Kubernetes Classroom Series – 12/Oct/2019

Kubernetes Options for Installation

1. Bare-Metal Installations:
   • Start from OS
   • Install Every Component
   • K8s the hard way
2. Other Managed Installations:
   • Start from OS
   • You need not install every component
   • Popular options
     • minikube (developer)
     • micro k8s (developer)
     • kube-admin
     • kops
3. Cloud-Provider Installations:
   • You will get k8s as a service.
   • You need to take care of nodes. Masters generally will be managed by cloud-providers
   • Popular options:
     • GKE
     • AKS
     • EKS Refer Here for complete list of providers

Installation Steps

• Setup Details:

  • Four ubuntu 16.04 machines
  • One will be master
  • other 3 will nodes
  • Ensure all the four machines can communicate with each other
  • We will be kubeadm and we will follow official k8s documentation over here

• Login into all nodes, become root user and install docker using steps from here

• Install the kubelet, kubeadm and kubctl on all nodes from here

• Necessary software’s are present on your nodes

• we need to create k8s cluster. For creating cluster. Refer documentation from here

• Login into master node as a root user

kubeadm init

• If kubeadm is able to install succesfully, it gives some info at the end which looks like

To start using your cluster, you need to run the following as a regular user:

  mkdir -p $HOME/.kube
  sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
  sudo chown $(id -u):$(id -g) $HOME/.kube/config

You should now deploy a pod network to the cluster.
Run "kubectl apply -f [podnetwork].yaml" with one of the options listed at:
  https://kubernetes.io/docs/concepts/cluster-administration/addons/

Then you can join any number of worker nodes by running the following on each as root:

kubeadm join 172.31.29.1:6443 --token o38veh.5g6zhgfaafg5zzk3 \
    --discovery-token-ca-cert-hash sha256:bd735d51e2beceaa9c4aba6f32ef6d91dabdfb51d93a9f8bcb41c51e9c5f45d9

• Pod networking is enabled only after you install pod network.

kubectl apply -f "https://cloud.weave.works/k8s/net?k8s-version=$(kubectl version | base64 | tr -d '\n')"

• Now login into nodes as a root user and execute kubeadm join command with necessary arguments which you have copied.

Basic Commands to execute after kubeadm installation in master

kubctl get nodes
kubectl get nodes -w # watch the output changes 
kubectl get nodes -o wide # more info
kubectl get pods # List all the pods you created
kubectl get pods --all-namespaces   # Lists all the pods in the cluster irrespective of who created

Can I use this setup in Enterprise?

NO Why?? Problem with current setup is we have single node for master. To use it in staging/prod environments, master needs to be Highly available.

For that refer below * options * steps

What is that I create in Kubernetes

Workers

• Pod:
  • atomic unit of work in k8s

Controllers

• Control the pods
• Eg: Deployment, Statefulset, Daemonsets