Runbook for API Failures and Silent Cron Jobs in a Backend Metrics Dashboard

Short summary

This runbook distinguishes three monitoring signals that must stay separate: API failure counts from metrics, business event counts from domain verbs, and schedule completeness from heartbeat pings. A cron job that never fires produces no signal at all, so heartbeat monitoring must live outside the metrics query path. The author compares Prometheus+Grafana, Datadog, Sentry, and Infrai+Healthchecks, recommending the latter for small SaaS teams wanting a stable HTTP contract with built-in missing-run detection.

• •Three distinct signals needed: API outcomes, business events, and heartbeat pings — collapsing them hides real failures
• •Heartbeat monitoring must be external to the metrics system so a failed component isn't asked to detect its own silence
• •Tool comparison favors Infrai+Healthchecks for small SaaS; Prometheus+Grafana for teams already operating collectors

Generated with AI, which can make mistakes.