Install Vertical Pod Autoscaler using Helm

This page shows how to install the Vertical Pod Autoscaler (VPA) using the Helm chart published by the Kubernetes Autoscaler project.

The VPA chart installs the three VPA components:

• the recommender
• the updater
• the admission controller

The chart version shown in the examples on this page can change over time as the Kubernetes Autoscaler project publishes new releases.

After you install the chart, you can create VerticalPodAutoscaler resources for workloads that you want VPA to manage.

Before you begin

You need to have a Kubernetes cluster, and the kubectl command-line tool must be configured to communicate with your cluster. It is recommended to run this tutorial on a cluster with at least two nodes that are not acting as control plane hosts. If you do not already have a cluster, you can create one by using minikube or you can use one of these Kubernetes playgrounds:

• iximiuz Labs
• Killercoda
• KodeKloud

To check the version, enter kubectl version.

Before you begin:

• Install Helm.
• Make sure that Metrics Server is deployed in your cluster. VPA uses the metrics.k8s.io API to make recommendations.
• Verify that your cluster is running Kubernetes v1.28 or later.

Add the Kubernetes Autoscaler Helm repository

Add the Helm repository that publishes the VPA chart:

helm repo add autoscalers https://kubernetes.github.io/autoscaler
helm repo update

Install the Helm chart

Install the VPA chart in the kube-system namespace:

helm upgrade --install vpa autoscalers/vertical-pod-autoscaler \
  --namespace kube-system \
  --create-namespace

By default, the chart configures Helm to manage the VPA admission webhook and to generate the required TLS certificates for that webhook.

If you want to review the chart values before installing, run:

helm show values autoscalers/vertical-pod-autoscaler

Verify the installation

Check that the Helm release is deployed:

helm list --namespace kube-system

The output is similar to:

NAME  	NAMESPACE   	REVISION	UPDATED                                 	STATUS  	CHART                         	APP VERSION
vpa   	kube-system	1       	2026-04-07 12:00:00.000000 +0000 UTC	deployed	vertical-pod-autoscaler-0.8.1	1.6.0

Check that the VPA Pods are running:

kubectl get pods -n kube-system

The output is similar to:

NAME                                                         READY   STATUS    RESTARTS   AGE
vpa-vertical-pod-autoscaler-admission-controller-xxxxx       1/1     Running   0          1m
vpa-vertical-pod-autoscaler-recommender-xxxxx                1/1     Running   0          1m
vpa-vertical-pod-autoscaler-updater-xxxxx                    1/1     Running   0          1m

If your release name is different from vpa, adjust the Pod names accordingly.

Confirm that the VerticalPodAutoscaler custom resource definition exists:

kubectl get customresourcedefinitions verticalpodautoscalers.autoscaling.k8s.io

Create a VerticalPodAutoscaler resource

After you install VPA, create a VerticalPodAutoscaler resource that targets a workload. The following manifest configures VPA for a Deployment named my-app. The example uses the explicit Recreate update mode because the Auto mode is deprecated:

apiVersion: autoscaling.k8s.io/v1
kind: VerticalPodAutoscaler
metadata:
  name: my-app-vpa
spec:
  targetRef:
    apiVersion: apps/v1
    kind: Deployment
    name: my-app
  updatePolicy:
    updateMode: Recreate

Apply the manifest:

kubectl apply -f my-app-vpa.yaml

To inspect the recommendation that VPA produces, run:

kubectl describe vpa my-app-vpa

Configure the chart

To customize the chart, provide your own values file when you install or upgrade the release:

helm upgrade --install vpa autoscalers/vertical-pod-autoscaler \
  --namespace kube-system \
  --create-namespace \
  --values vpa-values.yaml

For example, you can use chart values to:

• change the number of recommender, updater, or admission controller replicas
• set Pod scheduling constraints such as node selectors, affinity rules, and tolerations
• configure how the admission webhook is managed
• override container image tags or pull policies for the VPA components

For the full set of chart values, see the VPA Helm chart README.

Uninstall VPA

To remove the Helm release:

helm uninstall vpa --namespace kube-system

If you no longer need the VPA custom resource definitions, you can remove them separately:

kubectl delete customresourcedefinition verticalpodautoscalers.autoscaling.k8s.io

What's next

• Learn more about Vertical Pod Autoscaler.
• Read the HorizontalPodAutoscaler walkthrough.
• Review the upstream VPA installation documentation.