Incident response

Heartbeats

Track cron jobs and scheduled tasks with MonoDuty check-in URLs, grace periods, missed-check alerts, escalation, and check-in history.

Heartbeats

A Heartbeat tracks an expected check-in from a recurring job. It belongs to a workspace and owning team, may be linked to a service, and can create an incident trigger when a check-in is missed or its expected body does not match.

Create a Heartbeat

Create and manage heartbeats through the authenticated https://api.monoduty.com/api/heartbeats endpoint. Provide the workspace, owning team or owned service, name, interval in milliseconds, optional grace period, optional expected body, and whether to create the implicit incident trigger. The response returns the public check-in URL at https://mndty.com/ping/:token.

Intervals and grace periods are bounded by the API and plan entitlements. Do not publish the check-in token.

Send a check-in

curl -X POST https://mndty.com/ping/YOUR_HEARTBEAT_TOKEN

A matching check-in returns HTTP 200. If an expected body is configured, send that exact body; a mismatch returns HTTP 422 and records the Heartbeat as down. Bodies larger than 64 KiB are rejected.

Missed and recovered states

The scheduler evaluates next_expected_at and emits durable status-change events. An attached implicit trigger routes the resulting incident through the configured on-call and escalation workflow. Email, SMS, and voice are the current release-gated notification channels.

Pause a Heartbeat during planned maintenance; resuming resets its next expected window.