CapstoneDevOpsProject

This project deploys a simple node server into a Kubernetes cluster using the AWS infrastructure. It also uses a CI/CD pipeline to enhance development and continous delivery of features.

Architecture

The underlying architecture of the project is represented by the following diagram:

Installation

To setup the project you need to first deploy the underlying infrastructure required to support the application. This is done by executing the following script:

Windows

$ ./cloudformation/create_environment.bat

Linux

$ ./cloudformation/create_environment.sh

Once the infrastructure is deployed, you need to install kubectl and deploy the application using the following commands:

$ cd capstone_app
# configure kubectl
$ aws eks update-kubeconfig --name capstone-cluster

# deploy the service
$ kubectl create -f ./deployment.yml
$ kubectl create -f ./service.yml

# check service status and url
$ kubectl describe svc node

Important

Be sure that you use the same credentials to deploy the infrastructure and in the Jenkins credentials configurations. If they are not the same, you will have problems deploying the EKS Cluster with kubectl.

Cleanup

To make cleanup of the application and infrastructure you need to run first:

$ cd capstone_app
$ kubectl delete -f ./deployment.yml
$ kubectl delete svc node

Then you need to delete the infrastructure using the following command:

Windows

$ ./cloudformation/delete_environment.bat

Linux

$ ./cloudformation/delete_environment.sh

Jenkins

First, you need to login with the Jenkins default password. Then go to settings > manage plugins and install the following plugins:

• Blue Ocean
• AWS Steps
• Docker Pipelines

Then you need to setup AWS Credentials and DockerHub credentials. To trigger the workflow with pushes to the branch you should configure a webhook.

The method used in this proyect for deployment is rollback.

Evidence

The screenshots and urls that support this project are located at the evidence folder.