Deploying a Kubernetes Cluster on Ubuntu using MicroK8s

Kubernetes (K8s) is an open-source platform for automating deployment, scaling, and operations of application containers across clusters of hosts. Deploying a Kubernetes cluster can be a complex task, but with the help of MicroK8s, it can be done quickly and easily. MicroK8s is a lightweight, fast, and simple Kubernetes cluster that can be installed on any machine. In this blog post, we will cover how to deploy a high-availability Kubernetes cluster on three Ubuntu 22.04 virtual machines using MicroK8s.

Prerequisites

Before we begin, make sure you have the following prerequisites:

• Three Ubuntu 22.04 virtual machines with at least 2GB of RAM and 2 CPUs each.
• SSH access to all three virtual machines.
• A user account with sudo privileges on each virtual machine.
• A static IP address for each virtual machine.

If you haven't already set up the virtual machines, you can follow the steps in this blog post to create them: Install Ubuntu Server.

Install MicroK8s on All Three Virtual Machines

To install MicroK8s on all three virtual machines, follow these steps:

1. Install MicroK8s using the snap package manager:

   sudo snap install microk8s --classic --channel=1.29

2. Join the microk8s group to avoid using sudo for microk8s commands:

   sudo usermod -a -G microk8s $USER
   sudo mkdir -p ~/.kube
   sudo chown -f -R $USER ~/.kube
   su - $USER

3. Check the status of the MicroK8s cluster:

   microk8s status --wait-ready

For more detailed instructions on installing MicroK8s, you can refer to the official documentation: Get started.

Deploy a High-Availability Kubernetes Cluster

To deploy a high-availability Kubernetes cluster using MicroK8s, at least three nodes are required. In this example, we will use three Ubuntu 22.04 virtual machines. Follow these steps to deploy the cluster:

1. On the first virtual machine, initialize the cluster:

   microk8s add-node

2. Copy the join command from the output of the previous command. for example:

   From the node you wish to join to this cluster, run the following:
   microk8s join 192.168.1.230:25000/92b2db237428470dc4fcfc4ebbd9dc81/2c0cb3284b05
   
   Use the '--worker' flag to join a node as a worker not running the control plane, eg:
   microk8s join 192.168.1.230:25000/92b2db237428470dc4fcfc4ebbd9dc81/2c0cb3284b05 --worker
   
   If the node you are adding is not reachable through the default interface you can use one of the following:
   microk8s join 192.168.1.230:25000/92b2db237428470dc4fcfc4ebbd9dc81/2c0cb3284b05

3. On the second virtual machine, join the cluster by running the join command from the first virtual machine. Note: do not add the --worker flag to the join command as we want this node to be part of the control plane and make the cluster highly available.

4. Once the second node has joined the cluster, go back to the first virtual machine and add the third node by running the microk8s add-node command again.

5. Repeat the join process for the third virtual machine.

6. Set failure domains for the three nodes by running the following commands:

   on the first virtual machine:

   echo "failure-domain=1" > /var/snap/microk8s/current/args/ha-conf
   microk8s.stop
   microk8s.start

   on the second virtual machine:

   echo "failure-domain=2" > /var/snap/microk8s/current/args/ha-conf
   microk8s.stop
   microk8s.start

   on the third virtual machine:

   echo "failure-domain=3" > /var/snap/microk8s/current/args/ha-conf
   microk8s.stop
   microk8s.start

7. Verify that all nodes are part of the cluster by running the following command on the first virtual machine:

   microk8s kubectl get nodes

   you should see output similar to this:

   NAME          STATUS   ROLES    AGE   VERSION
   ubuntu-vm-1   Ready    <none>   2m    v1.29.0
   ubuntu-vm-2   Ready    <none>   1m    v1.29.0
   ubuntu-vm-3   Ready    <none>   1m    v1.29.0

8. Verify that the cluster is highly available by running the following command on the first virtual machine:

   microk8s status

   you should see output similar to this:

   microk8s is running
   high-availability: yes
   	datastore master nodes: 192.168.1.4:19001 192.168.1.5:19001 192.168.1.6:19001
   	datastore standby nodes: none
   addons:
   	...

Congratulations! You have successfully deployed a high-availability Kubernetes cluster on three Ubuntu 22.04 virtual machines using MicroK8s. You can now start deploying and managing your containerized applications on the cluster.

For more information on deploying a high-availability Kubernetes cluster using MicroK8s, you can refer to the official documentation: High Availability (HA).

Enabling Dashboard

To enable the Kubernetes dashboard, run the following command on the first virtual machine:

microk8s enable dashboard

Before accessing the dashboard, you need to get the token to log in. Run the following command to get the token:

microk8s kubectl describe secret -n kube-system microk8s-dashboard-token

Then run the following command to forward the dashboard port to your local machine:

microk8s kubectl port-forward -n kube-system --address 0.0.0.0 service/kubernetes-dashboard 10443:443

Note: we added --address 0.0.0.0 to allow access to the dashboard from any IP address. Make sure to secure the dashboard access by restricting the IP addresses that can access it.

Now you can access the Kubernetes dashboard by opening a web browser and navigating to https://<first-vm-ip>:10443. Use the token you obtained earlier to log in.

For more information on the Kubernetes dashboard, you can refer to the official documentation: Addon: dashboard.

Conclusion

In this blog post, we covered how to deploy a high-availability Kubernetes cluster on three Ubuntu 22.04 virtual machines using MicroK8s. We also enabled the Kubernetes dashboard to help you manage and monitor your cluster. With this setup, you can start deploying and managing your containerized applications on the Kubernetes cluster.