Find out when your cron job didn’t run
Uptime checks can’t see a backup that silently stopped. A heartbeat monitor works the other way round: your job calls us, and we alert you when it doesn’t.
How it works.
Create a heartbeat monitor
Copy its URL. Set how often the job should run (1 minute to 1 week) and how late it may be (1 minute to 1 day).
Add one line to the job
At the end of it, for example curl -fsS https://ping.pingwise.com/<token>. Anything that can make an HTTP request works.
Silence pages you
If no call arrives in time, the monitor goes down and you’re alerted on the channels you chose.
Report failures, not just silence.
Fail at once
Call the URL with /fail at the end and the monitor goes down immediately. The first 300 characters of the request body become the reason.
Check what the job says
POST the job’s output and set a keyword. «OK» must be there, or «ERROR» must not. A mismatch is a failure, with the reason shown.
Alerts where you work
Email, Slack, Teams, Discord, Telegram, Google Chat, PagerDuty, Pushover, ntfy, a webhook, or SMS and phone calls to EU and UK numbers, the same as every other monitor.
Schedules
Build and read cron schedules with the cron expression generator.
Questions
What does a heartbeat cost?
One credit per expected check-in. A job every 5 minutes uses 8,640 credits a month, and a daily job uses 30.
Is there a rate limit?
30 pings a minute per token.
Does it need an agent?
No. Anything that can make an HTTP request works: curl, wget, PowerShell, or your own code.
Related: Uptime monitoring · Heartbeat monitors in the docs