Skip to content

Health Check Failed

Error message

                Health check failed: HTTP 503 Service Unavailable
              

Quick summary

The health check endpoint returns a non-200 response, so the load balancer or orchestrator stops sending traffic to the service.

Why this happens

  • The application is overloaded and cannot respond to health checks in time

  • The health check endpoint has a bug or is not implemented correctly

  • A dependency (database, cache) is down, and the health check reports it

  • The health check timeout is too short for the response time

Minimal example

✗ Broken code
livenessProbe:
  httpGet:
    path: /health
    port: 8080
  timeoutSeconds: 1  # too short
✓ Fixed code
livenessProbe:
  httpGet:
    path: /health
    port: 8080
  timeoutSeconds: 5  # more realistic

The health check timeout is too short (1s), causing failures under load. The fix increases it to 5 seconds.

How to diagnose

  • Call the health endpoint directly: curl http://<host>/health

  • Check the application logs for errors around the health check

  • Verify the health check configuration (path, port, timeout)

  • Test if dependencies are available and responding

How to fix

  • Fix the application error that causes the health check to fail

  • Increase the health check timeout or interval

  • Fix or restart the dependency that the health check depends on

  • Ensure the health endpoint returns 200 when the service is healthy

How to prevent

  • Keep health checks lightweight — do not perform expensive operations

  • Set appropriate timeouts based on expected response times

  • Monitor health check failure rates and alert on trends

Related Resources

← Back to language