How Monitoring Works

At its core, HealthyCron uses a "heartbeat" or "dead man's switch" model. It expects an incoming HTTP request at regular intervals.

The Heartbeat Model

When your cron job or scheduled task completes successfully, it makes a simple HTTP GET or POST request to HealthyCron. This ping tells HealthyCron that everything is functioning correctly for that interval.

Job runs
   ↓
Pings /ping/uuid
   ↓
healthycron resets timer

Failure Modes

There are primarily two reasons why a job might fail, and both are caught by this model:

• The job never started. The server crashed, the cron daemon stopped, or the physical server lost power. Because the job never ran, the ping never triggered, and HealthyCron alerts you that it missed its schedule.
• The job started but crashed. An exception was thrown, or the script exited gracefully due to an error. Because it exited before reaching the ping at the end of the script, HealthyCron realizes the pulse is missing and alerts you.

Start, Success, and Fail States

For advanced tracking, you can emit telemetry at different stages of your job's lifecycle. By pinging at the start, you allow HealthyCron to measure exact execution duration.