Kubernetes Cluster Installation

 Execute following script on all Kubernetes Nodes (Control Node and Worker Nodes)

Login to every node and run following (I am using Ubuntu 20.04)

cat > k8s.sh

sudo apt update

sudo apt -y full-upgrade

[ -f /var/run/reboot-required ] && sudo reboot -f

swapoff -a

sed -i '/ swap / s/^\(.*\)$/#\1/g' /etc/fstab

echo $1 > /etc/hostname

hostnamectl set-hostname $1

echo -e "overlay\nbr_netfilter" > /etc/modules-load.d/containerd.conf

modprobe overlay

modprobe br_netfilter

echo -e "net.bridge.bridge-nf-call-iptables = 1\nnet.ipv4.ip_forward = 1\nnet.bridge.bridge-nf-call-ip6tables = 1" > /etc/sysctl.d/99-kubernetes-cri.conf

sysctl --system

apt-get update && apt-get install -y containerd

mkdir /etc/containerd && containerd config default | tee /etc/containerd/config.toml

systemctl restart containerd.service

systemctl enable containerd

systemctl status  containerd

apt-get install -y apt-transport-https curl

curl  -s https://packages.cloud.google.com/apt/doc/apt-key.gpg | apt-key add -

echo "deb https://apt.kubernetes.io/ kubernetes-xenial main" > /etc/apt/sources.list.d/kubernetes.list

apt-get update

apt-get install -y kubelet=1.23.0-00 kubeadm=1.23.0-00 kubectl=1.23.0-00

## Do not update or change below packages and put on hold

apt-mark hold kubelet kubeadm kubectl

systemctl enable kubelet

if [ "$1" == "control" ]

then

    ## Do not update or change below packages and put on hold

    apt-mark hold kubelet kubeadm kubectl

    ## Setup cluster on Control Plane

    kubeadm init --pod-network-cidr 192.168.0.0/16 --kubernetes-version 1.23.0

    mkdir -p $HOME/.kube

    sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config

    sudo chown $(id -u):$(id -g) $HOME/.kube/config

    kubectl create -f https://docs.projectcalico.org/manifests/tigera-operator.yaml

    kubectl create -f https://docs.projectcalico.org/manifests/custom-resources.yaml

fi

This script requires 1 Argument (NO validation has been done in script) which is used to update HOSTNAME on servers like (control, worker1, worker2)

NOTE: I have used CALICO for CNI network plugin

For Control Plane node

bash k8s.sh control

For Worker nodes

bash k8s.sh worker1

Now login to Control node and run following commands to verify cluster setup correctly

kubectl get nodes

NAME      STATUS     ROLES                  AGE   VERSION

control   NotReady   control-plane,master   54s   v1.23.0

kubectl get pods --all-namespaces

calico-system     calico-kube-controllers-69cfd64db4-5q6w8   1/1     Running   0          44s

calico-system     calico-node-s82tv                          1/1     Running   0          44s

calico-system     calico-typha-7785554894-6grh4              1/1     Running   0          44s

kube-system       coredns-64897985d-wzd4s                    1/1     Running   0          55s

kube-system       coredns-64897985d-xq59q                    1/1     Running   0          55s

kube-system       etcd-control                               1/1     Running   0          64s

kube-system       kube-apiserver-control                     1/1     Running   0          64s

kube-system       kube-controller-manager-control            1/1     Running   0          65s

kube-system       kube-proxy-gkdv7                           1/1     Running   0          55s

kube-system       kube-scheduler-control                     1/1     Running   0          65s

tigera-operator   tigera-operator-7d8c9d4f67-7dfnb           1/1     Running   0          55s

Run following command to get the K8s cluster token

kubeadm token create --print-join-command

kubeadm join 172.31.20.221:6443 --token 743g1x.3shw8jgfxb70l7xd --discovery-token-ca-cert-hash sha256:8d16a1aa8afbd56bb4eaa8dfbf25fcb5147207550e39268efa4a04f1a2483f77

###Copy the output of above command

###Login to Worker nodes and execute copied command (cluster join command)

###Once done with above command, GOTO CONTROL Server and check the output of nodes if WORKER Nodes Joined the cluster

root@control:~# kubectl get nodes

NAME      STATUS     ROLES                  AGE   VERSION

control   Ready      control-plane,master   13m   v1.23.0

node1     NotReady   <none>                 17s   v1.23.0

node2     NotReady   <none>                 1s    v1.23.0

root@control:~# kubectl get nodes

NAME      STATUS   ROLES                  AGE   VERSION

control   Ready    control-plane,master   14m   v1.23.0

node1     Ready    <none>                 83s   v1.23.0

node2     Ready    <none>                 67s   v1.23.0

Your cluster is Ready!!