Healthcheck Core
@nexload-sdk/healthcheck runs checks and collectors and returns a stable HealthReport. Use it for liveness, readiness, startup, protected diagnostics, and exporter input.
It owns orchestration: scope selection, timeouts, cancellation, retries, status aggregation, redaction, and JSON serialization. Runtime, framework, Payload, Prometheus, and OpenTelemetry behavior live in separate packages.
In plain language: a check can change service health, while a collector only adds measurements. A scope is the question being asked: liveness asks whether the process should restart, readiness asks whether it can receive traffic, startup asks whether initialization finished, and diagnostics exposes protected operational detail.
Choose the next page
Section titled “Choose the next page”- New project: Installation then Quick start.
- Designing probes: Concepts and Guides.
- Looking up a symbol: API.
- A check is timing out or changing status: Troubleshooting.
Non-goals
Section titled “Non-goals”Core does not start an HTTP server, inspect Node/Bun-specific resources, query Payload, or publish telemetry. It does not make dependency failures part of liveness unless you explicitly assign them there.
This site documents the current package version only. Use the package changelog for release history.