[StackOverflow/kubernetes] Readiness probe failing for pod

### ROOT CAUSE The readiness probe is failing because the application within the pod is not fully initialized or the probe configuration is incorrect. This could be due to: 1. The readiness probe endpoint not being accessible or returning a non-200 status. 2. The application not being fully started before the probe is executed. 3. Incorrect probe configuration (e.g., wrong path, port, or command). ### CODE FIX Update the readiness probe configuration in the pod specification to ensure it aligns with the application's actual state. Here's an example fix for an HTTP probe: ```yaml spec: containers: - name: my-app image: my-image readinessProbe: httpGet: path: /readyz # Ensure this endpoint exists and returns 200 when ready port: 8080 initialDelaySeconds: 15 # Delay first check by 15 seconds periodSeconds: 5 # Check every 5 seconds ``` **Steps to resolve:** 1. **Verify the endpoint**: Confirm the path `/readyz` exists in your app and returns 200 when the app is ready. 2. **Adjust delays**: Increase `initialDelaySeconds` if the app takes longer to start. 3. **Check logs**: Ensure no errors prevent the app from marking itself as ready. 4. **Debug**: Use `kubectl exec -- curl http://localhost:8080/readyz` to test the endpoint manually. If using an `exec` probe, ensure the command is correct: ```yaml readinessProbe: exec: command: ["/bin/sh", "-c", "app_healthy"] # Replace with your readiness check ``` Always test the probe in a pod to confirm it works as expected.