Cron expression to run every minute

Why every field is a wildcard

An asterisk means "match every value of this field". With all five set to * there is nothing left to filter on, so the job is due at the top of every minute. Cron has no seconds field, so this is the highest frequency the syntax can express — roughly 43,800 runs a month.

Using it

In a user crontab:

* * * * * /usr/local/bin/healthcheck.sh >> /var/log/healthcheck.log 2>&1

In a Kubernetes CronJob:

spec:
  schedule: "* * * * *"
  concurrencyPolicy: Forbid

Note concurrencyPolicy: Forbid. At one-minute intervals overlapping runs are a real risk, and this is the single most important setting to get right.

Guard against overlap

Plain cron will happily start a second copy of a job while the first is still running. At a one-minute cadence a job that occasionally takes ninety seconds will pile up until the machine falls over. Wrap the command in a lock:

* * * * * /usr/bin/flock -n /tmp/myjob.lock /usr/local/bin/myjob.sh

The -n flag makes flock exit immediately if the lock is held, so a slow run simply skips the next tick instead of queueing behind it.

When you probably want something else

• You need sub-minute timing. Cron cannot do it. Use a systemd timer with OnUnitActiveSec, or a long-running worker process.
• The job is only relevant during working hours. * 9-17 * * 1-5 cuts the run count by more than 80 per cent for the same coverage.
• The job polls an external API. One request a minute from every instance adds up fast; check the rate limit before committing to this schedule.
• The job is cheap but noisy. Every run emails its output to the crontab owner unless you redirect it. Set MAILTO="" at the top of the crontab or redirect explicitly, or you will generate 1,440 emails a day.

Variations