Do you want to know how to manage and orchestrate virtual machines within Kubernetes clusters? Let’s explore KubeVirt, that allows you to integrate virtualized workloads alongside containerized applications for enhanced flexibility and resource utilization.

In today’s software development, Kubernetes has taken its position as the de facto standard for container orchestration, offering great flexibility and scalability. However, many development teams face a dilemma: while Kubernetes excels at managing containerized workloads, what about existing Virtual Machine (VM)-based applications that are not easily containerized?

KubeVirt logo

This is where KubeVirt comes to play – a technology designed to bridge the gap between traditional VM workloads and the Kubernetes ecosystem. But why would one want to run VM workloads under Kubernetes instead of opting for conventional VM hosting solutions like Azure Virtual Machines? Let’s delve into the core challenges KubeVirt addresses and the reasons why professionals are turning to Kubernetes to orchestrate and manage their VM workloads.

Simple KubeVirt Architecture

KubeVirt extends Kubernetes by adding custom resources and controllers specifically designed to handle virtual machines (VMs). At its core, KubeVirt introduces two primary components: the VirtualMachineInstance (VMI) and the VirtualMachine (VM) custom resources.

KubeVirt also introduces custom controllers responsible for managing these custom resources and orchestrating VM lifecycle operations within Kubernetes clusters. These controllers monitor the state of VMIs and VMs, ensuring that the desired state specified in the resource manifests is achieved and maintained.

In terms of integration with Kubernetes architecture, KubeVirt extends the Kubernetes API server to support the creation and management of VM-related custom resources. This allows users to interact with VMs using standard Kubernetes API calls. KubeVirt deploys custom controllers as part of the Kubernetes controller manager. These controllers continuously reconcile the state of VMIs and VMs with the desired state specified by users, ensuring proper VM lifecycle management. KubeVirt integrates with the Kubernetes scheduler to determine the optimal placement of VM instances across nodes in the cluster based on resource requirements and constraints. Additionally, KubeVirt leverages Kubernetes networking and storage plugins to provide networking and storage resources to VM instances. This enables seamless integration with existing Kubernetes networking and storage solutions.

KubeVirt fundamentally changes how organizations handle VM workloads by smoothly incorporating them into the Kubernetes ecosystem. By leveraging Kubernetes’ robust orchestration capabilities, KubeVirt enables organizations to improve their agility, scalability, and efficiency while preserving the investments in existing VM-based applications. As professionals increasingly recognize the transformative potential of Kubernetes for managing diverse workloads, KubeVirt plays a crucial role in facilitating this shift towards a cloud-native future.

Let’s witness KubeVirt by deploying it on an Azure Kubernetes Service (AKS) cluster.

Before diving into the next steps, ensure you have the following prerequisites in place:

Step 1: First, login using az login, and set the right subscription with the following command:

az account set --subscription "<subscription-id>"

Step 2: First, let’s create the resource group. Use the command below to create your resource group:

az group create --name rg-p-weu-kubevirt-001 --location westeurope

Step 3: Next, we’ll create a new managed Kubernetes cluster by running the command below:

az aks create --resource-group rg-p-weu-kubevirt-001 \
              --name aks-p-weu-kubevirt-001 \
              --node-count 3 \
              --node-vm-size Standard_D8ds_v5 \
              --enable-cluster-autoscaler \
              --min-count 3 \
              --max-count 10 \
              --enable-addons monitoring \
              --generate-ssh-keys \
              --location westeurope

⚠️ Note: This setup is for demonstration purposes and does not adhere to best practices for production scenarios.

The process of creating a cluster may require a few minutes. Once completed, you are ready to continue onto the next steps.

Step 4: Next, we need to get the credentials for the AKS cluster and login (interactively) using the command below:

az aks get-credentials --resource-group rg-p-weu-kubevirt-001 --name aks-p-weu-kubevirt-001

Step 5: We will now install KubeVirt by using the KubeVirt operator, which manages the lifecycle of all the KubeVirt core components. Run the command below to deploy the KubeVirt operator to your cluster:

kubectl create -f https://github.com/kubevirt/kubevirt/releases/download/v1.2.0/kubevirt-operator.yaml

Step 6: Now, we’ll use kubectl again to deploy the KubeVirt custom resource definitions. Run the command below to do so:

kubectl create -f https://github.com/kubevirt/kubevirt/releases/download/v1.2.0/kubevirt-cr.yaml

By default KubeVirt will deploy 7 pods, 3 services, 1 daemonset, 3 deployment apps, and 3 replica sets.

Step 7: Deploying KubeVirt can take a few minutes. You can verify the deployment by running the command below:

kubectl get kubevirt.kubevirt.io/kubevirt -n kubevirt -o=jsonpath="{.status.phase}"

Additionally, you can verify the components by using the following command:

kubectl get all -n kubevirt

Get All KubeVirt

Basic VirtualMachineInstance operations can be performed with the kubectl utility. However, the virtctl command line interface is required to use advanced features such as serial and graphical console access. It also provides more convenient commands for starting and stopping VirtualMachineInstances, live migrating VirtualMachineInstances and canceling live migrations, and uploading virtual machine disk images.

mv /path/to/the/directory/virtctl-<version>-<os>-<architecture> /usr/local/bin/virtctl

3. Restart your terminal and run the virtctl command to validate it is working.