Udacity Cloud DevOps Engineer Nanodegree Program

Capstone Project

Project Overview

In this project, I applied the skills and knowledge which were developed throughout the Cloud DevOps Nanodegree program. These include:

• Working with AWS
• Using Jenkins to implement Continuous Integration and Continuous Deployment
• Building Jenkins pipelines
• Building Docker containers in the pipelines
• Working with eksctl and kubectl to build and deploy Kubernetes clusters

As a capstone project, the directions are rather more open-ended than they were in the previous projects in the program. I developed a CI/CD pipeline for micro services applications with rolling deployment.

Project Steps

• Create an AWS EC2 instance for Jenkins server and another AWS EC2 instance for a Jenkins agent

  • The Jenkins server itself is the "master" agent for build pipelines, so having another AWS EC2 instance as a Jenkins agent is optional.
  • After installing the Jenkins server, install necessary plugins such as Blue Ocean, Pipeline: AWS Steps, and Docker Pipeline.
  • Add credentials of Jenkins agent(s)' SSH private key(s), AWS, and dockerhub.
  • Do not forget to associate Elastic IP addresses to both two instances.
  • Add a new Jenkins agent
  • The Linux user jenkins needs to be added to the group docker.
    • sudo usermod -a -G docker jenkins
  • Install docker, hadolint, aws-cli, eksctl, kubectl, and other necessary system-level utilities in the Jenkins build agent(s).

• Use Jenkinsfile to build a Declarative Pipeline

  • Validate Jenkinfiles in VSCode
  • Jenkins Pipelines: What I Wish I Knew Starting Out
    • difference between single and double quoted strings
  • Stages
    • verify the build system
    • verify aws-cli v2, eksctl, kubectl
    • lint
    • build docker image
    • test the docker container
    • push docker image
    • deploy to AWS EKS
    • verify deployment
    • check rollout