Health Check

A health check is a lightweight request, usually to an endpoint such as /healthz or /health, that tells the caller whether an application is able to serve traffic. Load balancers use it to pick backends, orchestrators like Kubernetes use it to restart containers, and monitors use it to alert people.

Why it matters

A running process is not a working service. A good health check detects what a port check cannot: a lost database connection, a full disk or a stuck worker. A bad one either always says "OK" or fails on a minor dependency and takes a whole fleet out of rotation.

Example

GET /healthz HTTP/1.1

HTTP/1.1 200 OK
Content-Type: application/json

{"status":"ok","db":"ok","cache":"ok","version":"2026.10.1"}

Return 200 when healthy and 503 when not. Keep it fast (well under a second), cheap and unauthenticated or restricted by network. Kubernetes separates liveness (restart me) from readiness (send me traffic).

Common mistakes

• Checking a downstream third-party API on every call, so its outage makes your instances look dead.
• Making the check expensive, so frequent polling adds load.
• Exposing version and dependency details publicly.
• Only checking /healthz externally while real pages fail, so add a content check on a real page too.

Related

• Load balancer
• Synthetic monitoring
• HTTP status code
• गाइड: Load balancer health checks
• टूल: Website health check

यह पेज सामान्य जानकारी है, जिसे हमारी संपादकीय टीम ने लिखा है और दस्तावेज़ीकरण बदलने पर अपडेट किया जाता है। सर्वर और प्रदाता अलग-अलग होते हैं, इसलिए बदलाव पहले सुरक्षित तरीके से जाँचें। देखें: संपादकीय नीति और अस्वीकरण.