The heartbeat monitor is a dead-man's switch for scheduled jobs like cron and backups. This article focuses on how to set one up and the fields involved.
How it works (inverted)
Unlike other monitors, a heartbeat waits to be contacted. Your job pings a URL on a schedule – if the ping does not arrive, the monitor alarms. For the thinking behind this, see the separate concept article on heartbeats.
Expected interval
Set how often your ping is expected to arrive. This matches how often your job runs.
Grace seconds
The Grace seconds field (default 60) gives extra slack before the monitor alarms, so a job that runs a little long does not trigger an alert needlessly.
Ping URL
After you save, a Ping URL is shown. Have your cron job or script call it when the job succeeds. A simple approach is to put the call at the very end of the script so the ping is only sent on a successful run.