Linux Commands for Beginners: The Essential Terminal Guide With Examples

Most servers, containers and cloud machines run Linux, and they usually do not have a graphical desktop. That is why learning Linux commands is such a useful skill for developers and testers alike. You do not need to memorize hundreds. A few dozen will carry you a long way.

We tested the file, search and pipe examples in Git Bash on Windows, which provides the same core tools, and the output comes from that run. We did not run the commands for permissions, services, packages and remote access, because they depend on a real Linux system. Details can vary by distribution, so check your own system’s documentation.

Getting started safely

You do not need to install Linux to practise. Good options include:

• Windows: Windows Subsystem for Linux (WSL), or Git Bash for the basics.
• macOS: the built-in Terminal, whose commands are very similar, though some options differ.
• Any system: a container. Our Docker tutorial for beginners shows how to start one.

Practise in a throwaway folder, and remember that everything in Linux is case-sensitive: Notes.txt and notes.txt are different files.

Anatomy of a command

ls -l /var/log
^  ^  ^
|  |  argument: what to act on
|  option: how to behave
command: what to run

Short options can usually be combined, so ls -l -a and ls -la are the same. Long options use two dashes, as in --help. For help, run man ls for the full manual, or ls --help for a summary. Press q to leave a manual page.

Move around the file system

Press Tab to auto-complete file names, and the up arrow to bring back earlier commands. These two habits save more time than any other tip here. A file name beginning with a dot, such as .env, is hidden, and ls shows it only with -a.

Manage files and folders

Here is a short session from our run:

$ mkdir -p project/logs
$ cd project
$ touch notes.txt
$ ls
logs  notes.txt

$ cp notes.txt backup.txt
$ mv backup.txt old.txt
$ ls
logs  notes.txt  old.txt

$ rm old.txt
$ ls
logs  notes.txt

View file contents

For the next few examples, we created a small log file:

$ cat logs/app.log
ERROR db down
INFO started
ERROR disk full
INFO stopped

$ head -n 1 logs/app.log
ERROR db down

$ tail -n 1 logs/app.log
INFO stopped

$ wc -l logs/app.log
4 logs/app.log

tail -f is a daily tool for watching a live application log while you reproduce a bug. Press Ctrl+C to stop it.

Search with grep and find

grep searches inside files, and find searches for the files themselves:

$ grep ERROR logs/app.log
ERROR db down
ERROR disk full

$ grep -c ERROR logs/app.log
2

$ find . -name "*.log"
./logs/app.log

Pipes and redirection

This is the idea that makes the command line so powerful. Each tool does one small job, and you connect them:

$ echo "hello" > greeting.txt
$ echo "world" >> greeting.txt
$ cat greeting.txt
hello
world

Be careful with a single >, because it silently overwrites what was already in the file.

A real example: summarize a log

How many lines of each type are in our log? Combining four simple commands gives the answer:

$ cut -d' ' -f1 logs/app.log | sort | uniq -c
      2 ERROR
      2 INFO

Reading from left to right: cut takes the first space-separated word on each line, sort groups identical words together, and uniq -c counts each group. The sort step matters, because uniq only collapses neighbouring duplicates. Add | sort -nr to the end to list the most common first.

This kind of one-liner is a great help when you are investigating failures. It can quickly tell you whether an error happened twice or two thousand times, which is useful detail to put in a bug report.

File permissions

Run ls -l and you will see something like this:

-rwxr-xr-- 1 asha devs 512 Nov 1 10:00 deploy.sh

After the first character, which shows the file type, there are three groups of three characters, for the owner, the group and everybody else. Each can have r (read), w (write) and x (execute).

The numbers add up: read is 4, write is 2 and execute is 1. So 7 means all three, and 5 means read and execute. Avoid chmod 777, which gives everyone full access, as a quick fix for permission errors. Find the real cause instead. And use sudo carefully, since it removes the safety checks that normally protect the system.

Processes and system information

To install software, use your distribution’s package manager, such as sudo apt install curl on Debian and Ubuntu, or sudo dnf install curl on Fedora and RHEL.

Archives and remote work

$ tar -czf logs.tar.gz logs
$ tar -tzf logs.tar.gz
logs/
logs/app.log

In tar, c means create, z means gzip, f names the file and t lists the contents. To extract an archive, use tar -xzf logs.tar.gz. For working with other machines, these three are what you need:

These same commands are used inside Dockerfiles and CI/CD pipelines, so this knowledge pays off across DevOps work.

Common mistakes

• Using rm -rf without looking. Check your folder with pwd first, and be careful with spaces in paths.
• Overwriting with >. Use >> when you meant to append.
• Running everything with sudo. Use it only when it is required.
• Pasting commands you do not understand. Read them first, especially anything piping a download into a shell.
• Forgetting quotes. A file name with spaces, like "my file.txt", needs them.

Final thoughts

You do not learn the terminal by reading about it. Open a shell, make a practice folder and run each section’s commands. Then try to answer a real question using a pipe, like which error appears most often in a log. Start with navigation and files this week, add grep and pipes next, and the rest will follow as you need it.

Frequently asked questions

Keep learning