k3d

k3d is a lightweight wrapper to run k3s (Rancher Lab’s minimal Kubernetes distribution) in docker. k3d makes it very easy to create single- and multi-node k3s clusters in docker, e.g. for local development on Kubernetes.

Prerequisites

  • To use k3d, you will also need to install docker.
  • Install the latest version of k3d.
  • To interact with the Kubernetes cluster kubectl
  • (Optional) Helm is the package manager for Kubernetes

Installation

  1. Create a cluster and disable Traefik with the following command:

    Zip

    1. $ k3d cluster create --api-port 6550 -p "9080:80@loadbalancer" -p "9443:443@loadbalancer" --agents 2 --k3s-arg '--disable=traefik@server:*'
  2. To see the list of k3d clusters, use the following command:

    1. $ k3d cluster list
    2. k3s-default
  3. To list the local Kubernetes contexts, use the following command.

    1. $ kubectl config get-contexts
    2. CURRENT NAME CLUSTER AUTHINFO NAMESPACE
    3. * k3d-k3s-default k3d-k3s-default k3d-k3s-default

    k3d- is prefixed to the context and cluster names, for example: k3d-k3s-default

  4. If you run multiple clusters, you need to choose which cluster kubectl talks to. You can set a default cluster for kubectl by setting the current context in the Kubernetes kubeconfig file. Additionally you can run following command to set the current context for kubectl.

    1. $ kubectl config use-context k3d-k3s-default
    2. Switched to context "k3d-k3s-default".

Set up Istio for k3d

  1. Once you are done setting up a k3d cluster, you can proceed to install Istio with Helm 3 on it.

    1. $ kubectl create namespace istio-system
    2. $ helm install istio-base istio/base -n istio-system --wait
    3. $ helm install istiod istio/istiod -n istio-system --wait
  2. (Optional) Install an ingress gateway:

    1. $ kubectl label namespace istio-system istio-injection=enabled
    2. $ helm install istio-ingressgateway istio/gateway -n istio-system --wait

Set up Dashboard UI for k3d

k3d does not have a built in Dashboard UI like minikube. But you can still set up Dashboard, a web based Kubernetes UI, to view your cluster. Follow these instructions to set up Dashboard for k3d.

  1. To deploy Dashboard, run the following command:

    1. $ GITHUB_URL=https://github.com/kubernetes/dashboard/releases
    2. $ VERSION_KUBE_DASHBOARD=$(curl -w '%{url_effective}' -I -L -s -S ${GITHUB_URL}/latest -o /dev/null | sed -e 's|.*/||')
    3. $ kubectl apply -f https://raw.githubusercontent.com/kubernetes/dashboard/${VERSION_KUBE_DASHBOARD}/aio/deploy/recommended.yaml
  2. Verify that Dashboard is deployed and running.

    1. $ kubectl get pod -n kubernetes-dashboard
    2. NAME READY STATUS RESTARTS AGE
    3. dashboard-metrics-scraper-8c47d4b5d-dd2ks 1/1 Running 0 25s
    4. kubernetes-dashboard-67bd8fc546-4xfmm 1/1 Running 0 25s
  3. Create a ServiceAccount and ClusterRoleBinding to provide admin access to the newly created cluster.

    1. $ kubectl create serviceaccount -n kubernetes-dashboard admin-user
    2. $ kubectl create clusterrolebinding -n kubernetes-dashboard admin-user --clusterrole cluster-admin --serviceaccount=kubernetes-dashboard:admin-user
  4. To log in to your Dashboard, you need a Bearer Token. Use the following command to store the token in a variable.

    1. $ token=$(kubectl -n kubernetes-dashboard create token admin-user)

    Display the token using the echo command and copy it to use for logging in to your Dashboard.

    1. $ echo $token
  5. You can access your Dashboard using the kubectl command-line tool by running the following command:

    1. $ kubectl proxy
    2. Starting to serve on 127.0.0.1:8001

    Click Kubernetes Dashboard to view your deployments and services.

    You have to save your token somewhere, otherwise you have to run step number 4 everytime you need a token to log in to your Dashboard.

Uninstall

  1. When you are done experimenting and you want to delete the existing cluster, use the following command:

    1. $ k3d cluster delete k3s-default
    2. Deleting cluster "k3s-default" ...