basic-helm-spring-boot

A Sample OpenShift Pipeline for a Spring Boot Application Using Helm

This Jenkins pipeline provides an example of how to build a basic Spring Boot application using Helm. The pipeline runs using the CoP's jenkins-agent-helm agent and contains the following steps:

1. Checkout: Checks out the spring-rest application
2. Get Version From POM: Gets the version defined in the pom.xml file. This version is used to set your image's tag.
3. S2I Build: Performs an s2i build. This stage runs your unit tests, builds your jar file, and builds your container image. You could split this stage like the basic-spring-boot example if desired, but the jenkins-agent-helm agent doesn't have maven on it, so in this case, it's easier to perform an s2i build instead.
4. Deploy: Deploys your application to OpenShift.

This pipeline uses two different Helm charts called spring-boot-build and spring-boot. The spring-boot-build chart is used in the S2I Build stage to create and update your BuildConfig and ImageStream. The spring-boot chart is used in the Deploy stage to create and update your application's Kubernetes resources. While you could combine both of these charts into one, splitting into separate charts for building and deploying provides a greater separation of concerns.

Before you configure this pipeline, you must first build a Jenkins agent containing Helm. Let's look at how you can build the jenkins-agent-helm agent.

Building the jenkins-agent-helm Agent

While the pipeline itself only uses the spring-boot and spring-boot-build Helm charts, a third Helm chart is provided under the .helm/ folder called jenkins-agent-helm that you can use to easily build this Jenkins agent. First, install the jenkins-agent-helm Helm chart, which creates a BuildConfig and ImageStream:

helm install jenkins-agent-helm .helm/jenkins-agent-helm

The first build will automatically run. You can follow this build by running:

oc logs bc/jenkins-agent-helm -f

This ImageStream created by Helm already has the role: jenkins-slave label, so Jenkins will see this and automatically use this image to configure a new Jenkins agent.

Next, let's configure the Jenkins pipeline.

Configuring This Jenkins Pipeline

Because the JenkinsPipeline build strategy is deprecated in OpenShift 4.x, this doc will describe how you can configure this pipeline manually within the Jenkins console.

First, deploy a Jenkins instance to OpenShift. You can deploy Jenkins by running this command:

oc new-app jenkins-persistent

Once Jenkins is up, find and access your route URL:

oc get route jenkins -o jsonpath='{.spec.host}'

Then, follow these steps:

1. Click New Item on the lefthand side of the screen
2. Enter a name for this pipeline. The suggested name is basic-helm-spring-boot.
3. Select Pipeline and click Ok
4. Scroll down to the bottom of the screen to the Pipeline section. Change the "Pipeline script" dropdown to Pipeline script from SCM.
5. For "SCM", select Git and enter the Repository URL https://github.com/redhat-cop/container-pipelines.git
6. For "Script Path", enter basic-helm-spring-boot/Jenkinsfile
7. Click Save

Once you configure your pipeline, you can click Build Now on the lefthand side of your screen to start the pipeline.

Validating Your Deployment

You can validate your deployment was successful with the following commands:

ROUTE=$(oc get route my-app -o jsonpath='{.spec.host}')
curl $ROUTE/v1/greeting