Helm

Helm is the package management tool of choice for Kubernetes. Helm charts provide templating syntax for Kubernetes YAML manifest documents. With Helm we can create configurable deployments instead of just using static files. For more information about creating your own catalog of deployments, check out the docs at https://helm.sh/docs/intro/quickstart/.

K3s does not require any special configuration to use with Helm command-line tools. Just be sure you have properly set up your kubeconfig as per the section about cluster access. K3s does include some extra functionality to make deploying both traditional Kubernetes resource manifests and Helm Charts even easier with the rancher/helm-release CRD.

This section covers the following topics:

• Automatically Deploying Manifests and Helm Charts
• Using the Helm CRD
• Customizing Packaged Components with HelmChartConfig
• Upgrading from Helm v2

Automatically Deploying Manifests and Helm Charts

Any Kubernetes manifests found in /var/lib/rancher/k3s/server/manifests will automatically be deployed to K3s in a manner similar to kubectl apply. Manifests deployed in this manner are managed as AddOn custom resources, and can be viewed by running kubectl get addon -A. You will find AddOns for packaged components such as CoreDNS, Local-Storage, Traefik, etc. AddOns are created automatically by the deploy controller, and are named based on their filename in the manifests directory.

It is also possible to deploy Helm charts as AddOns. K3s includes a Helm Controller that manages Helm charts using a HelmChart Custom Resource Definition (CRD).

Using the Helm CRD

Note: K3s versions through v0.5.0 used k3s.cattle.io/v1 as the apiVersion for HelmCharts. This has been changed to helm.cattle.io/v1 for later versions.

The HelmChart resource definition captures most of the options you would normally pass to the helm command-line tool. Here’s an example of how you might deploy Grafana from the default chart repository, overriding some of the default chart values. Note that the HelmChart resource itself is in the kube-system namespace, but the chart’s resources will be deployed to the monitoring namespace.

apiVersion: helm.cattle.io/v1
kind: HelmChart
metadata:
  name: grafana
  namespace: kube-system
spec:
  chart: stable/grafana
  targetNamespace: monitoring
  set:
    adminPassword: "NotVerySafePassword"
  valuesContent: |-
    image:
      tag: master
    env:
      GF_EXPLORE_ENABLED: true
    adminUser: admin
    sidecar:
      datasources:
        enabled: true

HelmChart Field Definitions

Content placed in /var/lib/rancher/k3s/server/static/ can be accessed anonymously via the Kubernetes APIServer from within the cluster. This URL can be templated using the special variable %{KUBERNETES_API}% in the spec.chart field. For example, the packaged Traefik component loads its chart from https://%{KUBERNETES_API}%/static/charts/traefik-1.81.0.tgz.

Customizing Packaged Components with HelmChartConfig

To allow overriding values for packaged components that are deployed as HelmCharts (such as Traefik), K3s versions starting with v1.19.0+k3s1 support customizing deployments via a HelmChartConfig resources. The HelmChartConfig resource must match the name and namespace of its corresponding HelmChart, and supports providing additional valuesContent, which is passed to the helm command as an additional value file.

Note: HelmChart spec.set values override HelmChart and HelmChartConfig spec.valuesContent settings.

For example, to customize the packaged Traefik ingress configuration, you can create a file named /var/lib/rancher/k3s/server/manifests/traefik-config.yaml and populate it with the following content:

apiVersion: helm.cattle.io/v1
kind: HelmChartConfig
metadata:
  name: traefik
  namespace: kube-system
spec:
  valuesContent: |-
    image: traefik
    imageTag: v1.7.26-alpine
    proxyProtocol:
      enabled: true
      trustedIPs:
        - 10.0.0.0/8
    forwardedHeaders:
      enabled: true
      trustedIPs:
        - 10.0.0.0/8
    ssl:
      enabled: true
      permanentRedirect: false

Upgrading from Helm v2

Note: K3s versions starting with v1.17.0+k3s.1 support Helm v3, and will use it by default. Helm v2 charts can be used by setting helmVersion: v2 in the spec.

If you were using Helm v2 in previous versions of K3s, you may upgrade to v1.17.0+k3s.1 or newer and Helm 2 will still function. If you wish to migrate to Helm 3, this blog post by Helm explains how to use a plugin to successfully migrate. Refer to the official Helm 3 documentation here for more information. K3s will handle either Helm v2 or Helm v3 as of v1.17.0+k3s.1. Just be sure you have properly set your kubeconfig as per the examples in the section about cluster access.

Note that Helm 3 no longer requires Tiller and the helm init command. Refer to the official documentation for details.