Kubernetes Dashboard Management

Introduction

Kubernetes Dashboard is a web-based user interface that simplifies the management and monitoring of Kubernetes clusters. It provides an intuitive way to deploy and manage applications, as well as view and analyze cluster resources. This lab will guide you through deploying and accessing Kubernetes Dashboard.

Skills Graph

Start the Kubernetes Cluster

Before deploying the Kubernetes Dashboard, ensure that your cluster is running. We will use Minikube for this lab.

1. Open a terminal and navigate to the project directory:

   cd ~/project

2. Start the Minikube cluster:

   minikube start

   Minikube creates a local Kubernetes cluster, making it easy to test features like the Dashboard. The setup may take a few minutes.

3. Verify that Minikube is running by checking its status:

   minikube status

   If the cluster is not running, restart it with minikube delete followed by minikube start.

Deploy Kubernetes Dashboard

Kubernetes Dashboard is not included by default in the cluster. Deploy it using the official deployment YAML file.

1. Run the following command to deploy the Dashboard:

   kubectl apply -f https://raw.githubusercontent.com/kubernetes/dashboard/v2.3.1/aio/deploy/recommended.yaml

   This command creates all necessary resources for the Dashboard, including a Deployment, a Service, and Role-Based Access Control (RBAC) settings.

2. Verify that the Dashboard is running:

   kubectl get pods -n kubernetes-dashboard

   Look for a pod with the name kubernetes-dashboard and ensure its status is Running.

3. Verify the namespace for the Dashboard:

   kubectl get ns | grep kubernetes-dashboard

   The namespace kubernetes-dashboard should exist.

Create a Service Account and Role Binding

By default, Kubernetes Dashboard restricts access. To log in with admin privileges, you need to create a service account and bind it to a cluster role.

1. Create a file named dashboard-admin.yaml:

   nano ~/project/dashboard-admin.yaml

2. Add the following content to the file:

apiVersion: v1
kind: ServiceAccount
metadata:
  name: admin-user
  namespace: kubernetes-dashboard
---
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRoleBinding
metadata:
  name: admin-user
roleRef:
  apiGroup: rbac.authorization.k8s.io
  kind: ClusterRole
  name: cluster-admin
subjects:
  - kind: ServiceAccount
    name: admin-user
    namespace: kubernetes-dashboard

This file defines a service account admin-user and grants it admin privileges across the cluster.

Press Ctrl+X, Y, then Enter to save the file.

3. Apply the configuration:

   kubectl apply -f dashboard-admin.yaml

4. Verify the service account and role binding:

   kubectl get sa -n kubernetes-dashboard

   You should see the admin-user service account listed.

   kubectl get clusterrolebinding

   You should see the admin-user cluster role binding listed.

Access Kubernetes Dashboard

Now that the Dashboard is running and permissions are configured, access it via a browser.

1. Generate a login token:

   kubectl -n kubernetes-dashboard create token admin-user

   Copy the generated token. This will be used to log in.

2. Edit the Dashboard Service to expose it on a NodePort:

   kubectl edit service -n kubernetes-dashboard kubernetes-dashboard

   Find the type field under spec and change its value to NodePort. Save and exit.

   

3. Find the NodePort assigned to the Dashboard:

   kubectl get service -n kubernetes-dashboard

   NAME                        TYPE        CLUSTER-IP       EXTERNAL-IP   PORT(S)         AGE
   dashboard-metrics-scraper   ClusterIP   10.104.164.172   <none>        8000/TCP        20m
   kubernetes-dashboard        NodePort    10.108.222.153   <none>        443:30587/TCP   20m

   Note the port number listed under the PORT(S) column (e.g., 30587). This is the port you will use to access the Dashboard.

4. Retrieve the node's IP address:

   kubectl get node -o wide

   NAME       STATUS   ROLES           AGE   VERSION   INTERNAL-IP    EXTERNAL-IP   OS-IMAGE             KERNEL-VERSION      CONTAINER-RUNTIME
   minikube   Ready    control-plane   35m   v1.26.1   192.168.58.2   <none>        Ubuntu 20.04.5 LTS   5.15.0-56-generic   docker://20.10.23

   The INTERNAL-IP column provides the node's IP address.

5. Open the Firefox browser on the desktop and navigate to the following URL:

   https://<node-ip>:<node-port>

   Replace <node-ip> with the node's IP address and <node-port> with the port from Step 3.

   

6. If prompted with a security warning, proceed by selecting "Advanced" and then "Accept the Risk and Continue."

7. On the login page, select the Token option, paste (right click) the token from Step 1, and click "Sign In."

You should now be logged into the Kubernetes Dashboard, where you can explore cluster resources and manage workloads.

Summary

In this lab, you learned how to:

1. Deploy Kubernetes Dashboard to a local cluster.
2. Create a service account and bind it to a cluster role for admin access.
3. Access the Dashboard via a browser using a NodePort service.

With Kubernetes Dashboard, you can visually manage applications, monitor resources, and explore cluster configurations. This tool simplifies Kubernetes operations, especially for beginners transitioning from command-line workflows. Keep exploring its features to make the most of your Kubernetes environment!