# Health Checks And Probes

> Design meaningful liveness, readiness, and startup probes so orchestrators and load balancers route traffic correctly. Use when deploying to Kubernetes or any system that relies on health endpoints.

- Skill: `itsual/health-checks-and-probes` (Agent Skill)
- Install (CLI): `npx skillmds@latest add itsual/health-checks-and-probes`
- Raw SKILL.md: https://api.skillmd.com/api/skills/itsual/health-checks-and-probes/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: DevOps & Infra
- Author: itsual (https://skillmd.com/u/itsual)
- Updated: 2026-09-21
- Page: https://skillmd.com/skills/itsual/health-checks-and-probes

---


# Health Checks and Probes

## Overview

Bad health checks cause either unnecessary restarts or traffic being sent to broken instances. Make them accurate and cheap.

## When to Use

- Deploying services to Kubernetes or similar platforms
- Implementing /health, /ready, /live endpoints
- Debugging flapping or slow-to-start services

## Probe Types

- **Liveness** – Is the process alive? (should it be restarted?)
- **Readiness** – Can it accept traffic right now?
- **Startup** – Has it finished initializing?

## Practices

- Keep checks fast and free of heavy dependencies when possible
- Don’t make liveness depend on downstream systems (that causes cascading restarts)
- Make readiness reflect actual ability to serve useful traffic
- Include critical but cheap self-checks
- Log transitions for easier debugging

## Verification

- Probes correctly distinguish “starting”, “ready”, “alive but not ready”, and “dead”
- False positives/negatives are rare under normal conditions