Setup-kubernetes-cluster

Launch 3 EC2 servers.

Configure first as a master node and the other two as a slave.

Deploy one WordPress website

Addition

Attach persistent volume over the DB

Step 1 - Launch 3 ec2 Instance - one for master and 2 for slave

for master , choose the instance type t2xlarge and for the two slave , choose the t2small type

Step 2 - Configuring the master and slave nodes

On All EC2, the master and worker nodes:

Install and enable the docker service so that the docker service starts on system restarts.

yum install docker -y
systemctl enable docker --now

Create proper yum repo files so that we can use yum commands to install the components of Kubernetes.

cat <<EOF | sudo tee /etc/yum.repos.d/kubernetes.repo
[kubernetes]
name=Kubernetes
baseurl=https://packages.cloud.google.com/yum/repos/kubernetes-el7-\$basearch
enabled=1
gpgcheck=1
repo_gpgcheck=0
gpgkey=https://packages.cloud.google.com/yum/doc/yum-key.gpg https://packages.cloud.google.com/yum/doc/rpm-package-key.gpg
exclude=kubelet kubeadm kubectl
EOF

vim /etc/docker/daemon.json

{ "exec-opts": ["native.cgroupdriver=systemd"] }

sysctl command is used to modify kernel parameters at runtime. Kubernetes needs to have access to kernel’s IP6 table and so we need to do some more modifications

cat <<EOF | sudo tee /etc/sysctl.d/k8s.conf
net.bridge.bridge-nf-call-ip6tables = 1
net.bridge.bridge-nf-call-iptables = 1
EOF

Install kubelet, kubeadm and kubectl; start kubelet daemon

yum install -y kubelet kubeadm kubectl --disableexcludes=kubernetes
systemctl enable --now kubelet
yum install tc
systemctl restart docker
sudo sysctl --system

Only on the Master Node:

kubeadm config images pull
kubeadm init  --pod-network-cidr=10.244.0.0/16{network plugin cidr range}

Wait for few minutes. Following commands will appear on the output screen of the master node

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

This will also generate a kubeadm join command with some tokens. save the token at a place as it will be going to used for joining the worker with the master

On the Worker Node

kubeadm join 172.31.45.77:6443 --token ajrjbr.ovvaz6kohgyb0jr0 \
        --discovery-token-ca-cert-hash sha256:XYZXYZXYZXYZ

Now, wait for few minutes to let the pods in ready state

kubectl get nodes

Step 3 - Deploy one WordPress website

Now, for the wordpress, we will need a database , so here we are using mysql database

for this, we will create a file , I named it as mysql-deploy

For, Volume Claim , I created a file pv.yml and pvc.yml and pass.yml for secrets and then one file named as wordpress.yml for the wordpress deployment on the cluster

all these files are given in the same repo

Now, after creating these files

Now, will create these with kubectl command

kubectl create -f pass.yml
kubectl create -f pv.yml
kubectl create -f pvc.yml
kubectl create -f mysql-deploy.yml
kubectl create -f wordpress-deploy.yml

Now, will check the deployments, pods and nodes from the master

kubectl get pods

kubectl get svc

kubectl get pv

Note - if you are not able to see the nodes in the slave but able to see in ready state in master node , then run the below command in both the slaves -

export KUBECONFIG=/etc/kubernetes/kubelet.conf

and then you will be able to see the nodes in the slaves as well

kubectl config view 
kubectl get nodes

But this will be the temporary solution, we will again not able to see these after the restart so there is a file named as rc.d , we will add the same command in the file , so that it will automatically runs after the restart

cd /etc/rc.d/
ls
vim rc.local

after that , give the permission to run this

chmod +x rc.local

THANK YOU

SHRUTI VIJ

shrutiVij1502/Setup-kubernetes-cluster