> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.lakera.ai/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.lakera.ai/_mcp/server.

# Kubernetes Probes

You can deploy [self-hosted Check Point AI Guardrails](/docs/selfhosting) to Kubernetes.
AI Guardrails exposes endpoints for liveness, readiness, and startup probes.

For detailed Kubernetes deployment instructions, see
[Self-hosting AI Guardrails](/docs/selfhosting).

## Startup

Use `/startupz` to implement
[a startup probe](https://kubernetes.io/docs/tasks/configure-pod-container/configure-liveness-readiness-startup-probes/#define-startup-probes)
for your K8s deployment. The startup probe is recommended if you also implement
readiness or liveness probes, because those require the gunicorn workers to start.

Example usage:

```yaml
startupProbe:
  httpGet:
    path: /startupz
    port: 8000
  periodSeconds: 10
  failureThreshold: 30
```

## Readiness

Use `/readyz` to implement
[a readiness probe](https://kubernetes.io/docs/tasks/configure-pod-container/configure-liveness-readiness-startup-probes/#define-readiness-probes)
for your K8s deployment. If you implement both readiness and liveness probes, a higher
`failureThreshold` on the liveness probe is recommended.

Example usage:

```yaml
readinessProbe:
  httpGet:
    path: /readyz
    port: 8000
  periodSeconds: 5
  failureThreshold: 1
```

## Liveness

Use `/livez` to implement
[a liveness probe](https://kubernetes.io/docs/tasks/configure-pod-container/configure-liveness-readiness-startup-probes/#define-a-liveness-http-request)
for your K8s deployment. If you implement both readiness and liveness probes, a higher
`failureThreshold` on the liveness probe is recommended.

Example usage:

```yaml
livenessProbe:
  httpGet:
    path: /livez
    port: 8000
  periodSeconds: 5
  failureThreshold: 3
```