Linux & software2 min read

Schedule tasks with cron jobs and systemd timers

Run scripts on a schedule: the cron syntax explained clearly, typical examples, logging output and errors and when a systemd timer is the better choice.

Backups every night, cleanup every week, a script every five minutes: Linux has two tools for recurring tasks. Cron is simple and known everywhere; systemd timers take a little more effort but catch up on missed runs and log everything.

Cron: edit the crontab

crontab -e          # your own tasks
sudo crontab -e     # root's tasks
crontab -l          # list

The cron syntax

Every line has five time fields followed by the command:

┌ minute (0-59)
│ ┌ hour (0-23)
│ │ ┌ day of month (1-31)
│ │ │ ┌ month (1-12)
│ │ │ │ ┌ weekday (0-7, 0 and 7 = Sunday)
│ │ │ │ │
* * * * * command
Line Meaning
*/5 * * * * every 5 minutes
0 * * * * every full hour
30 3 * * * every day at 3:30
0 4 * * 1 every Monday at 4:00
0 2 1 * * on the first of every month at 2:00
@reboot once after every reboot

The time follows the server's time zone. timedatectl shows it; sudo timedatectl set-timezone Europe/Berlin changes it.

Example: nightly backup

30 3 * * * /usr/local/bin/restic-backup.sh >> /var/log/restic-backup.log 2>&1

>> … 2>&1 writes normal output and errors to a file. Without it they vanish and you never notice that something goes wrong. The script in the example comes from Backups with restic.

Typical pitfalls

  • Cron does not know your environment. PATH is very short. Write programs with their full path (/usr/bin/php instead of php) or set PATH=/usr/local/bin:/usr/bin:/bin at the top of the crontab.
  • % has a special meaning in the crontab. Write \%, for example date +\%F.
  • The script works by hand but not via cron: Usually paths or permissions. Test it once the way cron calls it: env -i /bin/sh -c '/usr/local/bin/myscript.sh'.

systemd timers: the more robust choice

A timer consists of two files: a service that does the work and a timer that starts it.

sudo nano /etc/systemd/system/restic-backup.service
[Unit]
Description=Nightly restic backup

[Service]
Type=oneshot
ExecStart=/usr/local/bin/restic-backup.sh
sudo nano /etc/systemd/system/restic-backup.timer
[Unit]
Description=restic backup every night

[Timer]
OnCalendar=*-*-* 03:30:00
Persistent=true
RandomizedDelaySec=15m

[Install]
WantedBy=timers.target
sudo systemctl daemon-reload
sudo systemctl enable --now restic-backup.timer
systemctl list-timers

The advantages over cron:

Cron systemd timer
Server was off at the scheduled time run is skipped catches up (Persistent=true)
Output and errors only with redirection automatically in the journal
Is the script still running? not visible systemctl status restic-backup
Start by hand type the command sudo systemctl start restic-backup
Runs twice at the same time possible never

journalctl -u restic-backup -n 50 shows the log of the last run.

OnCalendar examples

Value Meaning
hourly every full hour
daily every day at midnight
Mon *-*-* 04:00:00 Mondays at 4:00
*:0/15 every 15 minutes
*-*-01 02:00:00 on the first of the month at 2:00

Check whether an expression is right with systemd-analyze calendar "Mon *-*-* 04:00:00".

Get notified on failure

So that a failed backup does not stay hidden for weeks, let the script ping a heartbeat monitor at the end, for example with the Statusbot. If the ping does not arrive, you get a message.

Written by TryNxt team

We build and run TryNxt ourselves and write down here what we learn along the way.