CI | CD Pipelines

Overview

In modern software projects many teams utilize the concept of continuous integration and continuous delivery (CI/CD). By setting up a tool chain that continuously builds, tests, and stages software releases a team can ensure that their product can be reliably released at any time. OpenShift can be an enabler in the creation and managecment of this tool chain. In this lab we will walk through creating a simple example of a CI/CD pipeline utlizing Jenkins, all running on top of OpenShift!

Start by creating a new project

To begin, we will create a new project. Name the new project “cicd”.

        $ oc new-project cicd
        

Start by installing Jenkins

First we will start by installing Jenkins to run in a pod within your workshop project. Because this is just a workshop we use the ephemeral template to create our Jenkins sever (for a enterprise system you would probably want to use the persistent template). Follow the steps below:

        $ oc new-app --template=jenkins-ephemeral -p JENKINS_PASSWORD=password
        $ oc expose svc jenkins
        $ oc policy add-role-to-user edit -z default
        

        $ oc get routes | grep 'jenkins' | awk '{print $2}'
        

The OpenShift pipeline plugin

Now let’s make sure we have the OpenShift Pipeline plugin properly installed within Jenkins. It will be used to define our application lifecycle and to let our Jenkins jobs perform commands on our OpenShift cluster. It is possible that the plugin is already installed in your environment, so use these steps to verify if it is installed and install it if is not.

You can read more about the plugin here.

Our deployments

In this example pipeline we will be building, tagging, staging and scaling a Node.js webapp. We wrote all the code for you already, so don’t worry you won’t be coding in this lab. You will just use the code and unit tests to see how CI/CD pipelines work. And keep in mind that these principles are relevant whether your programming in Node.js, Ruby on Rails, Java, PHP or any one of today’s popular programming languages.

Fork the project into your own GitHub account

Create a dev deployment based on the forked repo

Goto the terminal and type the following:

$ oc new-app https://github.com/YOUR_ACCOUNT/openshift-workshops.git \
   --name=dev --context-dir=dc-metro-map
$ oc expose svc/dev

Create a test deployment based on a tag of the dev ImageStream

Goto the terminal and type the following:

$ oc new-app dev:readyToTest --name=test --allow-missing-imagestream-tags
$ oc expose dc/test --port 8080
$ oc expose svc/test

Setup Jekins jobs to use their openshift image stream (which is off your GitHub fork)

</div>

Connecting the pipeline for dev->test

echo "inside my jenkins job"

Watch me release!

At this point you should see the following scenario play out:

• Inside of Jenkins, you will click the dev pipline that was created we created. On the left-hand side you will see an option to . When you click this, the first job will begin to run.

• This first job will use the OpenShift Pipeline plugin to create a new tag of the image called “readyToTest”.

• When this job completes, a second job will execute. This second job cause the deployment to initiate of our test application and then scale the test application to 2 pods.

• You can see the history of this new tag by browsing to initiate two jobs in the pipeline with the final step being the new tag of “readyToTest”. The new tag can then be used for automatic or manual builds of the new test application. You can view the status of the new tag in OpenShift by browsing to Builds -> Images -> your image stream

Summary

Coming soon… Read more about usage of Jenkins on OpenShift here. Read more about the concepts behind pipelines in Jenkins here.