The Linux filesystem and everyday commands

Why Linux matters in DevOps

Linux is a free, open-source operating system. Most cloud servers run it, nearly every container image is built on it, and the machines that run CI pipelines (called runners or agents) are usually Linux too. Kubernetes nodes run Linux. If you work in DevOps, you will spend a lot of time typing commands on Linux machines you never see physically.

Distributions

A distribution (distro) is Linux packaged with tools, a package manager and a release schedule. You will meet these most often:

Rocky Linux and AlmaLinux are free rebuilds of Red Hat Enterprise Linux (RHEL). Ubuntu LTS (long-term support) versions come out every two years and get years of security updates, which is why servers use them.

The shell and the terminal

The terminal is the window where you type. The shell is the program inside it that reads your command, runs it and prints the result. The most common shell is bash; zsh is popular on laptops. The prompt, such as deploy@web1:~$, tells you the user (deploy), the machine (web1) and the current folder (~ means your home folder).

One tree, starting at /

Windows has drives like C: and D:. Linux has a single tree that starts at / (called root). Disks and other devices are attached, or mounted, somewhere inside that tree.

Absolute and relative paths

An absolute path starts with / and works from anywhere: /var/log/nginx/access.log. A relative path starts from the current folder: if you are in /var/log, then nginx/access.log points to the same file. Two special names help: . means "this folder" and .. means "the parent folder". So from /var/log/nginx, the path ../syslog means /var/log/syslog.

Everyday commands

pwd                          # print the current folder
ls -la                       # list all files, including hidden ones, with details
cd /var/log                  # go to a folder (cd alone goes home, cd - goes back)
mkdir -p /opt/shop/releases  # make a folder and any missing parents
cp app.conf app.conf.bak     # copy a file
cp -r config/ backup/        # copy a folder and everything inside it
mv old.log archive.log       # move or rename
rm notes.txt                 # delete a file
rm -r old-build/             # delete a folder and its contents
touch deploy.log             # create an empty file (or update its time)

Reading files

cat /etc/os-release          # print a short file (which distro is this?)
less /var/log/syslog         # scroll a long file: q quits, /word searches
head -n 20 app.log           # first 20 lines
tail -n 50 app.log           # last 50 lines
tail -f app.log              # follow: keep printing new lines as they arrive
find /var/log -name "*.log" -mtime +7    # .log files older than 7 days
find / -type f -size +500M 2>/dev/null   # files bigger than 500 MB

Editing files on a server

Servers have no graphical editor, so learn one terminal editor. nano is simple: type, then Ctrl+O to save and Ctrl+X to exit. vim is on almost every machine: press i to insert text, Esc to stop inserting, then type :wq and Enter to save and quit, or :q! to quit without saving.

Installing software

sudo apt update                # Ubuntu/Debian: refresh the package list
sudo apt install -y nginx      # install nginx without asking yes/no
sudo dnf install -y nginx      # RHEL/Rocky/Alma
apk add --no-cache curl        # Alpine, often inside a Dockerfile