Skip to content
Cron schedules

Cron job every hour: expression and example

The cron expression for every hour, on the hour is 0 * * * *. Put it at the start of a crontab line, followed by the command. Below: what each of the five fields does, a sample line you can copy, the mistakes people make with this schedule, and how to get told when the job doesn't run.

The expression

A cron expression has five fields separated by spaces: minute, hour, day of month, month, day of week. An asterisk means "every"; a number means "only this one"; */n means "every nth"; a-b is a range and a,b is a list. For every hour, on the hour the fields are:

FieldValueMeaning
Minute (0-59)0minute 0
Hour (0-23)*every hour
Day of month (1-31)*every day
Month (1-12)*every month
Day of week (0-7, Sunday is 0 or 7)*every day of the week

Example crontab line

Open the crontab with crontab -e and add one line: the expression, a space, then the command exactly as you'd type it in a shell. Aggregating stats, expiring sessions, fetching a feed, rebuilding a cache that's allowed to be an hour stale. A typical line looks like this:

0 * * * * /usr/bin/python3 /srv/app/aggregate.py >> /var/log/aggregate.log 2>&1

The >> file 2>&1 at the end sends both normal output and errors to a log file. Without it, cron tries to email the output to the user, which usually means it's lost.

Gotchas

  • On the hour is when everyone else's hourly jobs run too, including the system's own. If the server is slow at :00, pick another minute (17 * * * *) and the job gets the box to itself.
  • * * * * * is not hourly; it's every minute. Hourly needs a number in the minute field.
  • Daylight-saving changes: in a timezone that shifts clocks, one hour a year happens twice and one not at all. Hourly jobs on a UTC server avoid that.
  • Timezone: cron uses the server's system timezone, which on most cloud servers is UTC. Run date on the server to see what "now" means to cron.
  • Environment: cron runs with a minimal PATH and no shell profile. Use full paths (/usr/bin/python3, not python3) and set any variables at the top of the crontab.
  • Output: if a job prints anything and there's no redirect or MAILTO, cron tries to send mail. Set [email protected] at the top of the crontab or redirect output to a log.
  • Silence: cron does not tell you when a job fails or never starts. A job that errors every run looks exactly like a job that runs fine, unless something is watching for it.

Know when it didn't run

In short: Append && curl -fsS https://heyuptime.com/ping/<token> to the line, create a heartbeat monitor with a 1 hour period and 15 minutes grace, and you're alerted when the job is late.

A heartbeat monitor in heyUptime is a URL your job calls when it finishes. Because of the &&, the ping is only sent when the command exits successfully, so a crashing job and a job that never started look the same: no ping. The watched line for this schedule is 0 * * * * /usr/bin/python3 /srv/app/aggregate.py && curl -fsS https://heyuptime.com/ping/<token>.

Set the heartbeat's period to 1 hour and the grace to 15 minutes. If a ping doesn't arrive within period plus grace, you get an alert by email, Slack, Discord, Teams or Telegram that says when the job was last seen.

To try it before signing up, the free cron job tester gives you a temporary URL for 24 hours; add it to the line and watch the ping arrive.

Questions

What is the cron expression for every hour?

0 * * * *: at minute 0 of every hour, 24 times a day. @hourly is a shorthand most cron implementations accept.

How do I run every 2 hours instead?

0 */2 * * *, which runs at 00:00, 02:00, 04:00 and so on. See the every-2-hours page.

What's the difference between 0 * * * * and * * * * *?

The first runs once an hour, at minute 0. The second runs every minute. The minute field decides.

Watch this job for free.

Create a heartbeat with a 1 hour period, add one curl to the crontab line, and get told the first time it doesn't run. 14-day trial, no card.

Watch this job free

Updated October 8, 2026.