systemd Cheatsheet

Top commands.

Overview

systemd is the init system and service manager underneath every modern Linux distribution. Five primitive surfaces cover almost everything an operator does day-to-day: systemctl for service control, unit files for declarative config, dependency directives for startup ordering, timers for scheduled jobs, and resource limits for protection. Fluency in all five turns ad-hoc init scripts into reviewable service definitions.

• Service management via systemctl. start, stop, restart, status, enable. The verbs every Linux operator uses dozens of times a day.
• Unit files in /etc/systemd/system. Declarative service config that lives in source control alongside the application.
• Dependency directives. Wants, Requires, After, Before express startup ordering without shell hacks.
• Timers plus resource limits. systemd timers replace cron with traceable, journalled jobs; CPUQuota, MemoryMax, IOWeight contain runaway services.

The approach

Five commands carry most operational weight. Memorising them moves the team from cargo-culted shell snippets to confident service control.

• systemctl status app. Live status plus the last lines of journal output. The first command on every service issue.
• systemctl restart app. Clean restart that respects unit-file dependencies. Replaces ad-hoc kill -HUP patterns.
• systemctl enable --now app. Enable at boot and start immediately, in one command.
• systemctl daemon-reload plus list-timers. Reload after unit-file edits to avoid stale config; list-timers shows what is actually scheduled.

Why this compounds

Each unit file captures service knowledge in a reviewable format. Over the year, the team's service library becomes a working map of how the platform actually starts up; debugging starts from a definition, not a guess.

• Reproducible services. Unit files produce identical behaviour across hosts. Drift shrinks.
• Faster debugging. journalctl plus systemctl status answers most service questions in seconds.
• Resource protection. CPU, memory, and IO limits stop one bad service from taking the host down.
• Year-one investment, year-two habit. The first year builds fluency under pressure. By year two, writing a unit file is routine.