Building a K8s Cluster with Kubeadm

We build a K8s cluster for managing containers and use Kubeadm to simplify the proess of setting up a simple cluster.

Install Docker on all three nodes

Add the Docker GPG Key:

curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo apt-key add -

Add the Docker Repository

sudo add-apt-repository \

"deb [arch=amd64] https://download.docker.com/linux/ubuntu \

$(lsb_release -cs) \

stable"

Update packages:

sudo apt-get update

Install Docker:

sudo apt-get install -y docker-ce=18.06.1~ce~3-0~ubuntu

Hold Docker at this specific version:

sudo apt-mark hold docker-ce

Verify that Docker is up and running with:

sudo systemctl status docker

After running above commands, the Docker service status should be active (running).

● docker.service - Docker Application Container Engine

   Loaded: loaded (/lib/systemd/system/docker.service; enabled; vendor preset: enabled)

   Active: active (running) since Sun 2019-10-20 11:15:32 UTC; 19s ago

     Docs: https://docs.docker.com

 Main PID: 9869 (dockerd)

    Tasks: 21

   CGroup: /system.slice/docker.service

           ├─9869 /usr/bin/dockerd -H fd://

           └─9894 docker-containerd --config /var/run/docker/containerd/containerd.toml

Install Kubeadm, Kubelet, and Kubectl

Install the K8s components by running the following commands on all three nodes.

Add the K8s GPG Key

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

Add the K8s repo

cat << EOF | sudo tee /etc/apt/sources.list.d/kubernetes.list

deb https://apt.kubernetes.io/ kubernetes-xenial main

EOF

Update packages

sudo apt-get update

Install kubelet, kubeadm and kubectl

sudo apt-get install -y kubelet=1.12.7-00 kubeadm=1.12.7-00 kubectl=1.12.7-00

Hold the Kubernetes components at this specific version

sudo apt-mark hold kubelet kubeadm kubectl

Bootstrap the cluster on the Kube master node

Initialize kubeadm on the master node

sudo kubeadm init --pod-network-cidr=10.244.0.0/16

After a few minutes, you should see a kubeadm join command that will be used later

kubeadm join 10.0.1.101:6443 --token ioxxtp.zugcxykam7jhmlqe --discovery-token-ca-cert-hash sha256:1feab8ca98d50689b5a524c1271b43a7c712d66dab0d6ab7b68c9fd472921731

Set up the local kubeconfig

mkdir -p $HOME/.kube

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

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

Verify if Kube master node is up and running

kubectl version

You should see Client Version and Server Version as below

Client Version: version.Info{Major:"1", Minor:"12", GitVersion:"v1.12.7", GitCommit:"6f482974b76db3f1e0f5d24605a9d1d38fad9a2b", GitTreeState:"clean", BuildDate:"2019-03-25T02:52:13Z", GoVersion:"go1.10.8", Compiler:"gc", Platform:"linux/amd64"}

Server Version: version.Info{Major:"1", Minor:"12", GitVersion:"v1.12.10", GitCommit:"e3c134023df5dea457638b614ee17ef234dc34a6", GitTreeState:"clean", BuildDate:"2019-07-08T03:40:54Z", GoVersion:"go1.10.8", Compiler:"gc", Platform:"linux/amd64"}

Join the two Kube worker nodes to the cluster

Once the Kube master node is ready, then we need to join those two Kube worker nodes to the cluster.

Copy the kubeadm join command that was printed by kubeadm init command in the previous step. Make sure you run it with sudo

sudo kubeadm join 10.0.1.101:6443 --token ioxxtp.zugcxykam7jhmlqe --discovery-token-ca-cert-hash sha256:1feab8ca98d50689b5a524c1271b43a7c712d66dab0d6ab7b68c9fd472921731

Go back to the Kube master node, check if the nodes are joined the cluster successfully or not

kubectl get nodes

Verify the result. Three of nodes are expected to be here but in the NotReady state.

 NAME            STATUS     ROLES    AGE   VERSION

 ip-10-0-1-101   NotReady   master   30s   v1.12.2

 ip-10-0-1-102   NotReady   <none>   8s    v1.12.2

 ip-10-0-1-103   NotReady   <none>   5s    v1.12.2

Setu up cluster networking

To get them ready, we need to use flannel because K8s does not provide any defalt network implementation. Flannel is a very simple overlay network that satisfies the Kubernetes requirements. Many people have reported success with Flannel and Kubernetes.

Turn on iptables bridge calls on all three nodes

echo "net.bridge.bridge-nf-call-iptables=1" | sudo tee -a /etc/sysctl.conf

sudo sysctl -p

Apply flannel on Kube master node

kubectl apply -f https://raw.githubusercontent.com/coreos/flannel/bc79dd1505b0c8681ece4de4c0d86c5cd2643275/Documentation/kube-flannel.yml

Once flannel is installed. Verifiy the node status.

kubectl get nodes

After a short time, all three nodes should be in the Ready state.

NAME            STATUS   ROLES    AGE   VERSION

ip-10-0-1-101   Ready    master   85s   v1.12.2

ip-10-0-1-102   Ready    <none>   63s   v1.12.2

ip-10-0-1-103   Ready    <none>   60s   v1.12.2