Health Check
A lightweight endpoint or probe a system polls to decide whether an instance is ready to receive traffic.
A health check is a cheap, frequent probe (often just an HTTP GET /healthz) that a load balancer, orchestrator, or service registry polls to decide whether an instance should keep receiving traffic. Liveness checks ask "is this process still running at all," while readiness checks ask the stricter question "is it currently able to serve a request correctly" — a subtle but important distinction, since a process can be alive but temporarily unable to serve (e.g. still warming a cache).
Worked example: a periodic probe (HTTP /healthz, TCP connect) that lets a load balancer or orchestrator know whether an instance can serve traffic and pull it from rotation if not. Gotcha: a shallow check (process up) passes while the instance is broken (DB pool exhausted); but a deep check that calls dependencies can cascade — one slow database fails every instance’s check at once. Separate liveness (restart me) from readiness (route to me).