Linux Tutorial (5): Process Management dan systemd

Setiap program yang berjalan di Linux adalah proses, instance program dengan PID unik, memory space sendiri, dan lifecycle yang dikelola kernel. Di server modern, systemd mengontrol hampir semua service.

Anatomi Proses

# Proses saat ini
echo $$                    # PID shell ini
ps -p $$                   # detail proses shell
 
# Semua proses (BSD style)
ps aux
 
# Tree view: lihat parent-child
ps auxf
pstree -p
 
# Filter
ps aux | grep nginx
pgrep -a nginx             # PID + command line
pidof nginx

Kolom penting ps aux:

USER: owner proses
PID: Process ID
%CPU / %MEM: resource usage
STAT: state (R=running, S=sleeping, Z=zombie, D=uninterruptible)
START / TIME: kapan mulai, total CPU time

Monitoring Real-time

# top: interactive (q=quit, M=sort memory, P=sort CPU, k=kill)
top
 
# htop: lebih user-friendly
htop
 
# Snapshot sekali
top -bn1 | head -20
 
# Monitor proses spesifik
top -p $(pidof nginx)

Sinyal dan kill

# Graceful stop (SIGTERM = 15)
kill 1234
kill -15 1234
 
# Force kill (SIGKILL = 9): tidak bisa ditangkap/diabaikan
kill -9 1234
 
# Kill by name
killall nginx
pkill -f "python app.py"
 
# Daftar sinyal
kill -l

Sinyal	Nomor	Gunakan saat
SIGTERM	15	Stop normal, proses bisa cleanup
SIGKILL	9	Proses hang/zombie, last resort
SIGHUP	1	Reload config (nginx, syslog)
SIGSTOP	19	Pause proses
SIGCONT	18	Lanjutkan proses paused

Aturan: selalu coba SIGTERM dulu, tunggu 5–10 detik, baru SIGKILL.

Job Control

# Jalankan di background
sleep 300 &
jobs
 
# Suspend proses foreground
# Ctrl + Z
 
# Lanjutkan di background
bg %1
 
# Bawa ke foreground
fg %1
 
# nohup: proses tetap jalan setelah logout
nohup ./long-running.sh > output.log 2>&1 &
disown

Prioritas: nice dan renice

# Jalankan dengan prioritas rendah (nice 19 = paling rendah)
nice -n 19 ./backup.sh
 
# Ubah prioritas proses yang sudah berjalan
sudo renice -n -5 -p 1234   # lebih tinggi (butuh root untuk negative nice)

Range nice: -20 (prioritas tertinggi) hingga 19 (terendah). Default: 0.

systemd: Unit Files

systemd mengelola units: service, mount, timer, socket, dll.

# Status service
systemctl status nginx
 
# Start / stop / restart / reload
sudo systemctl start nginx
sudo systemctl stop nginx
sudo systemctl restart nginx
sudo systemctl reload nginx
 
# Enable / disable saat boot
sudo systemctl enable nginx
sudo systemctl disable nginx
sudo systemctl is-enabled nginx
 
# List semua service
systemctl list-units --type=service
systemctl list-unit-files --type=service

Menulis Service Unit Custom

File: /etc/systemd/system/myapp.service

[Unit]
Description=My Application Server
After=network.target postgresql.service
Requires=postgresql.service
 
[Service]
Type=simple
User=appuser
Group=appuser
WorkingDirectory=/opt/myapp
ExecStart=/opt/myapp/bin/server --port 8080
Restart=on-failure
RestartSec=5
Environment=NODE_ENV=production
LimitNOFILE=65535
 
[Install]
WantedBy=multi-user.target

sudo systemctl daemon-reload
sudo systemctl enable --now myapp.service
systemctl status myapp.service

journalctl: Log systemd

# Log service spesifik
journalctl -u nginx
 
# Follow real-time
journalctl -u nginx -f
 
# Sejak boot terakhir
journalctl -b
 
# Rentang waktu
journalctl --since "2026-01-20 08:00" --until "2026-01-20 12:00"
journalctl -u myapp --since "1 hour ago"
 
# Prioritas error ke atas
journalctl -p err -b
 
# Export
journalctl -u nginx --since today > nginx-today.log

Timer: Cron Modern

# /etc/systemd/system/backup.timer
[Unit]
Description=Daily Backup Timer
 
[Timer]
OnCalendar=daily
Persistent=true
 
[Install]
WantedBy=timers.target

# /etc/systemd/system/backup.service
[Unit]
Description=Daily Backup
 
[Service]
Type=oneshot
ExecStart=/opt/scripts/backup.sh

sudo systemctl enable --now backup.timer
systemctl list-timers

Zombie Process

Zombie (STAT=Z) = proses sudah mati tapi entry masih di process table karena parent belum wait().

# Identifikasi
ps aux | awk '$8 ~ /Z/ {print}'
 
# Fix: restart parent process, atau kill parent (zombie di-adopt init)

Latihan Praktis

Jalankan sleep 600 &, monitor dengan jobs, hentikan dengan kill
Buat systemd service untuk script bash sederhana, enable saat boot
Simulasi crash: buat service dengan ExecStart=/bin/false, set Restart=on-failure, amati journalctl
Buat systemd timer untuk menjalankan script setiap 5 menit

Rangkuman

Process management dan systemd adalah skill inti sysadmin. Kuasai ps, sinyal, job control, systemctl, dan journalctl, alat ini dipakai setiap hari di production.