Running Virtual Machines on OpenShift (OpenShift Virtualization)

Virtual machines are not enabled by default in OpenShift. To run VMs, OpenShift relies on OpenShift Virtualization, which is based on KubeVirt and integrates virtual machines directly into the Kubernetes control plane.

In this guide, you’ll learn how to enable virtual machines on Red Hat OpenShift, verify the required prerequisites, and confirm that your cluster is ready to run VM workloads.

Prerequisites

Before enabling virtualization, confirm the following:

• OpenShift 4.10 or later
• Cluster administrator access
• At least one worker node with hardware virtualization support
• Nodes expose /dev/kvm (no KVM = no VMs)
• Sufficient CPU and memory on worker nodes

Step 1: Verify Node Virtualization Support

Check whether your worker nodes expose KVM:

oc get nodes -o custom-columns=NAME:.metadata.name,KVM:.status.allocatable.'devices\.kubevirt\.io/kvm'

Expected result for workers:

worker0   1000
worker1   1000

If you see 0 or <none>, the node cannot run virtual machines.

To confirm directly on a node:

oc debug node/worker0
chroot /host
ls -l /dev/kvm
egrep -o 'vmx|svm' /proc/cpuinfo | head
exit

If /dev/kvm is missing, fix virtualization at the infrastructure layer before continuing.

Step 2: Install the OpenShift Virtualization Operator

Enable virtualization by installing the operator from Ecosystem > Software Catalog (OpenShift 4.20+)

You might be asked to setup up HyperConverged Cluster, just click next through this process.

This step deploys all virtualization components, including:

• KubeVirt
• CDI (Containerized Data Importer)
• VM controllers and APIs

Step 3: Monitor the Installation

Watch the operator pods come up:

oc get pods -n openshift-cnv -w

Wait until all pods are in a Running or Completed state.

Verify the HyperConverged object:

% oc get hyperconverged -n openshift-cnv
NAME                      AGE
kubevirt-hyperconverged   18m

Step 4: Confirm Virtualization Is Enabled

Check that KubeVirt is running:

oc get kubevirt -n openshift-cnv

Confirm the API resources are available:

oc api-resources | grep -i virtualmachine

You should see resources such as:

• virtualmachines
• virtualmachineinstances
• virtualmachineinstancemigrations

This confirms that your cluster is now capable of running VMs.

Now we can start VMS on the platform. Click Virtualization > VirtualMachines > Create > FromInstanceType

Once you launch the VM, it will get provisioned on the most suitable worker. You can monitor this process from the web ui as well.

Conclusion

OpenShift Virtualization makes it possible to run traditional virtual machines and modern container workloads side by side on the same platform. By enabling virtualization and verifying KVM support at the infrastructure level, you’ve laid the foundation for a flexible hybrid environment that can support both legacy applications and cloud-native workloads.

With virtualization now enabled, your OpenShift cluster is no longer limited to containers alone. You can begin migrating existing virtual machines, building new VM-based services, or gradually modernising applications at your own pace while continuing to use the same tools, processes, and access controls you already rely on in OpenShift.