Quickstart
You can have a job monitored in about two minutes: create a monitor, then have the job check in at the end of its run.
1. Create a monitor
In the dashboard, choose New monitor and give it two things:
- A name — whatever you'll recognize it by, like "Nightly database backup".
- A schedule — either a simple interval ("every 15 minutes") or a cron expression ("
0 3 * * *" for 3 a.m. daily), matching how the job actually runs.
CronWarden creates the monitor and gives it a unique check-in URL.
2. Check in at the end of the job
Add one line to the end of your script so it calls the check-in URL when it finishes:
# ... your job does its work ...
curl -fsS https://cronwarden.com/ping/your-ping-key
That's the whole integration. If the job stops running — the server is down, cron is misconfigured, the script crashes before the curl — the check-in never arrives, and CronWarden alerts you.
3. Report what the run did (optional, recommended)
A check-in proves the job ran. To prove it did its job, send the value it produced:
curl -fsS "https://cronwarden.com/ping/your-ping-key?value=$ROWS_SYNCED"
Once you set a rule for that value — a floor, a ceiling, or a maximum drift from its recent average — CronWarden judges the outcome, not just the heartbeat. See Reporting a value.
4. Get alerted
Connect the channels you want to hear from — Slack, PagerDuty, email, and more — on the Integrations page, and choose which fire on a warning versus a critical alert. See Integrations.
Next steps
- Check-in URLs — the start, success, and fail endpoints.
- Schedules & intervals — cron expressions, intervals, and timezones.
- Monitors as code — create monitors from the API or a crontab instead of the dashboard.