Quick reference
Files
ls -lah
cd /var/log
cp -a src/ dst/
mv old.txt new.txt
find / -name '*.log' 2>/dev/nullProcesses
ps aux | grep nginx
top -o %CPU
kill -TERM <pid>
systemctl status nginx
journalctl -u nginx -fDisk
df -h
du -sh /var/log/*
lsblk -f
free -h
mount | column -tNetwork
ss -tlnp
ip -br addr
curl -I https://example.com
dig example.com +short
tcpdump -i any port 443Explained
Files. ls -lah gives you the listing you actually want: hidden files, human sizes, permissions. cp -a preserves the things you forget: timestamps, symlinks, ownership. find with -name is still the fastest way to answer "where is that log".
Processes. ps aux | grep is crude and works. top -o %CPU answers "what is eating the box" without a dashboard. systemctl status and journalctl -u are the pair you use on any systemd host, and kill -TERM asks politely before -9 ends the argument.
Disk. df -h shows the filesystem, du -sh shows the folder, and the gap between them is usually a deleted-but-open file. lsblk -f shows what is actually mounted where. free -h is the first thing to check when a process disappears at night.
Network. ss -tlnp shows what is listening, ip -br addr is the short address view, curl -I proves the service answers, dig answers DNS, and tcpdump is the last resort that settles arguments.
The lesson: a cheat sheet is not a reference. It is the set of commands you can recall under pressure.
In one line
Know the command, skip the search. Five per area beats fifty you cannot recall.
A first look
uptime; df -h; free -hMore notes: Linux
Open in the notes app: search, filters, copy This note on GitHub