Create Monitor

A heartbeat monitor flips the model: instead of Spectra checking your service, your job pings Spectra to say “I ran.” If the expected ping doesn’t arrive in time, Spectra alerts you. Open Monitors → Add monitor and choose Heartbeat / cron.

1. Schedule and grace period

• Name — a label like Nightly backup.
• Schedule — how often the job is expected to run, as an interval or a cron expression.
• Grace period — how long after the expected time Spectra waits before declaring the beat missed. Size it to the job’s real-world variance so normal jitter doesn’t false-alarm.

2. Get the ping URL

Spectra gives you a unique inbound ping URL, e.g. https://spectrabolt.com/beat/<id>. Call it at the end of your job’s success path:

0 3 * * * /usr/bin/backup.sh && curl -fsS https://spectrabolt.com/beat/<id>

Any HTTP method works (GET, POST, HEAD). The UUID in the URL is the secret — keep it private.

3. Start / complete / fail (optional)

Send a status so Spectra can distinguish “never ran”, “still running”, and “ran but failed”:

• ?status=start when the job begins
• ?status=complete on success
• ?status=fail on error

This also enables max-duration alerts for jobs that hang.

4. Alerts and save

Attach the alert channels and Save.