All-Things-Docker-and-Kubernetes

Lab 030: Building Highly-Available Kubernetes Cluster

TODO

Pre-requisites

• Basic Understanding of Kubernetes
• AWS account
• AWS IAM Requirements

• AWS CLI, kubectl, and eksctl

• Additional CLI utilities:
  • yq
  • jq

Introduction

The main goals for this lab are:

• Build a highly available etc cluster
• Bootstrap the HA Kubernetes cluster with kubeadm
• Front the cluster with HAProxy loadbalancer
• Set up the worker node
• Learn to manage the Kubernetes cluster

System Requirements:

Ensure that the networking is set up and that all the virtual machines can talk to each other.

• 7 virtual machines (etcd, LB, worker nodes)
• 3 virtual machines (master nodes)
• 1 client VM

Lab Environment

• Kubernetes cluster managed through minikube.
• The cluster is using a single node EC2 instance in AWS.
• To setup, follow: Using Minikube to Create a Cluster.

Cleanup

The resources can be deleted by simply running a delete command on the manifest directory where the YAML files are located.

kubectl delete -f manifest 

We can also simply delete the EC2 instance in the AWS Management Console.

Resources

• Create Kubernetes Nginx Ingress Controller for External API Traffic

This site is open source. Improve this page.