Kubernetes workshop: Local environment
Prepare your local Kubernetes environment
Goal: Having a running local Kubernetes environment
Ensure that you fulfilled the requirements
Using k3d to create a cluster
In order to have an easily provisioned temporary playground we’ll make use of k3d which is a lightweight local Kubernetes instance.
(Note that they are alternatives to run a local Kubernetes cluster such as: kubeadm, microk8s, minikube)
After installing the binary you should enable the completion (bash or zsh) as follows (do the same for both kubectl and k3d).
1source <(k3d completion bash)
Then create the sandbox cluster named "workshop" with an additional worker
1k3d cluster create workshop -p "8081:80@loadbalancer" --agents 1
2INFO[0000] Prep: Network
3INFO[0000] Created network 'k3d-workshop' (ce74508d3fe09d8622f1ae83effd412d754dfdb441aa9d550723805f9b528c6b)
4INFO[0000] Created volume 'k3d-workshop-images'
5INFO[0001] Creating node 'k3d-workshop-server-0'
6INFO[0001] Creating node 'k3d-workshop-agent-0'
7INFO[0001] Creating LoadBalancer 'k3d-workshop-serverlb'
8INFO[0001] Starting cluster 'workshop'
9INFO[0001] Starting servers...
10INFO[0001] Starting Node 'k3d-workshop-server-0'
11INFO[0006] Starting agents...
12INFO[0006] Starting Node 'k3d-workshop-agent-0'
13INFO[0018] Starting helpers...
14INFO[0018] Starting Node 'k3d-workshop-serverlb'
15INFO[0019] (Optional) Trying to get IP of the docker host and inject it into the cluster as 'host.k3d.internal' for easy access
16INFO[0023] Successfully added host record to /etc/hosts in 3/3 nodes and to the CoreDNS ConfigMap
17INFO[0023] Cluster 'workshop' created successfully!
18INFO[0023] --kubeconfig-update-default=false --> sets --kubeconfig-switch-context=false
19INFO[0023] You can now use it like this:
20kubectl config use-context k3d-workshop
21kubectl cluster-info
As k3d is made to be used on top of docker you can see the status of the running containers. You should have 3 containers, one for the loadbalancing, one for the control-plane and an agent (worker).
1docker ps
2CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
34b5847b265dd rancher/k3d-proxy:v4.4.6 "/bin/sh -c nginx-pr…" About a minute ago Up About a minute 80/tcp, 0.0.0.0:43903->6443/tcp k3d-workshop-serverlb
4523a025087b3 rancher/k3s:v1.21.1-k3s1 "/bin/entrypoint.sh …" About a minute ago Up About a minute k3d-workshop-agent-0
5791b8a69bc1f rancher/k3s:v1.21.1-k3s1 "/bin/entrypoint.sh …" About a minute ago Up About a minute k3d-workshop-server-0
With kubectl you'll see 2 running pods: a control-plane and a worker
1kubectl get nodes
2NAME STATUS ROLES AGE VERSION
3k3d-workshop-agent-0 Ready <none> 2m34s v1.21.1+k3s1
4k3d-workshop-server-0 Ready control-plane,master 2m44s v1.21.1+k3s1
You can also have a look to the default cluster's components that are all located in the namespace kube-system
1kubectl get pods -n kube-system
2NAME READY STATUS RESTARTS AGE
3helm-install-traefik-crd-h5j7m 0/1 Completed 0 16h
4helm-install-traefik-8mzhk 0/1 Completed 0 16h
5svclb-traefik-gh4rk 2/2 Running 2 16h
6traefik-97b44b794-lcmh4 1/1 Running 1 16h
7coredns-7448499f4d-h7xvn 1/1 Running 2 16h
8local-path-provisioner-5ff76fc89d-qvpf7 1/1 Running 1 16h
9svclb-traefik-cbvmp 2/2 Running 2 16h
10metrics-server-86cbb8457f-5v9ls 1/1 Running 1 16h
The CLI configuration
The main interface to the Kubernetes API is kubectl. This CLI is configured with what we call a kubeconfig
you can have a look at its content wether by having a look at its default location is ~/.kube/config or running the command
1kubectl config view
2apiVersion: v1
3clusters:
4- cluster:
5 certificate-authority-data: DATA+OMITTED
6 server: https://0.0....
and you can check if the CLI is properly configured by running
1kubectl cluster-info
2Kubernetes control plane is running at https://0.0.0.0:43903
3CoreDNS is running at https://0.0.0.0:43903/api/v1/namespaces/kube-system/services/kube-dns:dns/proxy
4Metrics-server is running at https://0.0.0.0:43903/api/v1/namespaces/kube-system/services/https:metrics-server:/proxy
Kubectl plugins
It is really easy to extend the capabilities of he kubectl CLI.
Here is a basic "hello-world" example:
Write a dumb script, just ensure its name is prefixed with kubectl- and put it in your PATH
1cat > kubectl-helloworld<<EOF
2#!/bin/bash
3echo "Hello world!"
4EOF
5
6chmod u+x kubectl-helloworld && sudo mv kubectl-helloworld /usr/local/bin
Then it can be used as an argument of kubectl
1kubectl helloworld
2Hello world!
Delete our test
1sudo rm /usr/local/bin/kubectl-helloworld
You can find more information on how to create a kubectl plugin here
In order to benefit from the plugins written by the community there's a tool named krew
Update the local index
1kubectl krew update
2Adding "default" plugin index from https://github.com/kubernetes-sigs/krew-index.git.
3Updated the local copy of plugin index.
Browse the available plugins
1kubectl krew search
2NAME DESCRIPTION INSTALLED
3access-matrix Show an RBAC access matrix for server resources no
4advise-psp Suggests PodSecurityPolicies for cluster. no
5allctx Run commands on contexts in your kubeconfig no
6apparmor-manager Manage AppArmor profiles for cluster. no
7...
For the current workshop we'll make use of ctx ns
ctx: Switch between contexts in your kubeconfig (Really helpful when you have multiple clusters to manage)ns: Switch between Kubernetes namespaces (Avoid to specify the namespace for each kubectl commands when working on a given namespace)
1kubectl krew install ctx ns
2Updated the local copy of plugin index.
3Installing plugin: ctx
4...
Then you'll be able to switch between contexts (clusters) and namespaces.
1kubectl ns
2Context "k3d-workshop" modified.
3Active namespace is "kube-system".