Kubernetes-based Setup on AWS

Pre-requisites

The pre-requisites for VSP installation:

1. EC2 machine with:
   1. kubectl
   2. helm
   3. awscli
   4. docker

The pre-requisites for Kubernetes engine deployment on AWS EC2

1. AWS IAM User account with the role
2. AmazonEC2FullAccess

AWS EC2 for Kubernetes Architecture

EC2 Machine Creation

1. Access the AWS Dashboard: https://console.aws.amazon.com/ec2 using valid credentials
2. Navigate to EC2 > Instances > Launch an instance. Provide an appropriate Name
3. Under the Quick Start tab, select Ubuntu
4. Select Instance type as required for CMS. Example: t2.medium
5. Select the Number of instances as 3
6. Click Create new key pair if required for authentication credentials
7. Configure the Storage information as required and click Launch Instance
8. Modify the Security Group for existing machines as below
   1. Click on the Security Group ID and modify the Inbound rule by clicking Edit Inbound rules
   2. Modify the Security Group details as below
      NOTE

      Public IP may change every time. If it is not possible to log in to the machines, change the Source to MYIP and try to log in
9. Log in to all the three EC2 machines and install the required software to provision Kubernetes Clusters
   1. Modify the hostname for all the three machines:

      hostnamectl set-hostname <hostname>

      
      

Installation

1. Execute the commands below to install the required kubernetes software in all the three machines

   sudo su
   sudo apt-get update
   sudo apt-get install -y apt-transport-https ca-certificates curl
   sudo curl -fsSLo /usr/share/keyrings/kubernetes-archive-keyring.gpg https://packages.cloud.google.com/apt/doc/apt-key.gpg 
   echo "deb [signed-by=/usr/share/keyrings/kubernetes-archive-keyring.gpg] https://apt.kubernetes.io/ kubernetes-xenial main" | sudo tee /etc/apt/sources.list.d/kubernetes.list
   sudo apt-get install -y kubelet=<Kubernetes_Version>  kubeadm=<Kubernetes_Version> kubectl=<Kubernetes_Version> #Example: sudo apt-get install -y kubelet=1.19.0-00 kubeadm=1.19.0-00 kubectl=1.19.0-00
   apt update 
   apt install docker.io -y
   
   

2. Execute the commands below to provision Kubernetes Control plane on the master machine

   sudo kubeadm init --pod-network-cidr=<CIDR_Range_for_IPAddress> #Example: sudo kubeadm init --pod-network-cidr=10.244.0.0/16
   mkdir -p $HOME/.kube
   sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
   sudo chown $(id -u):$(id -g) $HOME/.kube/config
   kubectl apply -f https://github.com/coreos/flannel/raw/master/Documentation/kube-flannel.yml
   kubectl get nodes #To list nodes
   

3. Execute the command below to generate a join command to execute on the worker nodes

   kubeadm token create –print-join-command

4. On the Worker machines, execute the join command so that the machines join as workers in the Kubernetes cluster
   1. Example: kubeadm join 172.31.2.84:6443 --token fh9lqg.4ko0gz16owx1r8i6 --discovery-token-ca-cert-hash sha256:02fba06cea1124f3639e6aebbcf4f01464bd9dc22512fe50e208be8ed4d56582
5. On the master machine, execute the command below to list the nodes

   kubectl get nodes

6. The Kubernetes cluster is up and running for workload deployment