NOTE: There are two branches for this demo

• master branch uses Docker Hub for the registry
• ocir branch uses Oracle Compute Infrastructure Registry

Please select branch accordingly as the wercker.yml file differs for different registries.

This repository contains the code for the WorkBetter application demo. This demo will walk you through the process of developing an application in a containerised CI/CD pipeline and deploying it to a Kubernetes cluster running on Oracle Cloud Infrastructure.

You will take on the role of a developer, responsible for configuring the automated build and deploy process that involve details about the application itself. To containerise and automate the building and deploying of this application you will make use of Wercker Pipelines, Oracle Container Registry, and Oracle Cloud Infrastrastructure. The Kubernetes infrastructure will be configured and provisioned through Terraform Installer.

On every pipeline execution, the code goes through the following steps:

1. Code is pulled from GitHub, built, tested and deployed to Kubernetes
2. The Docker image is pushed to the internal Oracle Container Registry
3. The application is then tested
4. If passed, the image will be pulled from the registy and deployed to the Kubernetes cluster

Approximately 15 minutes

You are an application developer who is developing a container native HR application using JET for building the web application to be used in a browser from a mobile phone up to desptop. This web application is designed as a microservice to run in a Docker container and deployed to Kubenetes.

The reason why you want to adopt a container native approach has partially been driven by the need to go to market quicker, by delivering new features more frequently, but also more reliably. And you are able to do this by developing Microservices that has less dependencies on other services, as well as the footprint being smaller, easier to deploy and starts up faster. What’s also attractive with Microservices is that you can use the best programming language for the job. To be able to deliver this new style of container native application approach, you will need to adopt the Agile development practice to continuously integrate and deliver these services and features. The application basically consists of the user interface that is built using Oracle JET framework, packaged as a Docker image and deployed to a Kubernetes cluster hosted on Oracle Cloud Infrastructure.

You will use Wercker (Oracle Container Pipeline) as the core CI/CD software lifecycle management tool for your Agile development and DevOps pipeline. The WorkBetter application has already been created in Wercker with the Workflow and Pipelines created. This demo assumes you have a GitbHub and Wercker accounts and Brackets installed locally on your laptop. You must also have a provisioned Kubernetes cluster on Oracle Cloud Infrastructure.

• GitHub account
• Wercker account
• Oracle Cloud Infrastructure
• A Kubernetes cluster running on OCI
• An installation of Brackets with Bracket Git by Martin Zagora extension installed
• Some familiarity with GitHub
• Some familiarity with Wercker
• Some familiarity with Brackets

This demo requires the Oracle Container Engine for Managed Kubernetes Service (OKE) as the runtime platform.

• If you do not already have an OKE cluster, you must provision one before you can run the demo.
• Please follow the instruction here to provision your OKE cluster OKE Setup Link.

This demo also requires Wercker to orchestrate the build and deployment of the Work Better application to an OKE cluster. This involves configure Wercker pipelines and workflow.

• If you do not already have Wercker setup for Work Better, you must create and configure one before you can run the demo.
• Please follow the instruction here to create your Work Better application in Wercker Wercker Setup Link.

Work Better is a HR web application written in JET. This web application is designed as a microservice to run in a Docker container and deployed to Kubenetes. It is continuously integrated and deployed to a Kubernetes cluster running on OCI.

• Go to the Work Better dashboard

• Click on some of the items

  

Wercker is the next generation container lifecycle management tool. You can create Pipelines that enable full build, test and deployment workflows to execute. It supports Docker-native continuous integration and delivery of microservices and functions with a full set of collaboration tools. And Releases are deployed and managed through Kubernetes and built-in Container Registry Service. So that container images are stored and shared across multiple deployments on Oracle Cloud with our private registry. You can also operate Kubernetes deployments to manage and monitor production applications; scale up and scale down your clusters - secure and load balanced.

Also explain about the GitHub repository was added to the workflow when the application is created and configured in Wercker.

• Go to the Wercker WorkBetter application

• Describes the navigation tab along the top

• Click on the Runs tab

  

A Workflow in Wercker is how the automation of pipelines are managed. You can interconnect multiple pipelines together to form a workflow. A pipeline can be triggered by a GitHub push or by the previous pipeline.

• Click on the Workflow tab

• Explain each pipleine in the workflow:

  • The build pipeline is created by default and it is triggered whenever a commit is made to the GitHub repo or by a GitHub push.
  • The functional-test pipeline is triggered by a successful build.
  • The deploy pipeline is triggered by a successful functional test with a condition for changed made to the master branch only. This enables the testing of code in a new branch before being merged into the master for release.

  

You can define your application environment variables externally under the Environment tab. Some of the variables are common standard variables defined by Dockers or Kubernetes, such as the Kubernetes master node address for application deployment, and its token. Similarly, the Docker user and password can be defined here so that the image can be pulled from the registry for deployment.

• Click on the Environment tab

• Describe the variables

• Highlight the token or key are protected and cannot be displayed or copied for security reason

  

You can monitor and manage your Kurbernetes cluster through its dashboard. Not everything can be managed from this console as the Kubernetes cluster is provisioned on Oracle Cloud Infrastructure. For more details please open the OCI console where you will find the master, worker, etcd and load balancer nodes.

• Open the Kubernetes dashboard

• Select Namespace and corralate it with the NS environment variable

• Select Overview show all

• Select Service

• Highlight the External endpoints, this is the IP address to access the WorkBeter application

  

We want to demonstrate how we can trigger a new build by making a code change. The profile picture for Steven King is presented in a square frame. We can change this to be presented in a circular frame.

We can change this by changing a line of code we previously commented it out in the dashboard.html under WorkBetter/website/js/views in our GitHub repo.

• Go to the GitHub WorkBetter repo

• Open the dashboard.html file

• Locate lines 44 to 47 where the dashboard photo is defined

  

• All we got to do is swap the comments tags around like so

  

There are a number of ways to work with your code. Most developers will be working with an IDE or some OpenSource editors. Or you could make the change directly in the GitHub online editor.

Using the GitHub online editor is straight forward, however, this is not the preferred way of developing an application. When a new branch is created, it will trigger the build and functional-test pipeline, but without excuting the deploy pipeline.

• Instead of committing the code to the master branch, we will create a new branch

• Check the Create a new branch for the commit and start a pull request at the bottom of the file

• Enter v1.1 in the new branch field

• Click on Propose file change

  

• You have now made a change to the file in a new branch

Since you've made code changes in the GitHub online editor, you can now go straight to Step 9

As I have the Brackets editor installed and synchronised with my GitHub repo, I can use Brackets to illustrate how I create a new branch for the code change and push it back to the repo to trigger the build and functional-test pipeline, but without excuting the deploy pipeline.

• Open Brackets

• Create a new branch v1.1

  

• Open the dashboard.html file

• Locate lines 44 to 47 where the dashboard photo is defined

• Swap the comments tags around

• Save the file

  

We have now made the code change in the new v1.1 branch and need to push the changes back to GitHub.

• Click on Commit button to commit the changes

• Enter a commit message

• Click on Git Push button synchronise

  

The moment we commit the Git push, the build pipeline would be triggered in Wercker. Please take note of the pipelines to be executed. The build and functional-test should be the only pipelines to be executed as the deploy pipeline is executed only on changes to the master branch. Since we have made the changes in a new branch, we can therefore test it without deploying it to production.

• Switch back to Wercker

• Click on the Runs tab

• Observe the workflow

• Click into the pipelines and expand the steps for more detail

  

If the previous pipelines completed sucessfully then we can commit the changes into the master branch by merging the v1.1 to the master. This will automatically trigger a deployment by calling the deploy pipeline. Instead of using Brackets to merge the branches, lets use the GitHub editor instead as you can see the steps better visually.

• Switch to GitHub editor

• Create pull request

• Click on Merge pull request

  

Once we commit the merging of the master and v1.1, the workflow will be triggered. Please take note of the pipelines to be executed. This time we will execute the deploy pipeline as changes have been merged with the master branch.

• Switch back to Wercker

• Click on the Runs tab

• Observe the workflow

• Click into the pipelines and expand the deploy for more detail

  

You can now see the changes by reloading your browser page. The key message is how agile it is to continuously integrate and deploy new features to the end users. And we shown the support of CI/CD with Wercker (Oracle Container Pipeline). With Wercker, the deployment can be to any Kurbenetes cluster but it will be even better with Oracle Kubernetes Engine which is a managed Kubernetes environment.

• Reload WorkBetter dashboard in your browser

• Observse the profile photo changed to a circular frame