Access Ray Dashboard via kubectl Port-Forward without Public Exposure - Container Service for Kubernetes

Ray Dashboard is a web interface for monitoring and troubleshooting Ray clusters. It provides visual monitoring, performance tuning, and application management. In Container Service for Kubernetes (ACK), a Ray cluster created through the managed KubeRay Operator automatically creates a ClusterIP Service for the head pod. Because Ray Dashboard has no built-in authentication or authorization, do not expose this Service as LoadBalancer. Use kubectl port-forward to access the Dashboard securely from your local machine.

Important

KubeRay Operator in ACK is in invitational preview. To use this component, submit a ticket to apply.

Prerequisites

Before you begin, ensure that you have:

A Ray cluster running in ACK. For setup instructions, see Efficient deployment and optimization of Ray in ACK clusters

Access the Ray Dashboard

Note

Metrics display in Ray Dashboard is disabled by default.

Get the Service name for the Ray cluster head pod in your namespace:

kubectl get svc -n ${RAY_CLUSTER_NS}

Expected output:

NAME                           TYPE        CLUSTER-IP        EXTERNAL-IP   PORT(S)                                         AGE
myfirst-ray-cluster-head-svc   ClusterIP   192.XX.XX.188   <none>        10001/TCP,8265/TCP,8080/TCP,6379/TCP,8000/TCP   18m

Note the Service name (for example, myfirst-ray-cluster-head-svc). You use it in the next step.

Forward local port 8265 to port 8265 on the Ray cluster head Service. Run this command in a separate terminal — it blocks the current shell while active:
```
kubectl port-forward svc/myfirst-ray-cluster-head-svc --address 0.0.0.0 8265:8265 -n ${RAY_CLUSTER_NS}
```
The following output confirms the forwarding is active:
```
Forwarding from 0.0.0.0:8265 -> 8265
```
Open http://127.0.0.1:8265/ in your browser. The Ray Dashboard opens and displays the cluster overview.

What's next

To run distributed tasks on your Ray cluster, see Submit a Ray job.
To automatically scale Elastic Compute Service (ECS) nodes using the Ray autoscaler, see Elastic scaling based on the Ray autoscaler and ACK autoscaler.
To automatically scale Elastic Container Instance nodes using the Ray autoscaler, see Elastic scaling of Elastic Container Instance nodes based on the Ray autoscaler.