Module 1 — Linux Systems Administration

Goal: LPIC-1-equivalent working knowledge, learned entirely by doing on lab-box. By the end you can navigate, edit, manage users/permissions/processes/services/ packages/disks/logs, and write a small shell script — the daily bread of every cloud engineer.

All labs run on lab-box over SSH. Type everything.

1.1 The shell: your new dial tone

Theory (10 min): kernel vs shell vs terminal; why servers have no GUI; the prompt anatomy user@host:dir$.

Lab 1.1 — Finding your way

whoami                # who am I?
hostname              # which machine?
pwd                   # where am I?
ls                    # what's here?
ls -la                # everything, in detail
cd /                  # go to the root of the filesystem
ls
cd ~                  # back home
mkdir -p labs/module1
cd labs/module1

Explore the filesystem hierarchy — visit each and peek inside with ls:

Dir	What lives there	Telecoms analogy
`/etc`	configuration	the site folder with all the config sheets
`/var/log`	logs	the alarm/event log printer
`/home`	user files	your desk drawer
`/bin`, `/usr/bin`	programs	the tool van
`/tmp`	scratch space	the bench
`/proc`	live kernel/process info	live line-test readouts

Checkpoint: without notes: create ~/labs/module1/test/deep/dir in one command, create an empty file in it, then delete the whole tree. (mkdir -p, touch, rm -r.)

1.2 Files, viewing and editing

cat /etc/os-release              # dump a file
less /var/log/syslog             # page through (q to quit, / to search)
head -20 /etc/passwd
tail -f /var/log/syslog          # follow live — Ctrl-C to stop
cp /etc/hostname myhostname.txt
mv myhostname.txt hostname.bak
grep ubuntu /etc/passwd          # search inside files
find /etc -name "*.conf" | head  # find files by name
wc -l /etc/passwd                # count lines

Lab 1.2 — nano then vim

nano notes.txt — write three lines, save (Ctrl-O), exit (Ctrl-X).
vimtutor — complete lessons 1 and 2 (20 min, worth it: vim is on every server you will ever touch).
Survival vim: vim notes.txt → i to insert → type → Esc → :wq to save-quit, :q! to abandon.

1.3 Pipes and redirection — the superpower

ls /etc | wc -l                       # pipe: output of one → input of next
grep -c bash /etc/passwd
ps aux | grep ssh
history | tail -20
ls /etc > etclist.txt                 # redirect stdout to file
ls /nonexistent 2> errors.txt         # redirect stderr
ls /etc /nonexistent > all.txt 2>&1   # both
sort /etc/passwd | cut -d: -f1 | head # chains

Checkpoint: one command line that counts how many .conf files exist under /etc (hint: find /etc -name "*.conf" 2>/dev/null | wc -l — why the 2>/dev/null?).

1.4 Users, groups, permissions, sudo

Theory: every file has owner/group/other × read/write/execute. root is god; sudo borrows god powers per-command and is logged.

id
ls -l /etc/shadow            # note: only root can read it
sudo tail -3 /etc/shadow
sudo adduser trainee
sudo usermod -aG sudo trainee
su - trainee                 # become them (their password)
exit

Lab 1.4 — Permission surgery

touch secret.txt
chmod 600 secret.txt && ls -l secret.txt      # rw-------
chmod 644 secret.txt && ls -l secret.txt      # rw-r--r--
chmod +x script.sh                            # make executable
sudo chown trainee:trainee secret.txt

Decode octal: 7=rwx, 6=rw-, 5=r-x, 4=r--. Checkpoint: what is 755 and where have you already seen it today? (ls -l /usr/bin | head.)

1.5 Processes and services (systemd)

ps aux | head
top                          # q to quit; note load average, memory
htop                         # nicer (pre-installed on lab-box)
sleep 300 &                  # background job
jobs
kill %1
systemctl status ssh         # the service keeping you connected
systemctl list-units --type=service --state=running
sudo systemctl restart ssh   # yes, mid-session — what happens to you? why not?
journalctl -u ssh -n 20      # logs for one service

Telecoms mapping: systemd is the exchange's supervisory system — it starts services at boot, restarts them when they die, and keeps their event logs.

1.6 Packages, disks, networking basics

sudo apt update && sudo apt install -y tree cowsay
tree ~/labs | head
df -h                        # disk usage by filesystem
du -sh /var/log
free -h                      # memory
ip a                         # interfaces + addresses
ip route                     # routing table — find the default gateway
ss -tlnp                     # listening TCP ports (you'll use this constantly)
ping -c3 1.1.1.1
curl -I https://example.com  # HTTP HEAD — first taste of module 2

1.7 Shell scripting — automate like an engineer

Lab 1.7 — Your first real script

vim ~/labs/module1/healthcheck.sh:

#!/usr/bin/env bash
# healthcheck.sh — first automation. Run: ./healthcheck.sh
set -euo pipefail

echo "=== Health report for $(hostname) at $(date) ==="
echo "-- Uptime/load:"; uptime
echo "-- Disk:";        df -h / | tail -1
echo "-- Memory:";      free -h | awk '/Mem/ {print $3 " used of " $2}'
echo "-- Top 3 CPU processes:"
ps aux --sort=-%cpu | head -4 | tail -3

if df -h / | tail -1 | awk '{gsub("%","",$5); exit ($5>80)?1:0}'; then
  echo "DISK OK"
else
  echo "WARNING: root disk above 80%"
fi

chmod +x healthcheck.sh
./healthcheck.sh

Then schedule it:

crontab -e     # add:  */5 * * * *  ~/labs/module1/healthcheck.sh >> ~/health.log 2>&1
tail -f ~/health.log

You have just built a monitoring agent. In module 7 you'll replace it with Prometheus and see why the industrial version is better.

1.8 Sabotage session (instructor)

While the student is away, do two or three of these on lab-box; student diagnoses and fixes, narrating their reasoning out loud (field-engineer style):

chmod 000 their healthcheck script.
Fill /tmp with a big file (fallocate -l 6G /tmp/big); watch them find it with df/du.
Stop a service they now rely on: sudo systemctl stop cron.
Add a nonsense line to ~/.bashrc that prints a fake error at login.
Rename ls in their PATH: alias ls='echo command not found' in .bashrc.

LPIC-1 coverage map

This module covers, hands-on, the core of LPIC-1 topics 101–110: system architecture (lightly), Linux installation/package management (apt), GNU & Unix commands, devices & filesystems, shells & scripting, user interfaces (skipped GUIs deliberately), administrative tasks (users, cron), essential services (systemd, logging), networking fundamentals, and security basics (permissions, sudo, ssh). For formal certification, add reading on: RPM/dnf (Red Hat side), runlevels history, localisation, and printing (rarely used in cloud work).

Quiz

1. What does chmod 640 file allow, exactly?

Owner: read+write. Group: read. Others: nothing.

2. A service is "down". Give the three commands you run first.

systemctl status NAME, journalctl -u NAME -n 50, ss -tlnp (is anything listening on its port?).

3. Difference between | and >?

| feeds output into another command; > writes output into a file.

4. Why is there no GUI on servers?

Wasted resources, larger attack surface, and everything must be scriptable/automatable — GUIs can't be version-controlled or run by pipelines.