Heartbeat Monitoring

Ensure scheduled jobs and background workers report in on time, and alert the moment a run is missed or finishes outside its expected window. Heartbeat monitoring inverts the usual model: instead of NoDisrupt reaching out to check your service, your job checks in with NoDisrupt - so silence becomes the signal that something went wrong.

What it checks

A heartbeat monitor expects a periodic check-in from your job and watches for it to arrive on schedule:

• Missed runs - detection of when a scheduled process fails to report completion within its expected window.
• Late completions - alerts when a job finishes outside the timing window you expect, even if it eventually checks in.
• Stalled pipelines - quick identification of ETL and asynchronous workflows that have stopped making progress.
• Ongoing background work - confirmation that long-running sync processes continue sending heartbeats as they should.

Configuration

Heartbeat monitors are configured around when you expect to hear from your job:

• Expected schedule - how often the job is expected to check in, expressed as an interval or schedule that matches your cron or batch cadence.
• Grace period - how much later than expected a check-in may arrive before the run is treated as missed, absorbing normal variation in run times.
• Check-in endpoint - the unique URL your job calls when it completes; NoDisrupt records the check-in and resets the timer.

When no check-in arrives within the schedule plus the grace period, the monitor raises an incident and notifies your alert policies.