How Drumbeats Works

Drumbeats is a heartbeat monitor for background jobs. It answers one question fast: did the job actually run and finish on time? Jobs ping a per-monitor URL on start, success, failure, or progress. Drumbeats records every event, detects missed or hung runs, opens an incident when something breaks, and pages the notification group attached to the monitor.

This page is the conceptual overview. For the runnable setup walkthrough see Quickstart; for the HTTP surface see Ping API.

The 30-second model

1. You create a monitor in the Drumbeats dashboard and pick its type — Cron, Heartbeat, Event-driven, or Uptime.
2. Your job (or scheduler) sends an HTTP ping to the monitor's URL — https://api.drumbeats.io/v1/ping/<monitor-id>/<event>.
3. The Drumbeats Ping API records the event on the run timeline and updates the monitor state.
4. If the expected ping fails to arrive — or a failure event arrives — Drumbeats opens an incident and pages every channel in every notification group attached to the monitor.
5. A subsequent success ping (or, for Uptime monitors, a passing scheduled check) resolves the incident and sends a recovery message.

What you get

• Coverage for cron jobs, heartbeat workers, queue consumers, pipelines, and HTTP endpoints. One product, four monitor types — see Monitor types for the picker.
• Alerts when a job is late, missed, hung, or failed. Tunable per monitor via failure_tolerance, schedule_tolerance, grace_period_seconds, and max_duration_seconds.
• No SDK and no agent to install. One HTTP request per ping — curl, fetch, requests, or any other client works.
• Usage-based pricing in Beats, not per-monitor seats. Idle monitors cost nothing; the bill scales with actual ping volume. See Beats & usage.

Setup in three steps

1. Create a monitor in the dashboard and pick Cron, Heartbeat, Event-driven, or Uptime.
2. Copy the Ping URL from the monitor detail page: https://api.drumbeats.io/v1/ping/<monitor-id>.
3. Ping on start and finish from your job.

curl https://api.drumbeats.io/v1/ping/<monitor-id>/start
your_job_command_here
if [ $? -eq 0 ]; then
  curl https://api.drumbeats.io/v1/ping/<monitor-id>/success
else
  curl https://api.drumbeats.io/v1/ping/<monitor-id>/failure
fi

curl https://api.drumbeats.io/v1/ping/<monitor-id>/start
your_job_command_here
if [ $? -eq 0 ]; then
  curl https://api.drumbeats.io/v1/ping/<monitor-id>/success
else
  curl https://api.drumbeats.io/v1/ping/<monitor-id>/failure
fi

The Quickstart walks the same flow end-to-end with screenshots and a tested alert round-trip.

Where to go next

• Monitor types — pick the right type for the workload (Cron, Heartbeat, Event-driven, Uptime).
• Monitoring modes — scheduled vs event-driven from a workload perspective.
• Ping events — the four event names and what each one signals.
• Beats & usage — how usage is metered and what a typical workload costs.
• Payloads & logs — what gets stored alongside pings.
• Ping API — the HTTP surface this whole model rides on.