> ## Documentation Index
> Fetch the complete documentation index at: https://docs.enterprise.falkordb.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Replicated deployment with Sentinel

> Create a replicated FalkorDB deployment with Redis Sentinel high availability.

## When to use this

Use Sentinel when the deployment should support failover from one FalkorDB replica to another.

The Admin UI maps this flow to:

* `spec.topology: replication`
* `spec.replicas` from the FalkorDB replica count
* `spec.sentinel.enabled: true`
* `spec.sentinel.replicas` from the Sentinel node count

## Create from Admin UI

1. Open **Deployments** > **Create Deployment**.
2. Choose **Standalone**.
3. Choose **QA** or **Prod** to start from a Sentinel-enabled template, or choose **Full Setup** and enable Sentinel manually.
4. In **FalkorDB**, set the FalkorDB replica count. Prod requires at least `2`.
5. In **Sentinel**, enable Sentinel and set an odd number of Sentinel replicas. The UI defaults to `3`.
6. Configure Sentinel CPU, memory, storage, and optional node affinity.
7. Optionally add a Sentinel service in **Networking**, or set an external domain for stable per-pod DNS names outside the cluster — see [Networking](/databases/networking).
8. Complete auth, TLS, backup, parameters, and review.

## Template defaults

| Template | FalkorDB replicas | Sentinel | Sentinel replicas | Backup   |
| -------- | ----------------- | -------- | ----------------- | -------- |
| Dev      | `1`               | disabled | n/a               | disabled |
| QA       | `2`               | enabled  | `3`               | disabled |
| Prod     | at least `2`      | required | odd, at least `3` | enabled  |

## Failover status

The current Pods tab displays a **Switchover** confirmation for the primary pod, but this UI path currently only shows a success toast. There is no backend switchover operation wired in this codebase yet.

Until that is implemented, handle manual failover through your Kubernetes/KubeBlocks operating procedure and verify the resulting pod roles in the Admin UI.

## Validate

```bash theme={null}
kubectl get clusters.apps.kubeblocks.io -n <namespace>
kubectl get pods -n <namespace>
kubectl get opsrequests.operations.kubeblocks.io -n <namespace>
```

In the Admin UI, verify **Pods**, **Operations**, **Logs**, and **Metrics**.