Extensions

Extensions allow you to specify Helm charts or yaml manifests to be deployed during the creation or update of a cluster. We will illustrate the extensions’ capability on a single node cluster created with the k0s binary.

Pre-requisites

Creation of a VM

First create a single VM, named k0s-1, using Multipass:

multipass launch -n k0s-1

Extensions with Helm Charts in k0s configuration

Run a shell into the newly created VM:

$ kubectl shell k0s-1

Next download k0s:

ubuntu@k0s-1:~$ curl -sSLf get.k0s.sh | sudo sh

In the single node tutorial we talked about the default k0s configuration file which can be retrieved with the following command:

ubuntu@k0s-1:~$ k0s default-config > k0s.yaml

This file contains the default properties of all the cluster’s components, among them:

  • the API Server
  • the ETCD key value store
  • the network plugins
  • …

The following shows all the available properties

$ cat k0s.yaml
apiVersion: k0s.k0sproject.io/v1beta1
kind: Cluster
metadata:
  name: k0s
spec:
  api:
    address: 192.168.64.113
    port: 6443
    k0sApiPort: 9443
    sans:
    - 192.168.64.113
  storage:
    type: etcd
    etcd:
      peerAddress: 192.168.64.113
  network:
    podCIDR: 10.244.0.0/16
    serviceCIDR: 10.96.0.0/12
    provider: kuberouter
    calico: null
    kuberouter:
      mtu: 0
      peerRouterIPs: ""
      peerRouterASNs: ""
      autoMTU: true
    kubeProxy:
      disabled: false
      mode: iptables
  podSecurityPolicy:
    defaultPolicy: 00-k0s-privileged
  telemetry:
    enabled: true
  installConfig:
    users:
      etcdUser: etcd
      kineUser: kube-apiserver
      konnectivityUser: konnectivity-server
      kubeAPIserverUser: kube-apiserver
      kubeSchedulerUser: kube-scheduler
  images:
    konnectivity:
      image: us.gcr.io/k8s-artifacts-prod/kas-network-proxy/proxy-agent
      version: v0.0.21
    metricsserver:
      image: gcr.io/k8s-staging-metrics-server/metrics-server
      version: v0.3.7
    kubeproxy:
      image: k8s.gcr.io/kube-proxy
      version: v1.21.2
    coredns:
      image: docker.io/coredns/coredns
      version: 1.7.0
    calico:
      cni:
        image: docker.io/calico/cni
        version: v3.18.1
      node:
        image: docker.io/calico/node
        version: v3.18.1
      kubecontrollers:
        image: docker.io/calico/kube-controllers
        version: v3.18.1
    kuberouter:
      cni:
        image: docker.io/cloudnativelabs/kube-router
        version: v1.2.1
      cniInstaller:
        image: quay.io/k0sproject/cni-node
        version: 0.1.0
    default_pull_policy: IfNotPresent
  konnectivity:
    agentPort: 8132
    adminPort: 8133

Under the .spec property, add the following extensions property and its helm sub property:

extensions:
  helm:
    repositories:
    - name: traefik
      url: https://helm.traefik.io/traefik
    charts:
    - name: traefik
      chartname: traefik/traefik
      version: "9.11.0"
      namespace: default

This indicates to k0s that the Helm Chart for Traefik Ingress Controller needs to be installed during the bootstrap of the cluster.

Next, install the cluster providing this configuration file:

ubuntu@k0s-1:~$ sudo k0s install controller --config ./k0s.yaml --single

and start it:

ubuntu@k0s-1:~$ sudo k0s start

Finally, make sure Traefik has been deployed to the cluster:

ubuntu@k0s-1:~$ sudo k0s kubectl get po -A

You should get a result similar to the following one:

NAMESPACE     NAME                                  READY   STATUS    RESTARTS   AGE
default       traefik-1625417854-7f8d8d8987-p4z96   1/1     Running   0          2m58s
kube-system   coredns-5ccbdcc4c4-kb24b              1/1     Running   0          2m58s
kube-system   kube-proxy-t8jn9                      1/1     Running   0          2m44s
kube-system   kube-router-vmhv7                     1/1     Running   0          2m44s
kube-system   metrics-server-59d8698d9-259xl        1/1     Running   0          2m58s

In order to verify Traefik has been installed via Helm, first install the helm client using the following command:

ubuntu@k0s-1:~$ curl -o helm.tar.gz https://get.helm.sh/helm-v3.6.2-linux-amd64.tar.gz && \
                tar -xvf helm.tar.gz && \
                sudo mv linux-amd64/helm /usr/local/bin

Then run a root shell (that will make things easier in this example) and configure the helm client so it uses the kubeconfig file of the cluster:

ubuntu@k0s-1:~$ sudo su -
root@k0s-1:~$ export KUBECONFIG=/var/lib/k0s/pki/admin.conf

Finally, make sure Traefik Helm chart is listed:

root@k0s-1:~$ helm list

You should get an output similar to the following one:

NAME                    NAMESPACE       REVISION        UPDATED                                         STATUS          CHART           APP VERSION
traefik-1625417854      default         1               2021-07-04 18:57:37.035554595 +0200 CEST        deployed        traefik-9.11.0  2.3.3

As you can see, using the extensions property makes it very easy to deploy additional application through Helm Chart.

Extensions with yaml manifests

K0s also allows to deploy additional resources automatically by adding yaml manifests into the /var/lib/k0s/manifests/ folder.

To illustrate this, first create a mongo folder in /var/lib/k0s/manifests/ and create the following Deployment (based on the mongodb image) in the file mongo.yaml inside that new folder:

apiVersion: apps/v1
kind: Deployment
metadata:
  name: db
  namespace: default
spec:
  replicas: 1
  selector:
    matchLabels:
      app: db
  template:
    metadata:
      creationTimestamp: null
      labels:
        app: db
    spec:
      containers:
      - image: mongo:4.4
        name: mongo

After a few tens of seconds, make sure the Deployment has been correctly created:

ubuntu@k0s-1:~$ sudo k0s kubectl get deploy,po -l app=db