extensive customization options. Advocates for each editor often passionately
argue for their preferred choice, leading to heated online debates.
That said, the choice is yours. Explore all your options and weigh out the
tradeoffs and decide what works best for you. For now, let’s focus on learning
Vim, our recommended text editor for Shell Samurai users.

What is Vim?
Vim is a popular text editor and builds upon the VI text editor, which stands
for Visual. Vim is “VI Improved” and it probably comes installed with your
system already. If not, we can install it with our trusty friend, apt-get install
vim
Vim has a notorious reputation for being difficult to learn and even harder to
exit. It’s become so much of a meme that exiting Vim is a popular joke around
how hard it is to get out of Vim and back to the command line!
Despite Vim’s learning curve, I think that it is worth learning since it can help
edit text files much more quickly and efficiently. Don’t feel like you need to learn
Vim all at once. Our goal in Shell Samurai is to get you to the bare minimum
proficiency and then guide you to learn more on your own time, as needed. We’ll
provide some resources at the end of this section that should help to take your
Vim journey to the next level. With that, let’s dive into Vim.

120
T.me/nettrain
A Very Brief Introduction To Vim
You’ve installed Vim or it’s come with your system already. Start Vim with the
simple command, vim. Whoa! Immediately, we’ll see a screen with a bunch of
tildes along the left-hand side and some text containing the version and other
details:

121
T.me/nettrain
 How do we we actually start editing text? If you just start typing characters,
likely nothing will happen or you’ll get weird errors. Before doing anything, type
the i character on your keyboard. This will get you into insert mode. You can
verify this by looking in the bottom left corner. Now, you can type text like
normal. Go ahead and type a few sentences of whatever you’d like. Make sure to
type multiple lines.
You can use the arrow keys while in “insert” mode to move around and use
delete as normal. Once you’re happy with your text, hit the Escape key. The —-
INSERT -— text will disappear. To save the file, hit the : colon key on your
keyboard. You’ll see the colon show up in the bottom left corner. Type write and
then a name to call the file. Hit enter. You’ve saved the file!

122
T.me/nettrain
 Now, we’ll quit Vim. Hit Escape again and type the colon character and then
quit! Like so :quit!
The exclamation mark indicates we want to close Vim without saving. We just
saved, so that’s fine. Hit enter and we’re returned back to our familiar prompt
again. Well done!
Back at the terminal, feel free to cat or less your file to verify it was written.

Getting around Vim with HJKL

We’ll go back into our file again by typing vim <file_name>. We’re now in
default mode. By default, Vim lets you move through text files and move your
cursor with the keys HJKL. H moves left, L moves right, K moves up and J
moves down. This sounds really counterintuitive and hard, but once you’ve
learned it, you can move through Vim quickly. Give it a try and move around on
your text now. You can always use the arrow keys instead if you’d like.

123
T.me/nettrain
Quickly Moving to the Top, Bottom, Left and Right
Vim has a ton of shortcuts that let you edit text super quick. Let’s introduce a
few more. Put your cursor on the first character of a line of text. Now, type the $
symbol. You’ll see that the cursor has moved to the last character on the line.
Type the 0 key and notice that your cursor moves to the beginning of the line
again. Try moving back and forth a few times.
Let’s go the other direction now. Put your cursor anywhere on the first line of
text. Now, type capital G with Shift+G. If you have multiple lines of text, your
character will quickly shoot to the bottom of the text. Let’s go to the top again.
Type the lower-case g character twice. gg. One after another. Your cursor should
move back to the top of the text.
Like the HJKL keys, you might be thinking that this sounds insanely hard to
learn. It does take some practice and muscle memory, but it’s worth it in the
long-run. Try to stick with it.

Searching for text in Vim

If you’d like to quickly search for a word or series or words within Vim, hit
the / key and type your query. Then hit enter. Vim will jump to the first
occurrence it finds that matches your query. Hit lower-case n and Vim will jump
to the next place it finds the query. If you want to go back, hit upper-case N on
your keyboard.
If you’d rather search from the bottom to top, instead of hitting / and then
your query, instead hit the ? Key and type your query. The n key will move up
and N will move down to matches of the query.

Edit The File Again and Quit

I have one more task for you and then we’ll quit Vim. Select any line, go to the
end of it with the $ sign (or just move using the arrow keys or HJKL keys). Now,
hit the a key and you’ll see the —- INSERT —- text once again in the bottom left
corner. We’ve just entered insert mode again using the append command.
Append will let you add text right after the cursor. Type some text and when
you’re ready, hit Escape again.
Previously, we left Vim with the :write and :quit! commands. I didn’t tell you,

124
T.me/nettrain
 "
but there’s a short cut for both of these. Write can be abbreviated as w and quit
can be abbreviated as q. We can also combine them! You should be out of insert
mode now, type :wq and hit enter. You have just written and quit your file.

Continuing Vim Education

Vim is a lot to take in all at once. We’ve just barely scratched the surface. The
incantations of Vim can seem like a mystery to newcomers. The amazing part
about Vim is that you don’t need to know it all to get a lot of value. Learn Vim
slowly and incrementally and it’ll pay off. Below are some additional Vim
resources for new Shell Samurais. I haven’t personally used them all, but I have
hand-selected them all:

• vimtutor — This is command comes with Vim. Open your terminal, type
vimtutor, hit enter to accept the default location and follow along.
Vimtutor is interactive and right in your terminal. If you follow along,
you’re bound to learn enough about Vim to become a Vim ninja.
• Steps to Learn Vim - a short blog post with more Vim resources.
• Vim Adventures - an awesome Vim interactive javascript game. Browser
based.
• Vim Genius - This short interactive course guides beginners through basic
Vim shortcuts and commands in the browser. No Linux install needed.
Pipe -- Not Just for Plumbers

The pipe | operator allows us to connect the output or stdout of one

command to the input or stdin of another command.
Hot Tip : This probably sounds pretty similar to the redirect operator > we
looked at earlier. While both the pipe operator and redirect operator are used to
redirect output, they work in different ways. The > redirect operator redirects
output of a command to a file. The pipe | operator redirects output to the input
of another command. > for files. pipe | for command inputs. Easy.

Let’s look at an example now. We can ls a directory and we’ll get back a

125
T.me/nettrain
bunch of text. Try running ls on the /etc directory now. Then, let’s make that
text a little easier to read and pipe it to less. Run this:

root@linux_rocks:/# ls -l /etc | less

Now you can scroll through the files with the arrow keys (or j and k keys).
Remember, you can hit q to quit. We’ve just redirected the output of the ls
command to the input of the less command. Simple!
We can also use multiple pipe operators to keep chaining commands. We’ll
look at a little bit of an advanced use here, you might have not used all of these
commands yet, but stick with me:

stetson@linux_rocks:/# ps aux | sort -nk +4 | tail

USER PID %CPU %MEM VSZ RSS TTY STAT START TIME
COMMAND
root 786 0.0 0.0 14640 704 pts/0 S+ 22:14 0:00
sort -nk +4
root 787 0.0 0.0 1980 504 pts/0 S+ 22:14 0:00
tail
root 1 0.0 0.1 3744 2964 pts/0 Ss 13:10 0:00 /
bin/bash -
root 785 0.0 0.1 5472 2328 pts/0 R+ 22:14 0:00 ps
aux

Above, we use ps aux to list all running processes on our system. We then
pipe that output to the sort command. Sort is going to sort based on the 4th
column, which is memory usage. -k +4 says we want to sort on the 4th column,
while the n flag sorts in ascending order. Finally, we pipe the output to the tail
command, which will display the tail end of the input. Tail gives us the last 10
lines by default.
That was a pretty advanced use of pipe, so don’t feel intimidated if it was a bit
much to handle. The important thing to remember here is that we can use
multiple pipe operators. We could’ve kept chaining 10 or 50 pipes together. That
would’ve been a real doozy!
Oh, yeah, we can also use our redirect operator along with the pipe operator

126
T.me/nettrain
together. One more quick example and we’ll move on:

stetson@linux_rocks:/# ps aux | sort -nk +4 | tail > processes.txt

stetson@linux_rocks:/#
root@linux_rocks:/# cat processes.txt
USER PID %CPU %MEM VSZ RSS TTY STAT START TIME
COMMAND
root 792 0.0 0.0 14640 756 pts/0 S+ 22:19 0:00
sort -nk +4
root 793 0.0 0.0 1980 560 pts/0 S+ 22:19 0:00
tail
root 1 0.0 0.1 3744 2964 pts/0 Ss 13:10 0:00 /
bin/bash -
root 791 0.0 0.1 5472 2348 pts/0 R+ 22:19 0:00 ps
aux

This is the same example as above, only we’re sending the output to the
processes.txt file at the end.
Pipe is a powerful operator! By using the pipe command, the output of one
command becomes the input of the next command in the chain. This enables us
to create complex command pipelines that perform multiple tasks efficiently,
without the need for intermediate files or manual intervention.

Tee -- Not just for Old English Ladies and Golfers

The tee command is a pretty simple one, so I’ll be brief in this section. The tee
command allows us to duplicate output streams, directing to both the standard
output (stdout) and a specified file. Let's take a look at an example to understand
how it works:

stetson@linux_rocks:/tmp$ ls -l /etc/ | tee etc.txt

By executing this command, you will receive a directory listing of /etc on

your screen (stdout). Additionally, if you cat etc.txt, you’ll see the same text
there, too.
Without the tee command, simply running ls -l /etc > etc.txt would

127
T.me/nettrain
have only output to the file with no output on your screen. Tee offers an easy
way to view and store command output, which can be helpful for
troubleshooting and documentation purposes.

Grep - I will find you and I will grep you

Grep is an essential tool for Shell Samurais to process and search for text. With
grep, we can search files and text for certain words or characters that match a
pattern we specify.
Let’s make a file for ourselves to use grep on:

stetson@linux_rocks:~$ printf "apples\nbananas\ncarrots\noranges\n”

> foods.txt

We’re using printf here instead of echo because printf will interpret our \n
characters as a new line. Now, let’s search the file using grep:

stetson@linux_rocks:~$ grep carrots foods.txt

carrots

Grep quickly located the word in the file! If we searched for a word that didn’t
exist, we’d get no results back:

stetson@linux_rocks:~$ grep tomatoes foods.txt

stetson@linux_rocks:~$

There are a few other uses of grep that are very helpful.

We can use grep to search multiple files:

grep "search-pattern" file1.txt file2.txt file3.txt

We can grep for all files in a directory:

grep "search-pattern" file1.txt file2.txt file3.txt

128
T.me/nettrain
 We can grep recursively in directories, so that we check directories within the
directory we specify:

grep -r "search-pattern" /path/to/directory/

We can even grep and ignore case sensitivity on our earlier example:

stetson@linux_rocks:~$ grep -i CaRrOts foods.txt

carrots

Lastly, we can pipe output to grep and search that! Here, I grep for security in
the /etc directory and find the security directory.

stetson@linux_rocks:~$ ls -l /etc | grep security

drwxr-xr-x 4 root root 4096 Sep 21 2021 security

Grep is great to filter text, search for text strings and more! Grep even
supports regex, so we can find almost any text we want. Learn grep and become
an effective Shell Samurai.

Regular Expressions with Grep

Regular Expressions (often abbreviated as regex) are an incredibly powerful

tool for finding and manipulating text. In this section, we’ll use grep with regex
to search through text. Regular expressions are widely used in IT and software
development, featuring support in numerous programming languages, text
editors, and utilities like awk and sed (which we’ll cover in a later section).
Enough theory about regex, let’s dive into some examples. Create a file called
my_regex.txt with any text editor and add the following content (or similar text
spanning a few lines, ensuring there are no spaces at the end of each line):

Free like the bird's wings

Linux soars with open-source
Freedom to explore

129
T.me/nettrain
 - my_regex.txt

First we’ll match the beginning of a line with the ^ character:

stetson@linux_rocks:~$ grep ^Freedom my_regex.txt

Freedom to explore.

Next, let’s match the end of a line with the $ character:

stetson@linux_rocks:~$ grep wings$ my_regex.txt

Free like the bird's wings

With a period character, we’ll match a single character following an L:

stetson@linux_rocks:~$ grep L. my_regex.txt

Linux soars with open-source

We could match any number of characters with multiple periods. Here we

look for an “o” followed by any 4 characters:

stetson@linux_rocks:~$ grep o.... my_regex.txt

Linux soars with open-source
Freedom to explore

Let’s try a some more advanced regex. Create a file and add the following
content and call it words.txt:

babble
bubble
bobble

We can now use [] brackets to search for characters that match inside. Here’s
an example:

stetson@linux_rocks:~$ grep b[au]bble words.txt

babble
bubble

130
T.me/nettrain
 Our regex returned babble and bumble because we specified the a and u
characters. It does not return bobble since an “o” is not in the brackets.
We could use the ^ character inside our brackets to search for anything that
does NOT contain that character.
Let’s look for b[^u]bble now:

stetson@linux_rocks:~$ grep b[^u]bble words.txt

babble
bobble

We get babble and bobble, but NOT bubble, since we told grep we don’t want
the u character to be returned.

This is a very brief overview of Regular Expressions. I encourage you to go

read up on regular expressions because they can be super powerful. Regular
Expressions aren’t something that you’ll always know off the top of your head,
but will most likely Google or research while trying to search for a specific string.
I can’t count the number of times where I’ve had to Google something like
“Regex search for all lines containing 192” or the like.
Regular Expressions will be a powerful tool on your Linux Journey, and this
section should serve as a foundation for future learning.

cut
Cut is a text processing utility we can use to very quickly cut or chop text up
and output it to stdout. Let’s dive in to some examples now. Write the following
command in your shell:

stetson@linux_rocks:/tmp$ echo -e "This is my shell\nThere are many

131
T.me/nettrain
like it\nBut this one is mine" > shell_creed.txt

Note, the -e flag tells echo to listen to backslash escapes, like the \n one we’ve
used to specify a new line. Then, we redirected the output to a new file called
shell_creed.txt.

Next, we’ll use cut to cut the file:

stetson@linux_rocks:/tmp$ cut -c 1 shell_creed.txt

T
T
B

The c flag tells cut to output the first character of each line. Go ahead and try
changing the -c flag’s argument now:

stetson@linux_rocks:/tmp$ cut -c 4 shell_creed.txt

s
r
\s

Above, I’ve marked the space character with \s to show that cut will interpret
spaces and output them! The 4th character of the bottom line happened to be a
space.
What if we wanted the second word from every line? We can use the -f or
field flag with cut to ask for the second word in each line:

stetson@linux_rocks:/tmp$ cut -f 2 shell_creed.txt

This is my shell
There are many like it
But this one is mine

Wait! That didn’t work. We just got back the whole file. Why? By default, cut
is set to look for TAB delimiters, so we’ll have to tell it that we want it to use
spaces instead. Do so now with the -d or delimiter flag:

stetson@linux_rocks:/tmp$ cut -f 2 -d ' ' shell_creed.txt

is
are

132
T.me/nettrain
 this

Much better!

As you can see, cut is another text processing utility we can add to our tool
belt. As with any Linux shell tool, it becomes more powerful when it’s combined
with other commands via the redirect (>) or pipe operators (|).

Counting Lines with Word Count

At some point, you may find that you want to count the number of lines in a
file. It’s simple to do so, just use the wc or Word Count command. Without any
arguments, wc returns the word count, line count and byte count.

Here’s an example:

stetson@linux_rocks:/tmp$ wc file.txt
8059 8059 32236 file.txt

If you only want the lines in a file, you can use the -l flag:

stetson@linux_rocks:/tmp$ wc -l file.txt
8059 file.txt

Another common way to use wc is to pipe the output of cat to wc:

stetson@linux_rocks:/tmp$ cat file.txt | wc -l

8059

Easy! Let’s move on.

sed and awk

Sed and awk are two command-line utilities that are commonly used together
for editing text. They’re most often used for analyzing large text files or editing
text files quickly. For example, Sed can replace every instance of a word in a large
text file.

133
T.me/nettrain
 "
sed
Sed stands for stream editor and at this point, it’s become an ancient and
powerful tool. Sed was initially developed in 1971 at Bell Labs, just like Unix. Sed
is actually considered a programming language. It’s just that powerful and even
considered turing-complete.
Let’s take a brief side quest to learn about Turing completeness.

Hot Tip : Turing Complete is a computer science term meaning that a system
can perform any computation that a Turing machine can do. Wait a second. What
is a Turing machine? A Turing machine is a hypothetical, imaginary computer
that would work by writing to and reading from a tape. The machine can move
left or right along a tape and perform operations on values it reads from each cell
on the tape.
A Turing Machine is simple but can perform any mathematical operation that
a modern computer can. If a programming language is Turing complete, this
means it can do any operation that any other Turing complete language can do.
This concept is named after the British mathematician Alan Turing, who
introduced the idea of the Turing machine as a way to understand the limits of
computation. I’d recommend reading more about Turing Machines if you’re
interested in the history of computing!
Sed actually works in a simple way. It reads text line by line into a buffer and
performs instructions on each line. For example, sed could read a whole book as
text and replace every instance of the with samurai. Let’s take a look at our first
sed command and instructions.

stetson@linux_rocks:~$ echo "This is my sword" | sed s/sword/shell/

This is my shell

Here, we’ve used sed with some simple instructions to replace instances of
sword with shell. The text after sed is the sed command. S stands for substitute,
where sword is the text we want to look for and shell is what we want to replace.
Let’s look at a few other examples. Create a new text file with your favorite
editor and write the following text, separated by new lines. Save it as

134
T.me/nettrain
shell_creed.txt:

This is my sword
There are many like it
But this one is mine

We already looked at replacing text. We’ll do the same thing but modify our
file in the process:

stetson@linux_rocks:/tmp$ sed -i ’s/sword/shell/g’ shell_creed.txt

stetson@linux_rocks:/tmp$ cat shell_creed.txt
This is my shell
There are many like it
But this one is mine

Let me explain. The -i flag stands for in-place, it tells sed to modify the file
and save the changes to the original file.
Again, we are substituting with s and specifying the text to search for and the
text to replace. This time, though, we also give it the g flag to tell sed to perform
the substitution globally.
Let’s try another sed command to instead delete any lines containing shell:

stetson@linux_rocks:/tmp$ sed -i '/shell/d' shell_creed.txt

stetson@linux_rocks:/tmp$ cat shell_creed.txt
There are many like it
But this one is mine

This time, we’re still calling sed with the -i or in-place flag, but the d
command we give sed tells it to delete every line containing shell.
Sed is very powerful, but becomes more powerful when we combine it with
awk.

AWK
AWK is also considered a programming language. Where sed is often used to
modify and process text, awk is used to analyze text. AWK was also developed by
Bell Labs in the 70’s and carries ancient and powerful history with it. Awk’s
name comes from it’s developers, Alfred Aho, Peter Weinberger and Brian

135
T.me/nettrain
Kernighan.
Like sed, AWK reads text, line by line and scans each line for a user-specified
pattern. If it finds a match, it performs an action.
Let’s take a quick look at an awk command. Here, I’ll use awk to print only the
9th column from the output of ls -lh:

stetson@linux_rocks:~$ ls -lh
total 16K
-rw-rw-r-- 1 stetson stetson 10K Feb 27 00:28 important_file
-rw-rw-r-- 1 stetson stetson 0 Feb 27 00:27
moon_landing_photos.png
-rw-rw-r-- 1 stetson stetson 209 Feb 27 00:44 myArchive.tar.gz
-rw-rw-r-- 1 stetson stetson 0 Feb 27 00:27 secret_documents

stetson@linux_rocks:~$
stetson@linux_rocks:~$ ls -lh | awk '{print $9}'

important_file
moon_landing_photos.png
myArchive.tar.gz
secret_documents

Above, I feed the output of ls -lh to awk with the pipe operator. Then, I use
the awk action ‘{print $9}’ to tell awk that I only want the 9th column of each
line, where each column is separated by whitespace.
AWK can also perform a few other tricks. Suppose you have a file with a list
of numbers called numbers.txt:

10
20
30
40
50

To calculate the sum, use AWK as follows:

awk '{sum += $1} END {print sum}' numbers.txt

output:

136
T.me/nettrain
 150

AWK can also calculate averages of the same file:

awk '{sum += $1; count++} END {print sum / count}' numbers.txt

output:

25

All Together Now

I mentioned that awk and sed were a powerhouse when combined. Let’s look
at an example.
We can pipe the output of one command to the other to perform complex text
processing tasks.

With a file called inventory.txt:

item001, Laptop, 5
item002, Mouse, 12
item003, Keyboard, 8
item004, Monitor, 6

Let's say you want to replace all commas with tabs and then print only the
lines with a quantity greater than 10. You can achieve this using sed and AWK
together as follows:

sed 's/,/\t/g' inventory.txt | awk '$3 > 10 {print}'

In this example, the sed command replaces all commas with tabs, and then
the output is piped to the awk command. The awk command checks if the third
field (quantity) is greater than 10 and, if so, prints the entire line.

root@linux_rocks:/tmp# sed 's/,/\t/g' inventory.txt | awk '$3 > 10

{print}'
item002 Mouse 12

137
T.me/nettrain
 This has been a brief intro to using sed and awk. Combining the two tools
creates a powerhouse combo for any Shell Samurai to automate away painful
processes. Whether it’s processing web server logs or analyzing transaction
records, sed and awk should be in your tool belt.

Uniq and Sort

Continuing our discussion of text processing, let’s look at another two useful
tools together, sort and uniq. I’ve found myself often using these in tandem, so
that’s why we’re covering them that way!

First, sort. sort…sorts lines. Given a file fruits.txt which contains the
contents:

carrot
banana
apple
orange
kiwi
tomato

We can run the above text file through sort to get the list returned back in
alphabetical order:

stetson@linux_rocks:~$ sort fruits.txt

apple
banana
carrot
kiwi
orange
tomato

Provide the -r flag and we’ll get the list back in reverse alphabetical order:

stetson@linux_rocks:~$ sort -r fruits.txt

tomato
orange

138
T.me/nettrain
 kiwi
carrot
banana
apple

Now, let’s pretend that our file has duplicate lines we want to eliminate. Let’s
not pretend, let’s put a few duplicates in:

carrot
banana
apple
tomato
apple
orange
orange
kiwi
tomato
carrot
kiwi
carrot

We can now use uniq to remove duplicate lines:

stetson@linux_rocks:~$ uniq fruits.txt

carrot
banana
apple
tomato
apple
orange
kiwi
tomato
carrot
kiwi
carrot

Wait a minute! That didn’t remove the duplicate lines at all! Or maybe it did…
take a closer look. Uniq, by default, will only remove duplicated lines that are
next to each other. Our repeated lines containing orange were “condensed” to a
single line. However, we still have two carrots and two tomatoes.
That’s where sort comes in. We’ll run sort on the file first and then pipe the

139
T.me/nettrain
output to uniq:

stetson@linux_rocks:~$ sort fruits.txt | uniq

apple
banana
carrot
kiwi
orange
tomato

Much better! sort first made our file alphabetically ordered, and then we
piped the output to uniq, which got rid of the duplicate lines!
If we wanted only the unique lines, we can get those by providing the -u flag:

stetson@linux_rocks:~$ sort fruits.txt | uniq -u

banana

And if we want to get only the lines that are duplicates, we can give uniq the
-d flag:

stetson@linux_rocks:~$ sort fruits.txt | uniq -d

apple
carrot
kiwi
orange
tomato

uniq and sort are indeed two more powerful tool for processing text. Like
other Linux utilities, they become even more powerful when we utilize the pipe
operator | to send their output to other commands. Are you beginning to feel the
power of Linux’s modular design? One tool does one thing very well and has an
interface to talk to other tools.

Making Heads and Tails of it

Making heads and tails of Head and Tail

In your Shell Samurai journey, you’ll often have to read longer files with
many, many lines like logs. Instead of dumping the entire file to your terminal

140
T.me/nettrain
and scrolling through, head allows you to instead read just the first 10 lines:

stetson@linux_rocks:~$ head /var/log/bootstrap.log

2021-09-21 16:51:01 URL:http://ftpmaster.internal/ubuntu/dists/

focal/InRelease [264892/264892] -> "/build/chroot/var/lib/apt/lists/
partial/ftpmaster.internal_ubuntu_dists_focal_InRelease" [1]
gpgv: Signature made Thu Apr 23 17:34:17 2020 UTC
gpgv: using RSA key 3B4FE6ACC0B21F32
gpgv: Good signature from "Ubuntu Archive Automatic Signing Key
(2012) <[email protected]>"
gpgv: Signature made Thu Apr 23 17:34:17 2020 UTC
gpgv: using RSA key 871920D1991BC93C
gpgv: Good signature from "Ubuntu Archive Automatic Signing Key
(2018) <[email protected]>"
2021-09-21 16:51:07 URL:http://ftpmaster.internal/ubuntu/dists/
focal/main/binary-arm64/by-hash/SHA256/
e014b241021d8c0cf16cb94f44f4cafb7746507769baf36b77845d3ae22e2c1b
[940188/940188] -> "/build/chroot/var/lib/apt/lists/partial/
ftpmaster.internal_ubuntu_dists_focal_main_binary-
arm64_Packages.xz" [1]
2021-09-21 16:51:09 URL:http://ftpmaster.internal/ubuntu/dists/
focal/restricted/binary-arm64/by-hash/
SHA256/34a9d195f086d660d5d3768695c30d6be11b4f5f13656d014e84b30c16cc05e
f [1300/1300] -> "/build/chroot/var/lib/apt/lists/partial/
ftpmaster.internal_ubuntu_dists_focal_restricted_binary-
arm64_Packages.xz" [1]
2021-09-21 16:51:12 URL:http://ftpmaster.internal/ubuntu/pool/main/
a/adduser/adduser_3.118ubuntu2_all.deb [162792/162792] -> "/build/
chroot//var/cache/apt/archives/partial/
adduser_3.118ubuntu2_all.deb" [1]

If instead you’d like to adjust the amount of lines returned, we can give head
the -n flag and the number of lines we’d like to get back. Let’s try just three lines
of output:

stetson@linux_rocks:~$ head -n 3 /var/log/bootstrap.log

2021-09-21 16:51:01 URL:http://ftpmaster.internal/ubuntu/dists/

focal/InRelease [264892/264892] -> "/build/chroot/var/lib/apt/lists/
partial/ftpmaster.internal_ubuntu_dists_focal_InRelease" [1]
gpgv: Signature made Thu Apr 23 17:34:17 2020 UTC
gpgv: using RSA key 3B4FE6ACC0B21F32

141
T.me/nettrain
 What’s the opposite of head? Tail! Tail will also give us the last 10 lines of a
given file.

stetson@linux_rocks:~$ tail /var/log/bootstrap.log

Setting up libp11-kit0:arm64 (0.23.20-1build1) ...

Setting up libnettle7:arm64 (3.5.1+really3.5.1-2) ...
Setting up gpgv (2.2.19-3ubuntu2) ...
Setting up libtasn1-6:arm64 (4.16.0-2) ...
Setting up libstdc++6:arm64 (10-20200411-0ubuntu1) ...
Setting up libhogweed5:arm64 (3.5.1+really3.5.1-2) ...
Setting up libgnutls30:arm64 (3.6.13-2ubuntu1) ...
Setting up libapt-pkg6.0:arm64 (2.0.2) ...
Setting up apt (2.0.2) ...
Processing triggers for libc-bin (2.31-0ubuntu9) ...

Again, if we’d like to specify the amount of lines returned, we can use the -n
flag and specify the number of lines we get back as an argument.

stetson@linux_rocks:~$ tail -n 3 /var/log/bootstrap.log

Setting up libapt-pkg6.0:arm64 (2.0.2) ...

Setting up apt (2.0.2) ...
Processing triggers for libc-bin (2.31-0ubuntu9) ...

That’s head and tail! Simple, huh?

Spying on syscalls with strace

Strace is an advanced debugging tool, but you should definitely still know
about it! We already talked about system calls and how processes on Linux use
syscalls to talk to kernel space and get things done. It’s a little bit abstract to
understand by itself. Wouldn’t it be nice if there was a way to “see” system calls
as they happen? Enter strace.
Strace is a syscall tracer. It allows us to spy on system calls that a command or
process uses. Try running strace ls in your terminal now and you’ll get back a

142
T.me/nettrain
 "
ton of alien output. Don’t be scared, you don’t need to understand everything
returned right now.

Let’s try to take a look at the first line that comes back from strace:

execve("/usr/bin/ls", ["ls"], 0xffffeee36aa8 /* 9 vars */) = 0

The first line we see shows execve, which is the execute program system call
and it’s what executes the program ls.
All the text between the parentheses are the arguments that are passed to the
execve system call. We can see in the arguments the path of the ls program at /
usr/bin/ls
The last section, = 0, is the return value. 0 is considered a successful return
value and tells the system that the execve system call worked properly.
Feel free to read through the rest of the strace output and see if you can find
out more on your own. Each line you see is a system call and its arguments. If
you don’t know what a specific system call does, google it!
Strace can give you superpowers for spying on programs and processes. My
intent isn’t to go deep with it, but to make you aware of strace and how it can be
used to better understand the Linux operating system and debug issues.

Hot Tip : You should almost never use strace on a production web server or
database. It is very resource-intensive and can grind your application to a halt. Be
very certain and careful before using strace in production.

Finding Files with Find

find is a command-line tool that can help us search the vast directory
structure of Linux for specified directories or files based on criteria such as name,
size, type, or modification time. Since Linux directories are so sprawling, it just
doesn’t make sense to go about manually running ls on directories, reading and
grepping the output until you find the one you’re after.
Find generally takes two arguments, a directory to search through and a file or
folder name we’re looking for.

143
T.me/nettrain
 The basic syntax of the find command is as follows:

find [path] [expression]

The [path] is the starting directory where the search will begin, and
[expression] is a set of options and tests that define the search criteria. For
example, to find all files named "file.txt" in the current directory and its
subdirectories, you would use the following command:

stetson@linux_rocks:~$ find . -name 'file.txt'

./file.txt

Try creating a file in your home directory and then using find to find it now.

Find doesn’t just make us use the file name as a search parameter. For
instance, we can search for files based on their size using the -size option
followed by a size value and unit (c for bytes, k for kilobytes, M for megabytes,
and so on). The following command will find all files larger than 10 megabytes in
the / directory (you may need to run this as root or sudo):

root@linux_rocks:/home# find / -size +10M

Find doesn’t just search one directory deep, we can search from the root
directory for any file:

stetson@linux_rocks:~$ sudo find / -name secret_file.txt

/home/stetson/secret_file.txt

Another useful feature of find is its ability to execute actions on the files that
match our search criteria. By using the -exec option, you can specify a command
to run on each file that was found.
The following example will find all files with the ".log" extension in the /var/
log directory and delete them:

find /var/log -name "*.log" -exec rm {} \;

144
T.me/nettrain
 Above, {} is replaced with the file path of each matching file, and \; signifies
the end of the -exec command.

In conclusion, find is a valuable asset for navigating the Linux filesystem and
performing operations on files and directories. By mastering find’s options,
you’ll be much better equipped to handle file management tasks efficiently and
effectively. Remember to consult the man pages (man find) for more information
and examples of the find command's extensive capabilities.

Swap Space
Swap space is not a unique topic to Linux, but it is fundamental to
understand operating systems in general and Linux is no different! Swap space is
space on a hard disk that can be substituted for physical memory or RAM. It’s a
kind of virtual memory used like an “overflow” for RAM. When RAM runs out
of space, swap space is used instead.
It’s generally not considered a good thing if your system is frequently using
swap space rather than RAM. Hard disk storage is much slower to access than
RAM, so if our system is using it all of the time, it’s going to operate more
slowly!
You can use the swapon command to show where your swap space is
currently configured and the configured size:

stetson@linux_rocks:/var/log$ swapon
NAME TYPE SIZE USED PRIO
/swap file 1024M 0B -2

Here, my swap space is located at /swap and is 1024 MB. If we’d instead like
to see how much swap is being currently used, we can use the same command
that we’d use to check memory statistics, free -h:

stetson@linux_rocks:/var/log$ free -h
total used free shared buff/cache
available
Mem: 1.9Gi 388Mi 706Mi 386Mi 894Mi

145
T.me/nettrain
 "
1.1Gi
Swap: 1.0Gi 0B 1.0Gi

Check the second row, you’ll see that I have 1 Gig of swap free with nothing
used (second column).

Scheduling tasks with cron

I estimate that much of the world’s economy and modern technology are held
together by cron. I’m serious. It’s everywhere. What is cron? Cron is a time-based
scheduler built-in to Linux. It’s available on nearly every system by default. Cron
is not a command, but a system-managed process and daemon that runs in the
background to schedule tasks.
Cron allows administrators to define schedules and tasks in a file known as a
“cron job”.
As an example, a cron job might do any of the following.

• Email users every week with a report of their usage of a Software license
• Backup a database every night at midnight
• Delete old log entries every Wednesday at 12 PM UTC
• Send a report of the system’s memory and disk-usage every single minute
to a monitoring service

As you can see above, cron jobs always define a schedule and a task to
perform. The Cron table, or crontab is a configuration file that defines each cron
job. Crontab files are created for each user. Cron jobs will run with the same
permissions of the user that created them.

Logged in as any user, you can quickly display the crontab file with crontab
-l

Hot Tip : In the rare event, cron is not available on your system, you may
need to install it! Run apt-get update && apt-get upgrade and then apt-get

146
T.me/nettrain
install cron.

Once you list a crontab output, you should see something like the following:

0 0 * * * /path/to/backup-script.sh

This looks a bit alien. Let me explain. The first five fields, 0 0 * * *, specify
minutes, hours, days, months and day of the week. The asterisk is a wildcard for
any. The above example tells cron to run the backup-script.sh at minute 0 and
hour 0 (midnight) every day of the month, every month and every day of the
week.
Let’s try making our own crontab entry. Type the command crontab -e and
hit enter. The e flag is for edit. Cron should then either throw you directly into
an editor or ask you to select a text editor. You may get vim as your default text
editor. Use vim if you’re comfortable with it (be sure to review the vim chapter in
this book). If you’re not comfortable with vim, exit out of it by hitting Escape and
typing :q! and enter. Type select-editor and you can freely choose a different
editor, like nano:

stetson@linux_rocks:~$ select-editor

Select an editor. To change later, run 'select-editor'.

1. /bin/nano <---- easiest
2. /usr/bin/vim.basic
3. /usr/bin/vim.tiny
4. /bin/ed

Choose 1-4 [1]:

With our text editor selection out of the way, let’s edit our crontab file!

On a new line, type the following contents and save the file:

*/2 * * * * echo `date` >> /tmp/current_date.txt

This crontab entry is going to run every 2 minutes (*/2 ) for as long as our

147
T.me/nettrain
 "
system is up and print the current date out to a file at /tmp/current_date.txt.
After saving the crontab, you should see a message confirming that the new
crontab was installed:

stetson@linux_rocks:/tmp$ crontab -e
no crontab for stetson - using an empty one
crontab: installing new crontab

Now let’s wait for cron to run our command. After about 2 minutes have
passed, run cat /tmp/current_date.txt to verify the cron job ran:

stetson@linux_rocks:~$ cat /tmp/current_date.txt

Wed Feb 22 03:33:07 UTC 2023

Keep in mind that we could have given cron any shell script or file to run, but
we chose to instead directly configure our command in the crontab file.
Now, let’s get rid of the cron entry so that our disk doesn’t fill up. Open the
crontab once again with crontab -e, delete the line we just created and save the
file. Without doing this, the cronjob would run for as long as our system is up.

Congrats on your first cron job!

See if you can make more cron entries of your own!

Hot Tip : Cron syntax can be very confusing. If you asked me to create a
cron job that ran on every third Wednesday at 3:45 PM, I couldn’t do it. I would
guess that most Linux systems administrators couldn’t either! That’s alright,
though, because we have tools to help us! There are many “cron calculator” web
apps that help users translate from English to crontab. Simply Google “cron
calculator” or use a site like CronTab.Guru or CronHub. These tools help us
configure Cron schedules like a true Shell Samurai!
Although these examples were simple, cron is a powerful and fundamental
Shell Samurai skill. Many, many business operations run on cron today. Why
should we do manual work when we can get machines and robots to do it for us?
Cron can be helpful for automating processes like backing up files, sending

148
T.me/nettrain
emails and even more. Use cron and become productive and dangerous.

Secure Shell or SSH

No Linux administration book would be complete without covering SSH. SSH

or Secure Shell is the standard way in which you’ll remotely login to Linux
systems. SSH doesn’t only allow you to access just a shell, it’s also capable of file
transfer, port forwarding and other network proxy services. Data passed over
SSH is secure because the connection is encrypted and authenticated.
OpenSSH is open source and included on almost every version of Linux.
There are a few other implementations of SSH, but you’re best off using
OpenSSH for now.

A Brief Primer on Public Key Cryptography

SSH uses public key cryptography or asymmetric cryptography for authentication
and encryption. Public key cryptography allows a client (you!) and a remote
server to communicate securely without sharing a secret key.
Traditionally, encryption would work by using the same secret key to encrypt
and decrypt data. This means that both client and server would need access to
the same key. If anyone intercepted that key, they could also decrypt the data.
Not good!
With public key cryptography, there are two keys involved: a public key and a
private key. The public key is NOT a secret and is free to be shared. The Public
Key is used to encrypt data before sending. The Private Key on the other hand, is
a secret. You might have guessed that based on the name Private Key. Private keys
are used to decrypt data that has been received.
When a client and server communicate, they are using two pairs of Public and
Private Keys. For example, when a server wants to send data to the client, it uses
the client’s public key to encrypt the data and send it. Once the data is received
by the client, it’s decrypted with the private key. The process is reversed for the
other direction. The client sends data encrypted with the server’s public key. The
server then decrypts the data with its own private key.

149
T.me/nettrain
Installing SSH and Connecting
With that bit of theory out of the way, let’s setup SSH from scratch and
connect to our server!
First, install OpenSSH with sudo apt-get install openssh-server if you
don’t already have it. You can check by running which sshd and checking if you
get a path back like /usr/sbin/sshd.
Once you have OpenSSH installed, you’ll need to configure it. Using sudo or
the root user, open up the config file for sshd at /etc/ssh/sshd_config with
any text editor. Lines that begin with a # are comments and will not be applied.
Search for and uncomment the following lines and set the values to below. If a
line doesn’t exist already, be sure to add it.

Port 22
ListenAddress 0.0.0.0
PermitRootLogin no
PubkeyAuthentication yes
PasswordAuthentication yes

• Port - This is the address the SSH server will listen on

• ListenAddress - This is the address the server will listen on. 0.0.0.0 will
listen on all available network interfaces
• PermitRootLogin - We set this to no so that we can disable logins as root.
This is a best practice for security.
• PubkeyAuthentication - We want to use public keys, so we set this to yes!
• PasswordAuthentication - We turn off password authentication because
we’re going to be using keys!

After saving the sshd configuration, go ahead and restart the SSH server with
systemctl restart sshd

Generating an SSH Key Pair

Next, we need to generate a key pair! No matter what operating system
you’re working on, you should have ssh installed and available as a command.
Windows has ssh built into Command Prompt and Powershell, Mac has it in the
terminal and of course, if you’re on Linux, you have the ssh command available.

150
T.me/nettrain
 From your local system, run ssh-keygen with no arguments. There are
additional flags available for cryptographic algorithm, key-length and other
settings. We’ll stick with the defaults for now. ssh-keygen will also ask for a
password to protect the private key. You can hit enter to skip the password, but
you might want to provide one on production systems for additional security!

Here’s what a ssh-keygen output should look like:

stetson@linux_rocks:/$ ssh-keygen
Generating public/private rsa key pair.
Enter file in which to save the key (/home/stetson/.ssh/id_rsa):
Created directory '/home/stetson/.ssh'.
Enter passphrase (empty for no passphrase):
Enter same passphrase again:
Your identification has been saved in /home/stetson/.ssh/id_rsa
Your public key has been saved in /home/stetson/.ssh/id_rsa.pub
The key fingerprint is:
SHA256:hbT1XTCvr5cZ3JLEvjdeawG/QS+kgKtMfyupJ2T52lQ
stetson@linux_rocks
The key's randomart image is:
+---[RSA 3072]----+
| . . o..|
| . + . . + |
| o.. ... .|
| ... .+o |
| .S.E. =*.o|
| = .. . =Bo|
| = +.. +X|
| +o* . .B=|
| o=.o.. +=.|
+----[SHA256]-----+

Check out the randomart image at the end. This a visual fingerprint of the
public key, it’s an additional, security feature that can be used to verify public
keys by quickly looking at the fingerprint.

Move the public key to the server

The last step before connecting to our server is to upload our public key to the
server. This could be a ”chicken and egg” scenario if you don’t already have

151
T.me/nettrain
access to the server you want to transfer your key to. I’ll assume you do, so you
can go ahead and cat your public key (not private!) out and copy it to your
clipboard.

stetson@linux_rocks:/$ cat /home/stetson/.ssh/id_rsa.pub

stetson@linux_rocks:/$
ssh-rsa AAAAB3NzaC1yc2EAAAADAQABAAABgQDEvm9NET90wmDTHYTGF/
I4amzem4h0pkRNvJfust8Zh0uwvseWAB0sl8fUZw7TbH1kVI/rwF6Jd5PYe/
srU6VhcChjV9/
oxM9G6hOkhUXyp6A+X17vBQmWEq3ffquvUWdCEVS1pt+x13VXbXSwjY64awQUWW/
UERz0oRBQ97mG85AV334+08DGiMB3JkmOCjjI0YZfEmC9Krc15jYJLjPfB6QYTFyA4tgap
49kU59BSpc0zYvvSXHP50i4usKjV7BlXBm7ay2bTdhZzL9Y49Mi2+3bSUXDMkBuXSBISCx
2o2DNYRTpizdJ9dmyRxy97bJKGJHGu0ZHLESuzpj7Oz4Mcq86hhwAad1m1T2Gl4VrPkaAF
SQmDLAVXY3501EiipT7JHkFWKBNM4wuywP2DesRhh8vsPJWgI8GrYC83ZH4JxshE/
IDSTbtJ7X4A5UAI87BHP1iJtZlwNH/
i08HYkbYaoCvYHVRwlwNDY249SHdybhcojhCVo4bJbuKmlTUILE=
stetson@linux_rocks

The ssh-rsa at the beginning of the file specifies the algorithm used to
generate the key. The stetson@linux_rocks at the end of the file is a comment
that was auto-generated. We could put anything here.
On your remote server running OpenSSH, we’ll paste this key into our user’s
authorized keys file at /home/<user_name>/.ssh/authorized_keys. Create this
file if it does not exist already.
Finally, we’re ready to connect!
From your client, you can run ssh <user_name>@IP_Address — If all went
well, you’ll be asked to verify the public key is one you recognize:

The authenticity of host '198.51.100.2 (198.51.100.2)' can't be

established.
RSA key fingerprint is
ef:ef:c8:e7:88:bc:94:95:c2:77:11:bb:e7:7d:33:dd.
Are you sure you want to continue connecting (yes/no)? yes

After entering yes, you should be logged in to the remote server! If your local
ssh agent is not using your local private key for some reason, you can specify it
with the -i flag explicitly and provide the path:

152
T.me/nettrain
 stetson@linux_rocks:/$ ssh -i ~/.ssh/id_rsa.pem [email protected]

This has been a very brief primer on SSH! SSH has a few other tricks up its
sleeves, like copying files with Secure Copy (scp), network tunneling and
proxying and more!
Knowing how to use SSH is an absolute requirement for any systems
administrator and Shell Samurai. It takes a little time to come up to speed with
and I don’t know all of the intricacies myself, even! Be sure to reference online
documentation, man pages and other resources if you need help!

CURL - Not just for Olympians and Hair Stylists

curl or client URL is a tool for transferring data from the web. We can use
curl to download data or call HTTP APIs. Curl’s webpage boasts that it’s used in
over ten billion places from routers, cars, phones, tablets and audio equipment to
virtually any other digital device you can think of. Curl comes with just about
every Linux installation by default.
We’ll use curl in this section to learn about APIs.
Curl allows us to specify a URL and data we’d like to send to that URL. Let’s
make sure we have curl installed with sudo apt-get install curl, although it
should be on your system by default.
Curl’s general syntax is:

curl <url> [options]

Let’s try using curl now with https://jsonplaceholder.typicode.com/, a

free website used for testing APIs. We’re going to perform a GET operation with
curl to retrieve data from the Json placeholder API.
In your terminal, run:

curl https://jsonplaceholder.typicode.com/todos/1

153
T.me/nettrain
 stetson@linux_rocks:/$ curl https://jsonplaceholder.typicode.com/
todos/1
{
"userId": 1,
"id": 1,
"title": "delectus aut autem",
"completed": false
}stetson@linux_rocks:/$

We’ll immediately get back a response, followed by our prompt. The response
doesn’t have a newline at the end, which is why the prompt got tacked on to the
end.
This API response came back to us in JSON or Javascript Object Notation
format. It’s an extremely common format for the modern web, although lately
some websites and applications are using the newer GraphQL instead.
You’ll notice the response has keys like userId, id, title and completed.
Each key has a value like 1, 1, delectus aut autem and false.

A Brief API primer

If you’re not already familiar with an API, this section will act as a primer.
APIs are the backbone of the internet. Nearly every website you use daily has an
API working in the background to transfer data and deliver a seamless web
experience.
API stands for Application Programming Interface. In short, an API is a HTTP or
HTTPS URL that computers send data to and receive data back from. Clients
might send data to APIs that they want updated, which the API will request to be
updated on the back-end and potentially save it in a database.
When API or web requests are sent, they also specify a method. A method is a
verb that loosely describes what the API call is requesting to do.

There are 5 HTTP Methods:

• GET - Requests data back from the server to be read

• POST - Submits data to the server and requests that it be written
• PUT - Updates a specified resource on the server by replacing it with a

154
T.me/nettrain
 new representation of that data.
• PATCH - Applies modifications to a resource
• DELETE - Deletes the resource from the server

CRUD
You’ll also hear the term CRUD when describing web applications and in
discussions about APIs. CRUD stands for Create, Read, Update and Delete.
These are the 4 operations used in databases and web apps to transform data and
is a model used to describe web apps. Each of our 5 HTTP methods lines up with
Create, Read, Update or Delete.
For example, GET is a read operation. POST is an Update operation. PUT and
PATCH update data and of course DELETE…DELETES data.
With the CRUD design, we can create web apps that perform almost business
goal.

What’s happening when we send an API request?

It’s important to understand the 5 http methods and the concept of CRUD
because it’s how most modern web applications work. What is actually
happening when you send a request marked as a GET, POST, PUT, PATCH or
DELETE?
First, the server has a function or method that listens for your request and
determines what type of request it is. Then, it looks to see if there is a resource
that’s specified. This is usually an id specified as a number or integer. There may
be some additional data attached to the request that the server parses. Once the
server has parsed all the data it’s been sent, it will then request that some code be
ran to get the resource back, create it, update it or delete it. It does this by talking
to the database.
Let’s imagine we have an E-Commerce store that sells surfboards and
provides an API for front-end developers at Surf Board shops to use.
Surf Shops might have a server that sends GET requests along with the ID of
the Surf Board they want information about.

155
T.me/nettrain
 The HTTP endpoint or path would look something like:

https://bobssurfboards.com/api/v1/surfboards/1

In the above made-up example, we call everything after the .com a path
component. The API portion specifies that we’re talking to an API. v1 might be
used to specify a version of the API. surfboards represents our resource or thing
we want to use the API to operate on. Finally, the 1 integer represents a specific
surfboard resource in the database.
In our example, we might send a GET request to the above URL and get back
something that looks like this in JSON format:

{
"id": "1",
"name": "The Cruiser",
"description": "A versatile board that's great for beginners and
advanced surfers alike. Comes in a variety of sizes and colors.",
"price": 599.99,
"length": 8,
"width": 22,
"thickness": 3,
"material": "Epoxy",
"image_url": "https://www.bobssurfboards.com/images/surfboards/
cruiser.jpg"
}

If instead we wanted to modify the surfboard object (and we had permissions

to do so), we could create a PATCH request and send our own JSON along to the
server. Maybe we only want to modify the price and the description. We’d send a
PATCH request to https://bobssurfboards.com/api/v1/surfboards/1 along
with a JSON body attached to our request that might look like this:

{
"description": "ON SALE FOR A LIMITED TIME! A versatile board
that's great for beginners and advanced surfers alike. Comes in a
variety of sizes and colors.",
"price": 599.99,
}

I hope this has been a helpful primer for understanding more about APIs.

156
T.me/nettrain
Note that not every API is the same or acts the same. Each API provider should
have documentation that describes how they’d like for you to authenticate to the
API and what format requests and formats should follow.

Sending data via POST

If we wanted to send a POST request (or any other HTTP verb) we can specify
it with the -X flag in curl. Curl will also expect us to give it some data, which
we’ll provide with the -d or —data flags.
There are two formats we use to send data with requests. We can use
application/x-www-form-urlencoded, which is normally used to send to forms.
The other format is application/json.
If you don’t tell curl what format to use, it’ll use application/x-www-form-
urlencoded

Here’s how we’d send data using the first method:

curl -X POST -d "name=ShellSamurais&type=blog-post” https://

jsonplaceholder.typicode.com/posts

The -X flag is followed by the HTTP verb or method and the -d flag is
followed by the data we want to post! You can try POSTing to the
jsonplaceholder API now:

stetson@linux_rocks:/$ curl -X POST -d

"name=ShellSamurais&type=blog-post" https://
jsonplaceholder.typicode.com/posts
{
"name": "ShellSamurais",
"type": "blog-post",
"id": 101
}stetson@linux_rocks:/$

If instead we wanted to send data formatted as a JSON string, we can specify

the Content-Type with a -H or header flag. Our -d flag must also be provided,

157
T.me/nettrain
except this time we’ll give the data in JSON format:

stetson@linux_rocks:/$ curl -X POST -d '{"name": "Shell Samurais",

"type": "Blog Post"}' -H "Content-Type: application/json" https://
jsonplaceholder.typicode.com/posts
{
"name": "Shell Samurais",
"type": "Blog Post",
"id": 101
}stetson@linux_rocks:/$

Neither of these responses is actually updating anything on the JSON

Placeholder website. We do get a response as if it was updating data. This is all
just for demonstration purposes and to act as a real-world API might.

Beyond Basics
Curl is a great tool for testing APIs, downloading data like images and videos
and saving web pages. You can use curl in your own shell scripts and programs
to automate tasks, get data from the web and more. Curl can do so much and I
invite you to read up on curl’s official website to learn more about the powers of
this tiny tool.

jq - Processing JSON
jq is like sed for JSON data. JSON, or Javascript Object Notation is an
extremely common format for log files, API calls and the Javascript programming
language. Earlier, we looked at using curl to call APIs. When we curl an API
endpoint, we’ll normally get back a bit of JSON that’s a bit…ugly. jq makes
formatting and searching through JSON data easy. Let’s look at how to use it.
We’ll start by using a free API at https://api.publicapis.org/entries — this
API endpoint is provided by publicapis.org and simply returns a JSON document
of other APIs we could use. Go ahead and run curl https://
api.publicapis.org/ entries now in your terminal.
Whoa! As is tradition, we get back a LOAD of text in JSON format. This data
really sucks to parse through manually and try to read. Enter…jq.
First, make sure you have jq with apt-get install jq

158
T.me/nettrain
 Then, run the prior curl command again, but attach a pipe and send it to jq:

stetson@linux_rocks:~$ curl https://api.publicapis.org/entries | jq

‘.’

This time, you’ll notice that jq has formatted the text neatly, a term typically
called pretty printing. We use the ‘.’ at the end to tell jq to take the input and
produce it, unchanged to our terminal. jq is useful enough with just that trick,
but let’s try some more!

You’ll notice if you scroll all the way to the top of the file at https://
api.publicapis.org/entries (or load it with your browser) that we see
something like this:

{
"count": 1425,
"entries": [
{
"API": "AdoptAPet",
"Description": "Resource to help get pets adopted",
"Auth": "apiKey",
"HTTPS": true,
…

Let’s dive into that entries key. JSON is a key-value object store, and entries
contains a list of all the APIs that public APIs is providing us. Try this command
out now:

curl https://api.publicapis.org/entries | jq '.entries'

We could also specify the count key if we’d like, which would return the
number of entries (1425 as of this writing) that come back with:

curl https://api.publicapis.org/entries | jq '.count'

Back to the entries filter, let’s tell jq that we want just the first item:

159
T.me/nettrain
 stetson@linux_rocks:~$ curl https://api.publicapis.org/entries | jq
'.entries[0]'
\ % Total % Received % Xferd Average Speed Time Time
Time Current
Dload Upload Total Spent
Left Speed
100 274k 0 274k 0 0 848k 0 --:--:-- --:--:--
--:--:-- 845k
{
"API": "AdoptAPet",
"Description": "Resource to help get pets adopted",
"Auth": "apiKey",
"HTTPS": true,
"Cors": "yes",
"Link": "https://www.adoptapet.com/public/apis/pet_list.html",
"Category": "Animals"
}

As with many things in computer science, jq is zero-indexed, meaning that the

first entry is the referenced as 0. You can change the 0 index to 5, 10, or 500 and
get another entry:

stetson@linux_rocks:~$ curl https://api.publicapis.org/entries | jq

'.entries[500]'
% Total % Received % Xferd Average Speed Time Time
Time Current
Dload Upload Total Spent
Left Speed
100 274k 0 274k 0 0 383k 0 --:--:-- --:--:--
--:--:-- 383k
{
"API": "Hotstoks",
"Description": "Stock market data powered by SQL",
"Auth": "apiKey",
"HTTPS": true,
"Cors": "yes",
"Link": "https://hotstoks.com?utm_source=public-apis",
"Category": "Finance"
}

Diving deeper, if we only want one field from the returned data, we can chain
that on to the end and get it back. Let’s try it with the description key:

160
T.me/nettrain
 stetson@linux_rocks:~$ curl https://api.publicapis.org/entries | jq
'.entries[500].API'
% Total % Received % Xferd Average Speed Time Time
Time Current
Dload Upload Total Spent
Left Speed
100 274k 0 274k 0 0 892k 0 --:--:-- --:--:--
--:--:-- 892k
"Hotstoks"

How about the name of every API in the list? Use the following:

stetson@linux_rocks:~$ curl https://api.publicapis.org/entries | jq

'.entries[] | .Description'
"Resource to help get pets adopted"
"Collection of axolotl pictures and facts"
"Daily cat facts"
"Cat as a service (cats pictures and gifs)"
"Pictures of cats from Tumblr"
"Random dog facts"
"Random facts of Dogs"
"Based on the Stanford Dogs Dataset"
"Retrieve recent or notable birding observations within a region"

The above jq query uses [] to specify that entries is a list or array. The pipe
character pipes each item in the list to the next filter, .Description.

We won’t go into jq any deeper today, but hopefully this has shown some of
the possibilities of the tool! As always, jq becomes even more useful when
combined with other Linux tools and utilities! When you’re working with JSON
data and want to chop it up quickly, reach for JQ!

Tmux - Secret Background Windows

tmux or Terminal Multiplexer is a piece of software used to manage terminal
sessions. In addition, it can spin up long-standing terminals in the background of
the Linux operating system that can be saved and recalled later. Tmux cancaome
in handy for when you are doing work on a Linux system remotely via SSH and
need to keep a session open and to come back to later.

161
T.me/nettrain
 The name terminal multiplexer gives a hint to its usage. From one shell
window, we can arrange multiple windows or panes. Each of these windows is
opened in a session. We can save those sessions for later by de-taching from the
window and later re-attaching and pick our work back up again. This will all
make sense soon, so let’s get started with tmux!
First, install tmux with apt-get install tmux. Then, simply run the tmux
command. You’ll immediately see a “status bar” at the bottom of your screen,
but otherwise, your terminal should look the same. There’s some other
components in the status bar like the date as well. You can even customize the
status bar further, but let’s keep it simple for now.

Tmux starts us off with a single window and a single pane. Tmux uses the
concept of a prefix key and a command key. By default, the prefix key is Control+b.
This is often notated as C-b in documentation for tmux. Press Control+b now.

162
T.me/nettrain
Tmux is now listening for a command. Press % on your keyboard and the screen
will split in two vertically! If we instead wanted to split our window
horizontally, we could press Control+b and then the double quote key “.
Your screen is now split in half vertically or horizontally, so how do you get to
the other pane? Simple! Hit Control+b again, and then an arrow key in the
direction of the pane you’d like to navigate to. Tmux will highlight the currently
used pane.

Closing Panes
Feel free to go crazy and split panes horizontally and vertically until your
terminal is an unusable mess. Then, to close a pane, we can use Control+d or just
type exit to close the terminal pane.

Creating Windows
Panes are great but I honestly use windows more often. Let’s make one now.
Hit the prefix command again (Control+b) and then c on your keyboard for
create. You’ll see that your status bar has now updated to reflect the new
window:

163
T.me/nettrain
Navigating Windows
To select a window, hit Control+b and then the number of the window (0 or 1
in my screenshot above). You can also use n and p for next and previous.

Renaming Windows
One more tip, the default window name sh or bash isn’t very descriptive. You
can hit Control+b and then the comma to rename your window to something
more memorable. My windows below are named hacker stuff and Shell
samurais. Nice!

Saving Sessions for Later

To get out of Tmux, you can simply exit all of your windows and panes by
closing them with exit or Control+d…or, you could keep the session for later
and save it’s state. To save a session, simply hit Control+b and then d for detach.

164
T.me/nettrain
Instantly your status bar is gone and you’re thrust back into your “regular” shell.
Nice! Where did the session go?
To get our session back, we have to re-attach to it. If you run tmux again, you’ll
get another session, but not the one we were just in! First, list your sessions with
tmux ls:

stetson@linux_rocks:/$ tmux ls
0: 1 windows (created Tue Mar 7 03:15:24 2023)

Then, we can re-attach to the session with tmux attach -t 0. The 0 is the
session number we saw along the left of the tmux ls output. Like the default
window names, 0 isn’t a very easily rememberable name.
We can rename the session we listed out with tmux rename-session -t 0
hacking if we’d like. Then we’d be able to attach with tmux attach -t hacking

stetson@linux_rocks:/$ tmux rename-session -t 0 hacking

stetson@linux_rocks:/$ tmux ls
hacking: 1 windows (created Tue Mar 7 03:15:24 2023)
stetson@linux_rocks:/$ tmux attach -t hacking

If instead, you wanted to start a session and name it, just type tmux new -s
<session-name>

Congrats! You’ve learned yet another Linux utility to add to your Shell
Samurai belt! I hope tmux comes in handy on your journey. Like many tools,
we’ve only just scratched the surface of what they can do. With just this basic
knowledge, you can do a lot!

lsof
Linux uses integers or numbers to track files that are currently open as well as
their associated processes. When I say open, I don’t just mean a file that’s open in
your text editor, but any file that is currently being used by a process, daemon or
program running on the system. Remember to that in Linux, everything is a file.
This includes processes, network devices, disks and more. This means lsof can
track all of those resources!

165
T.me/nettrain
 lsof allows us to retrieve information about open files by accessing the
kernel’s internal file table. We can see details like the process id, user id, file
descriptor, file type, device numbers and more. This makes it an invaluable tool
for troubleshooting problems related to file access, network connections, and
resource management.
To use lsof, simply type the command followed by various options and
arguments to filter the results as needed. For example, to list all open files by a
specific process, use the -p flag followed by the process ID.
I’ll give this a try now by opening vim in one terminal session with vim
my_file.
Then, in another terminal session, I’ll run lsof with the process id:

root@linux_rocks:/# lsof -p 31449

COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME
vim 31449 root cwd DIR 0,162 4096 790887 /
vim 31449 root rtd DIR 0,162 4096 790887 /
vim 31449 root txt REG 0,162 2940960 1184991 /usr/bin/
vim.basic
vim 31449 root mem REG 254,1 1184991 /usr/bin/
vim.basic (path dev=0,162)
vim 31449 root mem REG 254,1 1054259 /usr/lib/
aarch64-linux-gnu/libnss_files-2.31.so (path dev=0,162)
vim 31449 root mem REG 254,1 1054432 /usr/lib/
aarch64-linux-gnu/libogg.so.0.8.4 (path dev=0,162)
vim 31449 root mem REG 254,1 1054450 /usr/lib/
aarch64-linux-gnu/libvorbis.so.0.4.8 (path dev=0,162)
vim 31449 root mem REG 254,1 1054350 /usr/lib/
aarch64-linux-gnu/libutil-2.31.so (path dev=0,162)
vim 31449 root mem REG 254,1 1054003 /usr/lib/
aarch64-linux-gnu/libz.so.1.2.11 (path dev=0,162)
vim 31449 root mem REG 254,1 1053527 /usr/lib/
aarch64-linux-gnu/libexpat.so.1.6.11 (path dev=0,162)
vim 31449 root mem REG 254,1 1053922 /usr/lib/
aarch64-linux-gnu/libltdl.so.7.3.1 (path dev=0,162)
vim 31449 root mem REG 254,1 1054409 /usr/lib/
aarch64-linux-gnu/libtdb.so.1.4.5 (path dev=0,162)
vim 31449 root mem REG 254,1 1054465 /usr/lib/
aarch64-linux-gnu/libvorbisfile.so.3.3.7 (path dev=0,162)
vim 31449 root mem REG 254,1 1053481 /usr/lib/
aarch64-linux-gnu/libpcre2-8.so.0.9.0 (path dev=0,162)

166
T.me/nettrain
 vim 31449 root mem REG 254,1 1054193 /usr/lib/
aarch64-linux-gnu/libc-2.31.so (path dev=0,162)
vim 31449 root mem REG 254,1 1054322 /usr/lib/
aarch64-linux-gnu/libpthread-2.31.so (path dev=0,162)
vim 31449 root mem REG 254,1 1054503 /usr/lib/
aarch64-linux-gnu/libpython3.8.so.1.0 (path dev=0,162)
vim 31449 root mem REG 254,1 1054201 /usr/lib/
aarch64-linux-gnu/libdl-2.31.so (path dev=0,162)
vim 31449 root mem REG 254,1 1054493 /usr/lib/
aarch64-linux-gnu/libgpm.so.2 (path dev=0,162)
vim 31449 root mem REG 254,1 1051099 /usr/lib/
aarch64-linux-gnu/libacl.so.1.1.2253 (path dev=0,162)
vim 31449 root mem REG 254,1 1053850 /usr/lib/
aarch64-linux-gnu/libcanberra.so.0.2.5 (path dev=0,162)
vim 31449 root mem REG 254,1 1051215 /usr/lib/
aarch64-linux-gnu/libselinux.so.1 (path dev=0,162)
vim 31449 root mem REG 254,1 1053484 /usr/lib/
aarch64-linux-gnu/libtinfo.so.6.2 (path dev=0,162)
vim 31449 root mem REG 254,1 1054224 /usr/lib/
aarch64-linux-gnu/libm-2.31.so (path dev=0,162)
vim 31449 root mem REG 254,1 1054134 /usr/lib/
aarch64-linux-gnu/ld-2.31.so (path dev=0,162)
vim 31449 root 0u CHR 136,2 0t0 5 /dev/pts/2
vim 31449 root 1u CHR 136,2 0t0 5 /dev/pts/2
vim 31449 root 2u CHR 136,2 0t0 5 /dev/pts/2
vim 31449 root 3u REG 0,162 12288 792734 /.myfile.swp

Above, we see output generated by running lsof and the process ID 31449.
The output shows information about the open files and file descriptors associated
with the vim process running as the root user. The numbers that Linux uses to
track open files are known as file descriptors.
Let's break down the output:

• COMMAND: The command (or program) associated with the process, in

this case, vim.
• PID: The process ID, which is 31449 in this example.
• USER: The user who owns the process, in this case, root.
• FD: The file descriptor (FD) number. Standard file descriptors are: 0 for
standard input (stdin), 1 for standard output (stdout), and 2 for standard

167
T.me/nettrain
 error (stderr). Other numbers represent additional open files or resources.
• TYPE: The type of the file or resource (e.g., DIR for a directory, REG for a
regular file, CHR for a character device).
• DEVICE: The device number for the file or resource.
• SIZE/OFF: The file size or offset.
• NODE: The inode number.
• NAME: The name or path of the file or resource.

In this output, you can see that the vim process has several file descriptors:

• cwd and rtd: The current working directory and the root directory of the
process, both pointing to the root directory /.
• txt: The executable (text) file associated with the process, located at /usr/
bin/vim.basic.
• Multiple mem entries: These represent memory-mapped files, such as
shared libraries and the executable itself, loaded into the process's address
space.
• 0u, 1u, and 2u: Standard input (stdin), standard output (stdout), and
standard error (stderr), respectively. They are all connected to the same
character device, /dev/pts/2, which is a pseudo-terminal.
• 3u: An additional open file descriptor pointing to a regular
file /.myfile.swp. This is likely a swap file created by vim to store changes
made to a file (.myfile) during an editing session.

More about File Descriptors

When a process opens a file or creates a new I/O object, the operating
system's kernel assigns a file descriptor to represent that particular resource. The
file descriptor then serves as a handle, allowing the process to read, write, or
perform other operations on the associated resource.We use the lsof or List
Open Files to look at a process’s open files.
File descriptors make it easier for programs to interact with I/O objects by
abstracting underlying details. When a program needs to perform an operation
on an open file or socket, it can simply refer to the associated file descriptor. The

168
T.me/nettrain
operating system then translates the file descriptor into the appropriate internal
data structures to carry out the requested operation.

The /proc directory

/proc Directory
Once again, in Linux, everything is a file, even processes! Information about
processes is stored in a special directory (also called a virtual filesystem) under the
root directory (/) called /proc. Try listing the contents of /proc now:

stetson@linux_rocks:~$ ls -l /proc
total 0
dr-xr-xr-x 9 root root 0 Dec
30 00:38 1
dr-xr-xr-x 9 root root 0 Mar
2 00:27 10
dr-xr-xr-x 9 root root 0 Mar
2 00:27 1025510
dr-xr-xr-x 9 root root 0 Mar
2 00:27 105
dr-xr-xr-x 9 root root 0 Mar
2 00:27 108
dr-xr-xr-x 9 root root 0 Mar
2 00:27 109
dr-xr-xr-x 9 root root 0 Mar
2 00:27 11

You’ll see a ton of output and you should also see a directory for every
process that you’d see with the ps command.
The /proc directory contains a set of files and directories, with each directory
usually named after a running process's ID. These directories contain files that
represent different components of the process's state, such as memory usage, file
descriptors, command line arguments, and more.

169
T.me/nettrain
 Some common files and directories found in /proc include:

• /proc/cpuinfo: Shows information about the CPU, such as its model,

vendor, and features.
• /proc/meminfo: Provides details about the system's memory usage,
including total memory, free memory, and memory used by buffers and
caches.
• /proc/mounts: Lists all currently mounted file systems and their mount
options.
• /proc/uptime: Displays the system's uptime in seconds
• /proc/version: Contains information about the kernel version, compiler
used to build it, and the operating system's release.

Let’s go deeper and look at a single processes’ status file in /proc. Type cat /
proc/1/status or replace 1 with any process or directory on your system:

root@linux_rocks:/# cat /proc/1/status

Name: bash
Umask: 0022
State: S (sleeping)
Tgid: 1
Ngid: 0
..truncated
Seccomp:2
Seccomp_filters: 1
Speculation_Store_Bypass: vulnerable
Cpus_allowed: f
Cpus_allowed_list: 0-3
Mems_allowed: 1
Mems_allowed_list: 0
voluntary_ctxt_switches: 2894
nonvoluntary_ctxt_switches: 66

I’ve truncated most of the output here for brevity since there are around 60
lines of output.

170
T.me/nettrain
 You’ll get back tons of output that wasn’t available in the ps output! The /
proc file system lets us peer into what the Linux kernel sees when it manages
resources and processes. Take a look at the man page on proc for more info with
man proc

The takeaway here is that the /proc directory exposes extra details about the
kernel's internal data structures. It provides tons of valuable information about
the system's hardware, running processes, and kernel settings. You can use /
proc for diagnostics and monitoring not possible from other simple commands!

troubleshooting low disk

How would you troubleshoot low disk space?

If you’re running out of disk space, you’re probably going to notice it quickly.
As servers and applications run low on disk space, they start to show weird
symptoms like slow response-times, error messages or other indicators that there
might be an issue.
How can we confirm that our system is low on disk and remedy the situation?
The df or Disk Free command is our friend! Df will allow us to determine how
much disk is being used on our system. You should commit to memory df -h,
which tells df to display disk usage information for all files systems in a human-
readable format.

Let’s try it out:

root@broken-waterfall-4525:~# df -h
Filesystem Size Used Avail Use% Mounted on
tmpfs 47M 1.7M 45M 4% /run
/dev/vda1 9.6G 9.5G 6.9M 100% /
tmpfs 234M 1.1M 233M 1% /dev/shm
tmpfs 5.0M 0 5.0M 0% /run/lock
/dev/vda15 105M 5.3M 100M 5% /boot/efi

171
T.me/nettrain
 tmpfs 47M 4.0K 47M 1% /run/user/0
tmpfs 47M 4.0K 47M 1% /run/user/1000

Looking over the summary from df, we can see that this system has a few
disks. /dev/vda15 is mounted on /boot/efi so it’s used as part of the boot
process. We can ignore the tmpfs lines because they are temporary storage
created in memory. /dev/vda1 looks interesting though…
Looking at the Avail column, it has only 6.9M of disk free. The Use% column
shows 100% as well, that could certainly get us into some trouble! Let’s see if we
can find out what’s going on.
There’s got to be some files somewhere that are getting larger and causing this
issue. After all, our server worked previously, so what changed? Let’s try to track
those files down!
We’ll use a similar command to df called du or disk usage. You can use du -h
on common directories where we might expect large files to gather like /var/
log. The h flag tells du to make the output human readable.
Let’s run du -h on our /var/log directory now:

root@broken-waterfall-4525:/var/log# du -h /var/log
4.0K /var/log/landscape
1.2G /var/log/journal/278483eb55e3c7011468e370639f28de
17M /var/log/journal
4.0K /var/log/private
4.0K /var/log/dist-upgrade
4.0K /var/log/caddy
104K /var/log/apt
48K /var/log/unattended-upgrades
4.0K /var/log/sysstat
105M /var/log/postgresql
1.3G /var/log

There’s quite a few files here, but the /var/log/journal/

278483eb55e3c7011468e370639f28de should jump out! Look at its size!
If this directory contained many more files and directories, we might want to
sort it to be a little more readable. For that, we can do du -h /var/log | sort
-rh . Sort uses the h flag to treat the input as “human-readable” sizes. For

172
T.me/nettrain
example, if the input contained sizes like 2.5K, 10M, and 1G, sort -h will
convert them to a common unit (e.g., bytes) and sort them accordingly, so that
the largest size comes first. The -r option sorts the input in reverse order, so we
get the largest files in our output first:

root@broken-waterfall-4525:/var/log# sudo du -h /var/log | sort -rh

1.3G /var/log
1.2G /var/log/journal/278483eb55e3c7011468e370639f28de
105M /var/log/postgresql
17M /var/log/journal
104K /var/log/apt
48K /var/log/unattended-upgrades
4.0K /var/log/sysstat
4.0K /var/log/private
4.0K /var/log/landscape
4.0K /var/log/dist-upgrade
4.0K /var/log/caddy

Now that we’ve found our culprit, we have a few options to deal with it:

• Delete the file

• Check log rotate settings or journal settings
• Make the disk bigger

If you encounter a situation like this in the real world, you should do all of
these. First deleting the file, after confirming it isn’t needed. Obviously don’t go
deleting files that may be critical to the system. This is just a log file, so it’s
probably ok to delete. In certain situations, logs might have to be kept for audit
purposes, so use your common sense. Also, the logs could contain other details
that might help us diagnose further issues.
Assuming the low disk space is causing our server to not function, it’s helpful
to “stop the bleeding” and remove the log files immediately, putting the server
back into normal operation.
We could just delete the directories with rm, but we’ll instead use sudo
journalctl --vacuum-size=100M. This command will clear all logs older than
the specified size (in this example, 100 MB). You can adjust the size as needed to

173
T.me/nettrain
free up the desired amount of disk space.

root@broken-waterfall-4525:/var/log# rm -rfv /var/log/journal/

278483eb55e3c7011468e370639f28de
Deleted archived journal /var/log/journal/
278483eb55e3c7011468e370639f28de/system.journal~ (500M).
Deleted archived journal '/var/log/journal/
278483eb55e3c7011468e370639f28de/user-1000.journal~ (700M).
Vacuuming done, freed 16.0M of archived journals from

In this case, the logs at /var/log/journal are handled by the journald

service. We can modify the configuration at /etc/systemd/journald.conf so
that we don’t encounter this issue again. Under the [Journal] section, we
could set the following setting:

SystemMaxUse=100M

Then, restart journald with systemctl restart systemd-journald.service

so that the configuration takes effect. This command sets the maximum disk
space that the journal can use. If the journal size exceeds this limit, journald will
delete the oldest logs first to make room for new logs.
Part of being an effective Sysadmin and Shell Samurai is not only in fixing
issues, but ensuring that they don’t happen in the same way again. There’s
nothing worse than fixing an issue and then waiting and having it happen again
and fixing it in the same way! Fix an issue once and be done with it.

Logs - Not Just for Lumberjacks

Understanding logging in Linux is another fundamental skill for

troubleshooting and operating a production system. Logs are text files that get
populated with the output from a program or your system describing what is
happening, often with timestamps attached.
For example, if you are running an Nginx web server, it’ll output logs to /
var/log/nginx/access.log and /var/log/nginx/error.log. These files will
contain a newline for every web request that comes in to your server, along with

174
T.me/nettrain
details like IP address, path requested, HTTP status code and more. If you have a
path that isn’t working, like myexamplewebsite.com/login you can use the logs
to troubleshoot and drill deeper into the issue.
In this section, we’ll talk about what kinds of of things can be logged, how to
view logs and how to configure logs.

Logging Locations
Generally, logs are written to the /var/log directory. However, some
applications are configured to write to different areas of the file system. The
syslog service is what generally sends logs to the system logger.

Log Rotate - Don’t let the disk fill up!

Every sysadmin has run into this scenario: Your Linux server is serving a web
application or doing some other background process and suddenly starts acting
kinda funny. You check the logs and see mysterious entries about disk space. Oh
no. So you dig in further and find that the disk has no space left! Do you simply
add more disk to remedy the problem? No, you keep troubleshooting and find
that the logs are filling your disk. Those devious logs. What was meant to help us
administer our system and keep an eye on health has suddenly betrayed us and
made our system unhealthy. No good.
Logrotate remedies this by…rotating logs. Well, not spinning them in circles,
but cleaning up old logs that we no longer need by compressing them, deleting
them or splitting them up. Generally, if your app is important, you’ll be keeping
logs somewhere else that is not on the Linux server. It could be in the cloud or
another log server you manage yourself. Logrotate lets us set certain parameters
like the max amount of disk space we’d like to keep for logs before it starts
deleting or compressing old ones.
Logrotate is a handy friend and an essential tool for administering any
production web app. Let’s take a look at how to install and configure logrotate.
Logrotate might already being running on your system! We can verify this by
looking in the logs directory at /var/log. List that directory with ls, do you see
files titled syslog.1.gz, syslog.2.gz, kern.log.1.gz, etc? If so, logrotate is

175
T.me/nettrain
probably already running!
How does our system know to rotate logs without us doing anything? Look in
the /etc/cron.daily directory for a file titled logrotate or similar. Go ahead
and cat that file and take a look. Cron will run this file daily and take care of
your log rotation!

Configuring Logrotate
If logrotate is already installed, the configuration will be stored at the /etc/
logrotate.conf file and /etc/logrotate.d directory. The contents of that file
and directory are pasted together to make a complete configuration.
Here’s an example configuration for a program called fail2ban, which keeps
random people from logging into your server:

/var/log/fail2ban.log {

weekly
rotate 4
compress
# Do not rotate if empty
notifempty

delaycompress
missingok
postrotate
fail2ban-client flushlogs 1>/dev/null
endscript

# If fail2ban runs as non-root it still needs to have write

access
# to logfiles.
# create 640 fail2ban adm
create 640 root adm
}

The configuration is fairly readable. The file at /var/log/fail2ban.log will

be rotated every week, 4 compressed copies will be kept.
Many times when you install a package or application, a logrotate
configuration is installed for you! You may find the need to tune it or a logrotate

176
T.me/nettrain
configuration might not always be created, so it’s good to know how to configure
it!

177
T.me/nettrain
 178
T.me/nettrain
 "

5. The Network

The Network
Linux and computing in general would be practically useless if we couldn’t
make computers talk to each other. Linux provides a robust system for managing
network connections and configuring the network. Understanding Linux
networking is essential for optimizing network performance, and
troubleshooting connectivity issues.

A Brief Overview of the OSI Model

In order to think about networks, an understanding of the OSI or Open
Systems Interconnection model is helpful! The OSI model describes how
computers talk and is a model that splits networking into 7 layers:

• Application - layer at which applications interact with the network

• Presentation - Data encryption occurs here as well as making sure data is
in a usable format
• Session - Maintains connections, responsible for ports and sessions
• Transport - Transmits data with transmission protocols like TCP and UDP
• Network - Decides which path the data will take based on IP addresses
• Datalink - Defines format of data on network, usually associated with
MAC addresses
• Physical - Transmits raw bits over a physical medium. Think of electric
passing over a wire or light passing through a fiber optic cable.

Learning the OSI model can be difficult. There are many mnemonics used to
remember it. The one I used is too lewd to mention in this book, but I’ll
recommend All People Seem to Need Data Processing for remembering the layers.
Hot Tip : When working with other systems administrators, you’ll hear the
OSI model referenced from time-to-time. For example: “That sounds like a layer
8 issue to me” is used jokingly to imply that there is user error. Layer 8 doesn’t

179
T.me/nettrain
really exist on the OSI model. Also, network devices often are thought of as
either Layer 3 devices like routers or Layer 2 devices, like switches. Routers work
on the network layer because they make intelligent decisions about where to
send packets based on IP addresses, which are a layer 3 concept.
Switches on the other hand, are a layer 2 device. They are usually quicker
than routers because they don’t make decisions on where to send packets based
on IP addresses, but rather based on MAC addresses. MAC addresses are a
physical id that is burned in to a network interface. Switches keep a local
database called the ARP table (Address Resolution Protocol) which maps IP
address and MAC address to an interface. Switches forward packets rather than
route them based on the destination MAC address.

Interfaces and IPs

To see which devices your Linux system has, we can use the ifconfig
command or interface configuration:

root@linux_rocks:/# ifconfig
eth0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 1500
inet 172.17.0.2 netmask 255.255.0.0 broadcast
172.17.255.255
ether 02:42:ac:11:00:02 txqueuelen 0 (Ethernet)
RX packets 24273 bytes 35763327 (35.7 MB)
RX errors 0 dropped 0 overruns 0 frame 0
TX packets 9121 bytes 509036 (509.0 KB)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0

lo: flags=73<UP,LOOPBACK,RUNNING> mtu 65536

inet 127.0.0.1 netmask 255.0.0.0
loop txqueuelen 1000 (Local Loopback)
RX packets 0 bytes 0 (0.0 B)
RX errors 0 dropped 0 overruns 0 frame 0
TX packets 0 bytes 0 (0.0 B)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0

You’ll see that the command gives sections for each interface on the system.
First, eth0, and then lo or loopback. In addition to the IP address and subnet
configurations we’d expect, we can see other statistics about the interface like it’s
mac address (ether 02:42:ac:11:00:02), received packets count, transmitted

180
T.me/nettrain
packets, errors and other statistics. You might also see an interface called wlan0,
which is a wireless interface.
The loopback address on the bottom at 127.0.0.1 is a special kind of IP
address typically used for testing and ensuring that the network stack works on
the system. The loopback address is only reachable locally and cannot be reached
from outside the system.

The Routing Table

Networks route packets based on a routing table. Huge routers in datacenters
make decisions about how to get network data to their destination based on a
routing table. They look at the destination IP address of a packet and perform a
lookup in their local routing table to decide where to send the packet. Linux
hosts work the same way.
View your local routing table with ip route:

root@linux_rocks:~# ip route
default via 172.17.0.1 dev eth0
172.17.0.0/16 dev eth0 proto kernel scope link src 172.17.0.2

The Linux routing table is pretty simple by default. We see 2 entries, first, a
default route which says that any packet that doesn’t match other entries will get
routed to 172.17.0.1. The device used to send those packets is eth0.
Second, there is an entry that defines the local network, 172.17.0.0/16. This
rules means that the host can reach the subnet directly from its eth0 interface.
Another way to display routes is via the route -n command. This method is
a little older and generally ip route is used, but route -n is still supported. The
output we get from route -n is largely the same:

root@linux_rocks:~# route -n
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref
Use Iface
0.0.0.0 172.17.0.1 0.0.0.0 UG 0 0
0 eth0
172.17.0.0 0.0.0.0 255.255.0.0 U 0 0
0 eth0

181
T.me/nettrain
 We do see some different fields, though. Namely, the Flags column has U and
UG underneath. U means the network is up. UG means that the network is up and
is a gateway.

ARP Table
ARP stands for Address Resolution Protocol. The ARP table is a local database
that maps MAC addresses and IP addresses to an interface to know where to
forward packets. We can view our linux arp table with arp -a:

root@linux_rocks:~# arp -a
? (172.17.0.1) at 02:42:71:e3:bd:c5 [ether] on eth0

Above, we see the MAC address of the gateway at 172.17.0.1 is at

02:42:71:e3:bd:c5 on device eth0.

Checking Basic Connectivity with ping

Ping is probably the first networking command any systems administrator
learns. Ping checks if packets can reach a host using the ICMP or Internet Control
Message Protocol. Ping is super simple to use, simply specify a domain or IP
address with ping <host>. When you’re done sending packets, press Control+C
to stop the operation.

root@linux_rocks:~# ping google.com

PING google.com (172.217.4.206) 56(84) bytes of data.
64 bytes from lga15s48-in-f206.1e100.net (172.217.4.206):
icmp_seq=1 ttl=37 time=26.9 ms
64 bytes from lga15s48-in-f206.1e100.net (172.217.4.206):
icmp_seq=2 ttl=37 time=26.8 ms
64 bytes from lga15s48-in-f206.1e100.net (172.217.4.206):
icmp_seq=3 ttl=37 time=27.3 ms
^C
--- google.com ping statistics ---
3 packets transmitted, 3 received, 0% packet loss, time 2011ms
rtt min/avg/max/mdev = 26.837/27.003/27.250/0.177 ms

Ping shows the domain and IP address that was resolved along with a line for
each packet it sends with a sequence number, time to live and time in

182
T.me/nettrain
milliseconds it took to get a reply. Ping also is handy for checking the latency to a
host by observing the time it took to reply.
Ping gives us a tidy summary at the end of how many packets it sent and time
to send packets as rtt or round trip time.
Remember that many hosts may block ICMP or ping packets. So if you’re
sending a ping to a host and it isn’t replying, it may still be online but just
ignoring your requests.

Traceroute
You might already be familiar with the tracert command on Windows
systems. Traceroute accomplishes the same goal of observing how packets are
routed from your local host to a destination address. The way traceroute works
is really interesting. Traceroute sends packets with ever-increasing TTL or Time
to Live value. Time to Live packets are used by networks to prevent packets from
being sent around in a loop forever. When a device receives a packet, it
decrements the TTL by 1. If the value ever reaches 0, the packet is then discarded
and a message is sent back to the source.
For example, traceroute would sends it’s first packet with a TTL of 1 to the
destination. When the first router or layer 3 device gets that packet, it will
decrement the TTL field in the packet by 1 and thus the TTL is 0. Since the value
of the TTL is 0, the router will discard the packet and send back an ICMP time
exceeded message to the source. Traceroute then records the IP address of that
router and time it took to get a message back. Traceroute increases the TTL to 2
and sends the packet again, repeating the process with the next router in the
“chain”. This process repeats until the packet reaches its destination.
That’s a lot of theory, let’s see an example!
Use traceroute and any domain name or IP address now:

root@linux_rocks:~# traceroute google.com

traceroute to google.com (142.250.190.110), 30 hops max, 60 byte
packets
1 172.17.0.1 (172.17.0.1) 0.144 ms 0.029 ms 0.080 ms
2 192.168.1.1 (192.168.1.1) 30.871 ms 30.780 ms 30.695 ms

183
T.me/nettrain
 3 142-254-144-193.inf.spectrum.com (142.254.144.193) 39.544 ms
49.353 ms 49.090 ms
4 lag-63.mdtwohrs02h.netops.charter.com (24.29.4.197) 56.695 ms
56.641 ms 56.612 ms
5 lag-44.fpkhoh0301r.netops.charter.com (65.29.38.110) 48.874 ms
48.833 ms 48.782 ms
6 lag-11.mcr11blasoh56.netops.charter.com (65.29.33.66) 54.551
ms lag-11.dytnoh5501r.netops.charter.com (65.29.33.206) 25.436 ms
19.323 ms
7 lag-28.rcr01clmkohpe.netops.charter.com (65.29.1.44) 25.019 ms
21.453 ms 26.597 ms
8 lag-25.chctilwc00w-bcr00.netops.charter.com (107.14.17.252)
35.699 ms lag-15.chctilwc00w-bcr00.netops.charter.com (66.109.6.68)
28.653 ms lag-27.vinnva0510w-bcr00.netops.charter.com (66.109.6.66)
31.493 ms
9 lag-11.chcgildt87w-bcr00.netops.charter.com (66.109.6.20)
33.576 ms lag-31.chcgildt87w-bcr00.netops.charter.com (66.109.10.82)
33.445 ms lag-31.asbnva1611w-bcr00.netops.charter.com (107.14.18.82)
36.200 ms
10 74.125.147.54 (74.125.147.54) 38.337 ms 24.30.201.159
(24.30.201.159) 38.136 ms 74.125.50.44 (74.125.50.44) 36.028 ms
11 * * *
12 108.170.246.33 (108.170.246.33) 34.563 ms 142.251.52.64
(142.251.52.64) 33.461 ms 108.170.240.97 (108.170.240.97) 33.444 ms
13 142.251.60.205 (142.251.60.205) 29.742 ms 108.170.246.66
(108.170.246.66) 30.687 ms 108.170.243.175 (108.170.243.175) 27.983
ms
14 ord37s35-in-f14.1e100.net (142.250.190.110) 27.861 ms
142.251.49.21 (142.251.49.21) 39.974 ms ord37s35-in-f14.1e100.net
(142.250.190.110) 33.954 ms

Each line in this output is a router between me and google.com. We see the
DNS name of the device and IP address along with response times to that device.
What’s interesting about traceroute output is that you can glimpse into
network infrastructure a little bit. For example, taking one of these first lines,
lag-63.mdtwohrs02h.netops.charter.com, I can tell that this is probably a link
aggregation port with id 63 and the device name is likely mdtwohrs02h. Netops is
most likely the team at Charter doing Network Engineering. While I can’t
decrypt the exact syntax of charter’s device names, it’s fun to peer inside the
network a bit.
Another line has asbnva which stands for Ashburn, Virginia. There’s a really

184
T.me/nettrain
great slide deck I’ll recommend put out by Nanog (North American Network
Operator’s Group) called A Practical Guide to (Correctly) Troubleshooting with
Traceroute that goes way in-depth on some of these hostnames and the history.

Sniffing Packets with tcpdump

Tcpdump is one of my all time favorite tools. You might have heard of the
GUI tool Wireshark before. Wireshark allows you to listen and record each and
every packet on your machine as they are sent and received. Wireshark allows
you to get deep into analyzing packets and see each and every header, source
and destination address and all sorts of other details.
tcpdump is like Wireshark for the command line. Let’s give it a try now. Install
tcpdump with sudo apt-get install tcpdump.
Once you have tcpdump installed, let’s fire it up! Use sudo tcpdump -i eth0
(or whatever your interface is, you can list them with ifconfig or ip addr. You
should now see some output from your shell. In the rare event that you don’t see
any output, you might have to generate some network traffic. From another shell
window, try curling your favorite website via curl https://shellsamurai.com,
installing some packages or anything else that might generate traffic.

root@linux_rocks:~# tcpdump -i eth0

tcpdump: verbose output suppressed, use -v or -vv for full protocol
decode
listening on eth0, link-type EN10MB (Ethernet), capture size 262144
bytes
13:40:43.138045 IP linux_rocks.37145 > 192.168.65.5.domain: 27231+
A? shellsamurai.com. (34)
13:40:43.139202 IP linux_rocks.37145 > 192.168.65.5.domain: 49826+
AAAA? shellsamurai.com. (34)
13:40:43.139847 IP linux_rocks.59845 > 192.168.65.5.domain: 59682+
PTR? 5.65.168.192.in-addr.arpa. (43)
13:40:43.295953 IP 192.168.65.5.domain > linux_rocks.37145: 27231
2/0/0 A 23.21.234.173, A 23.21.157.88 (66)
13:40:43.319902 IP 192.168.65.5.domain > linux_rocks.59845: 59682
NXDomain 0/0/0 (43)
13:40:43.388302 IP 192.168.65.5.domain > linux_rocks.37145: 49826
0/1/0 (95)
13:40:43.390099 IP linux_rocks.35556 >
ec2-23-21-234-173.compute-1.amazonaws.com.https: Flags [S], seq

185
T.me/nettrain
3979911459, win 64240, options [mss 1460,sackOK,TS val 3904437079 ecr
0,nop,wscale 7], length 0

tcpdump puts out a lot of data, so I’ve omitted a lot above, but we can quickly
start to interpret what tcpdump is telling us. The first few lines of the output
above we can see my system querying DNS for shellsamurai.com’s IP address.

Let’s breakdown a single line of the output:

• 13:40:43.138045 IP linux_rocks.37145 > 192.168.65.5.domain:

27231+ A? shellsamurai.com. (34)

• 13:40:43.138045: The timestamp of the packet capture, showing the time

when the packet was captured.
• IP linux_rocks.37145 > 192.168.65.5.domain: The source and
destination IP addresses and port numbers for the captured packet. In this
case, the source IP address is linux_rocks with a randomly assigned source
port number of 37145. The destination IP address is 192.168.65.5, and the
destination port number is domain, which is the well-known port number
for DNS traffic (port 53).
• 27231+ A? shellsamurai.com. (34): The DNS query message itself. The
27231 is the query ID, which is used to match responses to queries. The A?
indicates that this is an IPv4 address query, and shellsamurai.com. is the
domain name being queried. The (34) at the end shows the length of the
query message in bytes.

If we wanted to save output from tcpdump, we can save it to a file with sudo
tcpdump -w /tmp/my_tcpdump. Tcpdump will dump packets to the file until
until you send an interrupt with Control+C. Then, you can move the file back to
another system and analyze it with Wireshark.

This is just a quick primer on tcpdump! There are tons of options and
configurations that can be used to filter traffic. One of my favorite resources for

186
T.me/nettrain
tcpdump is Daniel Miessler’s blog post “A tcpdump Tutorial with examples” —
be sure to check it out if you are interested in learning tcpdump more effectively!

TCPDUMP license plate spotted in San Francisco

187
T.me/nettrain
 DNS - The Yellow Pages of the Internet

It’s not DNS

There’s no way it’s DNS
It was DNS
-unknown

DNS was once described to me as the “Yellow Pages of the Internet”. Like a
phone book (people still use these?) that matches someone’s name to their phone
number, DNS or Domain Name System does the same for Domain Names to
map them to an IP address.
Instead of internet users needing to have Rainman-like ability to remember
that reddit is at 151.101.1.140, instagram is at 157.240.254.174 and amazon is
located at 205.251.242.103, we can have DNS do the heavy lifting!
DNS is a huge distributed database of all of these mappings between IP
address and domain name. In this section, we’ll look at the basic operations of
DNS, how Linux handles DNS and a few tools to configure DNS.

How does DNS work?

DNS relies on a few different components to do its job. Name servers are one of
those primary components. A name server keeps configuration of DNS settings
and answers queries or questions for clients that send it message. A DNS server
can be authoritative or recursive. Authoritative means that the server holds the
record you’re looking for, it is the authority. A recursive name server goes back
and asks other servers for DNS settings, which may in turn ask another server
until we reach an authoritative server that has the records.

188
T.me/nettrain
 Zone files are the actual information stored the domain. Zone files keep a
record name, TTL, Class, Type and Data. Let’s look at an example:

shellsamurai.com. 377 IN A 23.21.157.88

• Record Name: shellsamurai.com — this is the domain the record is about

• TTL - 377 - This is the time in seconds after which the record is thrown
away and a new one retrieved.
• Class Type - IN - Namespace of the record, IN is used for the internet
• Record Type - A - This is the type of info stored in the record, you might
have heard of MX records for mail and A for address before
• Data - 23.21.157.88 - This field contains the actual value for the record. This
A file has an IP address associated.

Resolvers vs Authoritative Nameservers

The DNS server that we query (called a resolver) is different from the servers
that actually contain the DNS records (authoritative nameservers).
There are generally three levels of authoritative DNS servers:

• root nameserver - Responsible for all domains. Does not contain zone
records. Contains a list of TLD nameservers
• .com TLD nameserver - Responsible for all domains ending in .com
• example.com nameserver - Responsible for example.com and all
subdomains

The root name server delegates queries down to the .com nameserver, who also
delegates queries to the example.com nameserver. This design allows DNS to be
decentralized.

Life of a DNS Request

Let’s go deeper into the DNS request process and examine how your system

189
T.me/nettrain
might find the record for shellsamurai.com. When your system makes a
request to shellsamurai.com, the process is kicked off:

First, your system will look at its local cache to see if there is already an IP
address stored for the domain. If there isn’t, the request gets forwarded on to a
recursive DNS resolver.
Then, the recursive resolver checks its local cache to see if it has an entry for
the domain. If it does, it responds with it. If it doesn’t have the entry cached, it
queries the root DNS servers to find who the authoritative DNS server is that
handles .com domain names.
The recursive resolver then queries the TLD (Top Level Domain) DNS server
that handles .com domains and obtains the IP address of the nameserver that is
responsible for shellsamurai.com
The recursive resolver finally then queries the authoritative DNS server to
find the IP address mapped to the website.
Once the authoritative server finds the address, it provides it to the recursive
server who will cache the result and send the IP address back to the local DNS
resolver, who also caches it and provides it to your local system.

/etc/hosts
The file at /etc/hosts acts as a kind of override for host to IP address
mappings. Before Linux calls out to DNS servers, it looks locally to see if there
are any domain to IP address mappings already in place. Here’s an example:

stetson@linux_rocks:/$ cat /etc/hosts

127.0.0.1 localhost

/etc/resolv.conf

190
T.me/nettrain
 The file at /etc/resolv.conf used to be the primary way to adjust your DNS
server settings. It’s usually managed for you by the system these days. If you
want to add a name server and change the config, you can use the nameserver
keyword:

stetson@linux_rocks:~# cat /etc/resolv.conf

# DNS requests are forwarded to the host. DHCP DNS options are
ignored.
nameserver 192.168.65.5

Here, my system has its DNS server set to 192.168.65.5. You probably don’t
need to adjust this setting on your own.

Exploring DNS further with dig

Dig is a command we can use for querying DNS servers and troubleshoot
DNS issues. In your shell, go ahead and type dig followed by any domain name
and hit enter:

root@linux_rocks:~# dig shellsamurai.com

; <<>> DiG 9.16.1-Ubuntu <<>> shellsamurai.com

;; global options: +cmd
;; Got answer:
;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 38503
;; flags: qr rd ra; QUERY: 1, ANSWER: 2, AUTHORITY: 0, ADDITIONAL:
0

;; QUESTION SECTION:
;shellsamurai.com. IN A

;; ANSWER SECTION:
shellsamurai.com. 377 IN A 23.21.157.88
shellsamurai.com. 377 IN A 23.21.234.173

;; Query time: 106 msec

;; SERVER: 192.168.65.5#53(192.168.65.5)
;; WHEN: Sat Mar 11 14:49:20 EST 2023
;; MSG SIZE rcvd: 66

191
T.me/nettrain
 Dig gives us a lot of output by default, which we can filter with other flags
and options. For now, let’s stick with the output we’ve got:

• ; <<>> DiG 9.16.1-Ubuntu <<>> shellsamurai.com: This line shows

the version of dig being used and the domain name being queried.
• ;; global options: +cmd: This line shows the global options being
used with the dig command.
• ;; Got answer: indicates that the query was successful.
• ;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 38503: This
line shows the header information for the DNS response, including the
opcode, status, and ID of the query.
• ;; flags: qr rd ra; QUERY: 1, ANSWER: 2, AUTHORITY: 0,
ADDITIONAL: 0: This line shows the flags and counts for the DNS
response. In this case, the qr flag indicates that this is a response, the rd
flag indicates that recursion was requested, and the ra flag indicates that
recursion was available. There was 1 query made, 2 answers were
provided, and there were no authority or additional records returned.
• ;; QUESTION SECTION:: This line shows the question section of the DNS
response, which is the domain name being queried and the record type
being requested (in this case, an A record).
• ;shellsamurai.com. IN A: This line repeats the domain name and record
type being queried, with additional information about the record class (IN,
which stands for Internet).
• ;; ANSWER SECTION:: This line shows the answer section of the DNS
response, which lists the IP addresses associated with the queried domain
name.
• shellsamurai.com. 377 IN A 23.21.157.88: This line shows the first
answer, which is the IP address 23.21.157.88 associated with the domain
name "shellsamurai.com". The number 377 indicates the time-to-live (TTL)
for this record.
• shellsamurai.com. 377 IN A 23.21.234.173: This line shows the second
answer, which is the IP address 23.21.234.173 associated with the domain
name "shellsamurai.com". The number 377 again indicates the TTL for this

192
T.me/nettrain
 record.
• ;; Query time: 106 msec: This line shows the time it took for the DNS
query to be completed.
• ;; Query time: 106 msec: This line shows the time it took for the DNS
query to be completed.
• ;; SERVER: 192.168.65.5#53(192.168.65.5): This line shows the IP
address and port number of the DNS server that provided the response.
• ;; WHEN: Sat Mar 11 14:49:20 EST 2023: This line shows the date and
time when the DNS response was received.
• ;; MSG SIZE rcvd: 66: This line shows the size of the DNS response in
bytes. In this case, the response contained 66 bytes of data.

Overall, dig can give us a ton of information about DNS servers that resolve
our query and is very handy for troubleshooting any kind of DNS issue.

DNS can be a confusing topic to understand if you’ve never dealt with it

before. Add in Linux an you’ve got a whole other level of confusion. Hopefully
this section has helped to de-mystify the inner workings of DNS and how your
Linux system handles DNS. Knowing DNS well is a critical Shell Samurai and
Systems Administrator skill!
Transferring files with rsync and scp

scp - Transfer files over SSH

scp or secure copy is a command-line tool used for transferring files between
systems securely. It uses the Secure Shell (SSH) protocol to encrypt data in transit
and even provides authentication and encryption for access to files. I’ve used scp
many times for pulling logs, packet captures and source code to my local system.

Here’s the basic syntax we can use for scp:

scp [options] [source] [destination]

Options are any additional flags we’d pass to change how scp works.

193
T.me/nettrain
 Source is the source file or directory we want to transfer.
Destination refers to the location where we want to copy our file or directory.

We’ll need SSH setup and configured on the systems we’re transferring files
to or from so be sure to brush up on the SSH section of the book to get SSH up
and running. To follow these examples, you’ll need two systems to transfer files
between. Setting up two linux virtual machines might be a good idea. Windows
has a few packages for SCP that can be used as a client as well.

Once you’ve ensured SSH is up and running, let’s try transferring a file:

stetson@localhost:~$ scp shell_samurais.txt [email protected]:/tmp

shell_samurais.txt
100% 0 0.0KB/s 00:00

Above, I copy the file shell_samurais.txt from my local system to the

remote system at 192.168.1.100. Then, on your remote system, verify the file
was received:

root@linux_rocks:/tmp# ls -l
total 4
-rw-r--r-- 1 root root 22 Mar 11 15:56 shell_samurais.txt
root@linux_rocks:/tmp# cat shell_samurais.txt
Hello Shell Samurais!

If instead we wanted to copy an entire directory, we can provide the -r flag to

tell scp to recursively transfer an entire directory:

stetson@localhost:~$mkdir -p my_dir/shell/samurais/secrets/
cloudkeys
stetson@localhost:~$ touch my_dir/shell/samurais/secrets/cloudkeys/
secret_key.txt
stetson@localhost:~$ scp -r my_dir root@localhost:/tmp

Above, I create a few nested directories and a file inside them. Then, use scp
with the -r flag to transfer recursively to my Linux server.

194
T.me/nettrain
rsync
rsync is yet another file transfer tool we can use, short for remote
synchronization. It’s very similar to scp but has a major distinction. rsync will
make what’s called a diff and only transfer the differences over. If you were
making an upload that got interrupted, you can simply rsync again and the files
that didn’t get transferred will get copied over. Additionally, rsync can transfer
only the parts of files that haven’t been transferred yet, making it much quicker
and more efficient than scp. rsync is often configured in cron as a simple backup
solution.

rsync also uses SSH to transfer files. rsync’s syntax is simple to use and
similar to scp:

To transfer local files to a remote server:

stetson@linux_rocks# rsync /my_local_directory root@localhost:/

my_remote_directory

To transfer remote files to your local system:

stetson@linux_rocks# rsync [email protected]:/remote/

directory /local/directory

Well done! This has been a brief overview of a few methods of transferring
files with Linux tools that are built right into most distributions. Indeed, file
backup and transfer is an important concept for keeping critical files safe. Learn
it, live it, love it.

195
T.me/nettrain
 196
T.me/nettrain
 6. Real World Samurai Skills and
Interview Questions

Intro
In this chapter, we’ll look at what I call some “real-world” Shell Samurai
skills. Although most everything in this book so far is applicable in the real-
world, our goal in this chapter is to run through a few tutorials that you can use
in a career working with Linux.
In addition, this chapter contains some common Interview Questions and
scenarios you might run into!

197
T.me/nettrain
 Installing Nginx Web server

Nginx (pronounced engine-x) is one of the most popular pieces of Web Server
software. In addition to just serving content, Nginx can do much more, like
acting as a load balancer and reverse proxy. Nginx is free and open-source and
was first released in 2004. Nginx was ranked first in usage in June 2022 ahead of
Apache and Cloudflare for web severs. There’s a 1 in 3 chance any website you
visit is using Nginx!
Nginx boasts some additional features and tricks, including its ability to
handle over 10k simultaneous connections while using a very low amount of
memory.
Enough talk about Nginx’s popularity and shiny features, let’s take a look at a
simple Nginx install.

Installing Nginx
As is customary and should be familiar to you by now, we need to first install
Nginx. We can do so with apt-get install nginx. You’ll be prompted to
review new packages and dependencies that will be installed before being asked
if you’d like to continue. Hit Y and then enter to confirm:

sudo apt-get install nginx

Reading package lists... Done
Building dependency tree
Reading state information... Done
The following additional packages will be installed:
fontconfig-config fonts-dejavu-core iproute2 libatm1 libbsd0
libcap2 libcap2-bin libelf1 libfontconfig1 libfreetype6 libgd3
libicu66 libjbig0 libjpeg-turbo8 libjpeg8 libnginx-mod-http-
image-filter libnginx-mod-http-xslt-filter libnginx-mod-mail
libnginx-mod-stream libpam-cap libpng16-16 libtiff5 libwebp6
libx11-6 libx11-data libxau6 libxcb1 libxdmcp6 libxml2 libxpm4
libxslt1.1 nginx-common nginx-core tzdata
Suggested packages:
iproute2-doc libgd-tools fcgiwrap nginx-doc ssl-cert
The following NEW packages will be installed:
fontconfig-config fonts-dejavu-core iproute2 libatm1 libbsd0

198
T.me/nettrain
libcap2 libcap2-bin libelf1 libfontconfig1 libfreetype6 libgd3
libicu66 libjbig0 libjpeg-turbo8 libjpeg8 libnginx-mod-http-
image-filter libnginx-mod-http-xslt-filter libnginx-mod-mail
libnginx-mod-stream libpam-cap libpng16-16 libtiff5 libwebp6
libx11-6 libx11-data libxau6 libxcb1 libxdmcp6 libxml2 libxpm4
libxslt1.1 nginx nginx-common nginx-core tzdata
0 upgraded, 35 newly installed, 0 to remove and 0 not upgraded.
Need to get 13.9 MB of archives.
After this operation, 54.9 MB of additional disk space will be
used.
Do you want to continue? [Y/n] Y

The installer will then ask a few questions to confirm your timezone and
region, answer accordingly:

Please select the geographic area in which you live. Subsequent

configuration questions will narrow this down by presenting a
list of cities, representing the time zones in which they are
located.
1. Africa 3. Antarctica 5. Arctic 7. Atlantic 9. Indian
11. SystemV 13. Etc
2. America 4. Australia 6. Asia 8. Europe 10. Pacific
12. US
Geographic area: 12

Please select the city or region corresponding to your time zone.

1. Alaska 3. Arizona 5. Eastern 7. Indiana-Starke 9.

Mountain 11. Samoa
2. Aleutian 4. Central 6. Hawaii 8. Michigan 10.
Pacific
Time zone: 5

Once Nginx has been installed, verify that it’s up and running with systemctl
status nginx. If you see output saying nginx has been disabled or isn’t
running, go ahead and run systemctl start nginx and systemctl enable
nginx to enable it to start on boot up.
Next, navigate to your system’s IP address in your browser as http://
192.168.1.1 to check if Nginx is up and running! As a reminder, you can check
your ip address using ifconfig. You can also use curl ifconfig.me to return

199
T.me/nettrain
your systems public ip address. Ifconfig.me is a free service that simply returns
your public IP.
If all went well, you should see Nginx’s default landing page:

This is great but doesn’t help us much if we can only see the default config
page! We want to serve our own webpages on Nginx!
We can navigate to /var/www/html and ls the contents, we’ll see a
index.nginx-debian.html file there:

root@linux_rocks:/var/www/html# ls
index.nginx-debian.html

Let’s rename it with .bak on the end so we can keep that file. Then, touch a
new index.html file inside of /var/www/html:

root@linux_rocks:/var/www/html# mv index.nginx-debian.html
index.nginx-debian.html.bak
root@linux_rocks:/var/www/html# touch index.html

Now, open your favorite text editor and we’ll write some HTML of our own
inside index.html! I’m using Tailwind for some basic CSS styling in this
example, but feel free to omit that tag if you just want to see a page quickly:

200
T.me/nettrain
 <!DOCTYPE html>
<html lang="en">
<head>
<title>My first Nginx Web Page!</title>
<script src="https://cdn.tailwindcss.com"></script>
</head>
<body>

<h1 class="text-2xl text-orange">

Hello Shell Samurais!
</h1>

<div class="container">
Welcome to my web page
</div>
</body>
</html>

Save the file and close it. Go back to your browser and hit refresh!

Nice! We’ve just served our own web page from scratch with Nginx, well
done! There are many, many more ways to configure Nginx, but I wanted to keep
it brief and simply introduce you to this software.
As mentioned, around 30% of websites today are using Nginx in some form.
Nginx has many more ways you can use it to configure to act as a proxy or load-

201
T.me/nettrain
balancers, among other use-cases. Adding Nginx to your tool belt is a
straightforward way to make yourself more valuable as a Shell Samurai!

Using Git and Version Control

While Git is not super in-scope to understanding Linux, I think you should be
aware of it because you might need to use Git in your career at some point.
Working with code and configuration files on your local system is great, but
what happens when you want to work with others? How can you modify web
page, configuration files and source code collaboratively? Should you zip up
your code and email it to your coworker for them to make their changes? Use
USB drives? Dropbox or Google Drive? Keep your code on a shared drive and
just modify it? What happens when a change is made that completely breaks
your application? What would happen if two developers made a change to the
same file?
Before Version Control, these were very real problems that developers and
systems administrators had to grapple with. Today, we don’t need to worry
about these problems because a solution has been built for us. Version control
allows team members to collaboratively work on code, keeping a running log of
changes that were made, access different versions, and a lot more.
Git is the most common and widely used version control system. In this
section, we’ll make our own Git repo, commit some code to it and start to
understand basic concepts about Git.

An Overview of Git
In addition to developing the Linux kernel, Linus Torvalds also originally
developed Git in 2005 for helping with development on Linux! Git, the software,
is open-source. You may have heard that Microsoft acquired Github and assume
this means they own git too. Github is simply a Git hosting platform. There are
other git hosting platforms, like Gitlab and Bitbucket. You don’t have to host
code on Github.
Git, the software, is a distributed version control system. This means that every
developer keeps a complete copy of the codebase on their local machine. They
don’t need to be connected to a file share or central server to make changes or

202
T.me/nettrain
contributions to code.
Git has a concept of branching and merging, which allows developers to make
isolated changes to the codebase to add new features or bug fixes. Then, they can
merge their code back into the main or master codebase.
Every single change that is made to the codebase is logged and tracked over
time. If anything goes wrong or you want to see the history of a piece of code, Git
makes it easy to do.

Core Git Concepts

Before we go further, we’ll define a few terms necessary for understanding
git.
A repository is a central storage area where all the code and files in a codebase
live. Developers save work here, collaborate with others and track their changes.
Changes made to the codebase are known as commits. A commit is like a
snapshot of the codebase at a certain time along with a short message describing
the change. Commits are given unique ids that can be later referenced.
Branches are how developers can work on different versions of the codebase. A
developer will create a new branch, make changes and then later ask for review
on their code from others and finally merge their code into the main or master
branch.

Getting started with Git and Github

Let’s try git out for ourselves. First, install git on your system with apt-get
install git.
Next, sign up for Github.com in your browser. Github has changed their
onboarding and sign up a lot, but the product remains totally free.
Once you’ve signed up for Github, you’ll be dumped right into Github’s
homepage and see a big green button called “Create repository”. Go ahead and
click it now:

203
T.me/nettrain
 You’ll see a page with a myriad of options. Name your Repository anything
you’d like and select Private. I’ve decided to set my repository to public and call
it my_first_repo.

A Quick Obvious Caution I shouldn’t really need to say: IF YOU MAKE

YOUR REPOSITORY PUBLIC, ANYONE ON THE INTERNET CAN SEE IT. So
don’t add any sensitive data there. Okay, with that out of the way, let’s continue.

204
T.me/nettrain
 When you’re happy with your repository name and privacy settings, hit
Create repository and you’ll be greeted with some instructions:

205
T.me/nettrain
Creating a New Repository on the Command Line
Hit the copy button underneath the instructions “…or create a new
repository on the command line” to save those commands to your clipboard.

Let’s explain the commands you just copied line-by-line:

echo "# my_first_repo" >> README.md

This line echos the text my_first_repo to a file called README.md. Whatever we
put in our README.md file will be formatted in the markdown format. Markdown
is a text format that allows us to define lists, titles and other text features like

206
T.me/nettrain
links. Putting a # before a piece of text will make it a large Heading 1.

git init
This command will create a new git repository on our local system.

git add README.md

This command adds the README.md file to staging. This tells git that we’d like
to include this file in our next commit.

git commit -m "first commit"

This is our first commit! Git takes the flag -m and a message that describes the
changes being made.

git branch -M main

This commands renames our current branch to main. By default, git names
the default branch master. Github changes their default name to main instead.
This is an optional step. You can name your main or master branch whatever
you’d like.

git remote add origin https://github.com/shell-samurai/

my_first_repo.git
This command adds a new remote repository (Github.com) as the name
origin. This is the connection between your local git repository and the remote
repository. This will allow us to later push our changes up and pull changes
down.

git push -u origin main

Finally, this command tells git we’d like to upload our local commits to our
remote repository, origin. The -u flag is shorthand for —set-upstream option.
origin main specifies the origin upstream we defined earlier and main is the
branch we’re pushing to on that repository.

Now that we know how to create a repository, let’s actually do it! Create a
new folder anywhere on your system. You can call it whatever you want, but I
like to name the folder the same as my repository, my_first_repo in this case. I’ll
put my folder under my home directory and then a projects folder:

207
T.me/nettrain
 stetson@linux_rocks:~$ mkdir -p projects/my_first_repo

I provide the -p flag here because the projects folder didn’t exist yet and I
wanted to create both it and the my_first_repo folder. Go ahead and cd into
your new folder.

Configuring Git
We’re almost ready to commit our changes, but git needs a little more setup.
First, we’ll set our git email and name:

stetson@linux_rocks:~$ git config --global user.email

"[email protected]"
stetson@linux_rocks:~$ git config --global user.name "Stetson"

Before we push our code, Git wants us to authenticate ourselves so it knows

that the account we just made is ours. There are multiple ways to authenticate to
Github including through SSH and a GitHub cli tool called GH. Github
documents some of those methods here.
For our use-case, I think a Personal Access Token is the quickest and easiest
way to get setup. To create one, click on your user profile in the top right corner >
Then Click Settings.

208
T.me/nettrain
Click Developer Settings on the left-hand pane on the bottom.

209
T.me/nettrain
Finally, click tokens (classic):

210
T.me/nettrain
 Now click Generate Token, Generate new token (classic)

Now, give the access token a name and expiration date. Github will ask what
permissions the key should have. I’ll select everything under repo and

211
T.me/nettrain
write:packages. The other permissions shouldn’t be necessary. Scroll to the
bottom and click generate token.

Copy the token and keep it somewhere safe! Github won’t show it to us again
but you can always generate another. This token will act as our password to
authenticate from the command-line.
Here’s mine with a few characters censored in-case I forget to delete this
token and some devious reader steals it!

212
T.me/nettrain
 Back to our command-line, run the previous commands we copied from the
repository page. Those commands will add a README.md, commit it and push
it up to Github. Either type these by hand to get some extra practice or copy
paste them into your shell. When you get to the step to git push -u origin
main, git will ask for your username and password. Enter your username and
then enter the personal access token as the password.
If all went well, you should see your code being uploaded to the remote
origin we setup:

stetson@linux_rocks:~/projects/my_first_repo$ git push -u origin

main
Username for 'https://github.com': shell-samurai
Password for 'https://[email protected]':
Enumerating objects: 3, done.
Counting objects: 100% (3/3), done.
Writing objects: 100% (3/3), 234 bytes | 234.00 KiB/s, done.
Total 3 (delta 0), reused 0 (delta 0)
To https://github.com/shell-samurai/my_first_repo.git
* [new branch] main -> main
Branch 'main' set up to track remote branch 'main' from 'origin'.

Back in your browser window, refresh the repository we created and you’ll
see our first file, README.md listed along with the text inside it displayed:

213
T.me/nettrain
 Let’s get some more practice in. Go back to your command line now and edit
the README.md file and push it back up to Github. If you want to get fancy,
research Markdown formatting and implement some styling in your file!

Once again, here’s a reminder on how to push to Github once you’ve made
your changes:

stetson@linux_rocks:~/projects/my_first_repo$ git add README.md

Add our file, README.md to staging to tell git we’re about to add it to a
commit.

stetson@linux_rocks:~/projects/my_first_repo$ git commit -m

‘Updated readme.md’

Create a new commit and give it a message.

stetson@linux_rocks:~/projects/my_first_repo$ git push

Username for 'https://github.com': shell-samurai
Password for 'https://[email protected]':
Enumerating objects: 5, done.

214
T.me/nettrain
 Counting objects: 100% (5/5), done.
Delta compression using up to 4 threads
Compressing objects: 100% (3/3), done.
Writing objects: 100% (3/3), 362 bytes | 362.00 KiB/s, done.
Total 3 (delta 1), reused 0 (delta 0)
remote: Resolving deltas: 100% (1/1), completed with 1 local
object.
To https://github.com/shell-samurai/my_first_repo.git
889beec..a8d6010 main -> main

Push the code! Note that we don’t have to set the upstream as origin and
branch as main again since we already did that previously. Entering our personal
access token every time is kind of a pain though…I recommend setting up one of
the other authentication methods like SSH or using the gh client if you’re going
to be using git more than for just this tutorial.

Pulling Code
Let’s say that someone else has modified our code base and we want to get
those changes to our local system. How can we do that?
First, let’s make some changes we don’t have on our system. Go to your
README.md file and click the pencil icon to edit the markdown file in your
browser.

Then, make a change, and commit it:

215
T.me/nettrain
Back to your shell, run git pull:

stetson@linux_rocks:~/projects/my_first_repo$ git pull

remote: Enumerating objects: 5, done.
remote: Counting objects: 100% (5/5), done.
remote: Compressing objects: 100% (3/3), done.
remote: Total 3 (delta 1), reused 0 (delta 0), pack-reused 0
Unpacking objects: 100% (3/3), 718 bytes | 143.00 KiB/s, done.
From https://github.com/shell-samurai/my_first_repo
a8d6010..7b3821e main -> origin/main
Updating a8d6010..7b3821e
Fast-forward
README.md | 2 ++

216
T.me/nettrain
 1 file changed, 2 insertions(+)

If you examine the README.md file, you’ll see that git has pulled the changes
down to your local codebase! Nice!
Want to see the repository we worked on together? It’s available here:

https://github.com/shell-samurai/my_first_repo

Congrats on Gitting Good at Git

If you’ve made it through this section, congrats! Git isn’t easy to understand
for beginners but I wanted to cover the bare minimum because I think it’s
important to get exposure to. Many SysAdmin roles are evolving into DevOps
and Cloud Administrator roles, which definitely have a pre-requisite of knowing
the absolute basics of Git.
There’s much we didn’t cover here, like branching, pull requests and more.
For extra credit, I implore you to dive deeper into git and learn more features.
For now, this primer should git (haha) your hands dirty!

217
T.me/nettrain
 Interview Questions and Scenarios

Although this book is primarily a technical one focused on configurations and

kernels and other pieces of the Linux operating system, people do happen to be a
big part of that system! This section offers some advice and guidance on
interview questions I’ve encountered in the real-world. The first few sections are
about conversational style interviews and the Common Linux Interview
questions section are cut and dry questions about pieces of the Linux os.

Scenario Based Questions

One type of interview that’s becoming popular are “real-world” scenarios or
labs that model a problem you might run into while on the job. Recently, I had an
interviewer that had setup a Linux system in the browser and had me share my
screen while I worked on the system. The root issue ended up being that the disk
was filling up with log files and I had to find a rogue process, kill it and modify it
so that the disk space didn’t fill up all the way.
In these types of interviews, I find it’s best to talk aloud and narrate my
thought process and the actions I’ll be taking next. For example, “Ok, so the disk
is filling up, I want to find which files are the biggest so I’ll navigate to /var/log
and list the files and their sizes.” “Now I’ll look for the logging process with ps”.
It feels really strange at first, but narrating like this can help the interviewer to
understand your thought process and guide you in the right direction. Contrary
to popular belief, good interviewers want to help you succeed.

Quiz Type Questions or Discussion Questions

Another type of interview question are standard questions like — “How can I
find a file?”. In my opinion, some of these questions can get a bit too “quiz-ey”. If
you find an interviewer pummeling you with abstract questions that are
something you would usually just Google, I wouldn’t worry too much. That type
of company is probably not worth working for.
There’s also definitely a difference between abstract trivia questions and

218
T.me/nettrain
questions like “Describe how the Linux boot process works”. When interviewers
blindly toss questions at you that are abstract, it’s usually a red flag for me.

Situational Discussion Interview Questions

One of the most common interview questions I’ve gotten is formatted as “tell
me about a time when you…”. It might come up either as discussion from your
resume or something you mention in passing. If you put on your resume
“migrated a legacy Redis server cluster to the cloud”, be prepared to talk about
that project.
If you’re not already familiar with the acronym STAR, it can be very helpful
for navigating these questions. STAR stands for Situation, Task, Action, Result.

Let me give an example:

Tell me about a time when you faced a difficult technical problem?

Situation:
Well, my team maintained a legacy server that was constantly experiencing
downtime issues and causing all sorts of trouble.

Task:
I was tasked with taking the lead on the project, planning the migration and
executing on it.

Action:
I put together a task force of 3 people, met with them weekly, drafted a
maintenance plan, led the migration and moved the application to the cloud.

Result:
As a result, the application’s stability was improved! The move from the old
hardware was a success. The company application is much more stable now. My
supervisor commended me in my annual review and I was able to build
leadership skills and rapport with my team.

219
T.me/nettrain
 The STAR method helps you think about and provide a clear, structured
response that sell your skills and experience. It also allows you to show how you
worked well with a team in the past.
Hopefully you have some examples where projects went well for you. Maybe
you can’t think of a project that went well. That’s alright, if you’re able to reflect
on what went wrong and what could go better next time. No matter what, be
sure to highlight the successful parts and reflect on what could have gone better.
STAR is just one component of the interview process but it’s a tactic I have
used over and over again along with millions of other professionals who want to
sell their skillset to organizations. Use it.

Common Linux Interview Questions

Now, let’s dive into some common Linux interview questions you might get.
Some of these might fit into the quiz-show type questions I said I hated, but I
think they’re still important to know. Some times you just have to play the game.
My advice for all of these questions is to not just simply memorize and answer
and repeat it back like a robot. Truly understand the concepts behind them and
you’ll have far less trouble remembering obscure details.
Many of these questions will be ones you already know the answer to if
you’ve read Shell Samurai. Consider this section a reference.

What is Linux?
Linux is an open-source, Unix-like operating system developed in the 1990s
by Linus Torvalds. It’s widely used in servers, supercomputers, mobile devices,
and other embedded systems.

What is a shell in Linux?

A shell is a command-line interface that provides users with access to the
operating system's services and tools. The shell interprets or takes user input and
executes commands on behalf of the user. Some examples of common shells in
Linux are Bash, Zsh, Ksh, and Tcsh. Bash is the most common and is installed on

220
T.me/nettrain
most Linux distributions by default.

What is the difference between Linux and UNIX?

Linux is a Unix-like operating system that was developed independently of
UNIX, however it shares many of the same design principles and features. UNIX
is a proprietary operating system that was developed in the 60s and is still used
today.

What are the different types of Linux distributions available?

There are tons of Linux distributions, including Debian, Ubuntu, Red Hat
Enterprise Linux, CentOS, Fedora, Arch Linux, and more. Each distribution has
its own set of features, package management system, and community support.

What are the basic components of a Linux system?

The basic components of a Linux system include the kernel, shell, utilities, file
system, and user interface. The kernel manages system resources and provides
services to applications. The shell provides users with an interface to the system
and utilities provide tools for managing the system. The file system is used for
storing and managing data.

What does the root account do in Linux?

The root account is the default administrator account in Linux. It has full
access to all system resources and can perform any operation on the system. It is
used to perform system-level tasks that require escalated privileges, like
installing software and packages, configuring settings, and managing users and
groups.

What is sudo? How does it work?

Sudo is a command that allows users to execute commands with elevated
privileges, similar to root. It’s used to delegate admin tasks to non-root users and
limit the risks associated with using the root account. Sudo works by allowing
users to execute specific commands as the root user or another user with elevated
privileges. You can use sudo by prefacing any command with sudo.

221
T.me/nettrain
What is the difference between a process and a thread in Linux?
A process is an instance of a program that is executing on the system. It has its
own memory space and resources, and it can communicate with other processes
through inter-process communication mechanisms. A thread is a lightweight
process that shares the same memory space as its parent process. It allows for
concurrent execution of multiple tasks within a single process.

What is a daemon? How does it work?

A daemon is a background process that runs continuously on the system and
performs specific tasks, such as handling network connections, monitoring
system events, or managing hardware devices. Daemons typically run with
elevated privileges and do not have an interactive user interface.
What are some common Linux commands? (My dear reader, you should
know many of these by now)
Some common Linux commands include ls which lists directory contents, cd
to change directories, pwd will print your working directory. Then you have
mkdir which creates directories, rm removes file or directories, while cp can copy
files. If we wanted to move files we’d use mv, but we could also rename files with
mv. Touch will create an empty file. Cat displays files or concatenates them. Grep
is a handy tool for searching for patterns in files. Finally ps can display process
statuses!

Author’s Obvious tip: You probably won’t fire all of these commands off from
memory, but I’d be surprised if you did!

How can you generate a private and public keypair?

You can generate a private and public keypair using the ssh-keygen
command. The command generates a private key and stores it in the user's home
directory. The corresponding public key is generated and can be transferred to
servers for authentication..

222
T.me/nettrain
Advanced Linux Interview Questions
What is systemd and what does it do?
systemd is a system and service manager used to control the startup process
and manage system services in Linux. It provides a consistent and unified way to
start, stop, and monitor services, and it can be used to manage system resources
like timers and sockets. systemd is used by many modern Linux distributions,
including Red Hat, Fedora, Debian, and Ubuntu.

What is a kernel module?

A kernel module is a piece of code that can be dynamically loaded into the
Linux kernel to add or extend functionality. It is typically used to provide device
drivers, file systems, or network protocols. Kernel modules can be loaded or
unloaded at runtime, without the need to reboot the system.

What does the strace command do?

Strace is used to trace system calls and signals made by a process in Linux. It
can be used to diagnose issues with an application, analyze behavior, or debug
issues with system calls.

What does the lsof command do in Linux?

lsof or list open files is used to list open files and network connections in
Linux. It can be used to identify which processes are using a file, determine
which network ports are in use, or diagnose issues with file access or network
connectivity.

What does ssh do?

Ssh is used to securely connect to a remote system over the network. It
provides a secure shell session that allows users to execute commands, transfer
files, or run remote applications.

How do private and public keys work?

Private and public keys are a pair of keys used in asymmetric cryptography to
encrypt and decrypt messages. The private key is kept secret and used to decrypt

223
T.me/nettrain
messages encrypted with the corresponding public key, which can be freely
distributed. The security of this system is based on the mathematical relationship
between the two keys, making it impossible to derive the private key from the
public key. Private and public keys can be used to secure communication over
untrusted channels and for authentication.

What is the difference between hard and soft links in Linux?

Hard links and soft links (or symbolic links) are two types of links that can be
created between files in Linux. A hard link is a direct link to a file that points to
the same data blocks as the original file. It has the same file permissions and
ownership as the original file, and it cannot be created for directories. A soft link,
on the other hand, is a pointer to a file that points to the original file by its path
name. It has its own file permissions and ownership, and it can be created for
both files and directories.

What is the purpose of the /proc filesystem?

Everything in Linux is a file! The /proc filesystem is a virtual file system that
provides a view into the system's processes and kernel data structures. It allows
users to read or modify kernel variables, system information, and process status
in real-time. It does not store any data on disk, but rather provides a dynamic
interface to the kernel's internal data structures.

What is cron and crontab used for?

Cron is used to schedule recurring tasks in Linux. It allows users to specify a
set of commands or scripts to be performed at specific intervals, like every hour,
every day, or every week. The crontab file uses a syntax that specifies the timing
and frequency of the job, as well as the command to be executed.

What is the purpose of the iptables command in Linux?

The iptables command is used to set up and manage the firewall rules in
Linux. It allows users to specify rules to filter, block, or allow network traffic
based on predefined rules.
Iptables works by configuring chains of rules, where each rule specifies the
type of traffic to match and what action to take on the matched traffic. The rules

224
T.me/nettrain
can be set up to allow or block traffic based on the source or destination IP
address, port number, protocol, and other factors.
iptables can also be used to forward traffic to other hosts or machines on the
network.
What is a container and how does it work?
A container is a lightweight, isolated runtime environment that allows
multiple applications or services to run on a single host without interfering with
each other. It achieves this isolation by using Linux namespaces to create
separate network, process, and file system environments, and by using cgroups
to limit resource usage. Containers can be created and managed using tools like
Docker and Kubernetes.

225
T.me/nettrain
 Outro

Shell Samurai
Master the Linux Command Line

Outro and Additional Resources

Thank You Again
We are finally to the end! Well done! I hope you enjoyed Shell Samurai! I want
to thank you again for purchasing and sticking through this book. Maybe you
didn’t even make it all the way through, perhaps you picked through the
interesting parts and you’ll revisit others later. That’s one of my favorite ways to
read a book.
Either way, I hope that you learned something new about Linux.
The world of open-source and Linux has completely changed the world and
isn’t slowing down any time soon. Knowing how to use Linux can help you to
advance in your career or start a new one.

I’d love to hear your feedback on Shell Samurai! You can reach out at
[email protected] or join our Private Discord Server.

226
T.me/nettrain