Building a CI/CD pipeline, GitOps pt3

Learning new tech skills and software takes time and consistency. This post covers the process I followed to learn and implement a new technology.

Task: Install Jenkins via Helm and build a pipeline that builds and pushes a Docker image from a Gitea repository.

Step 1: Planning

• Identify the steps and resources needed to deploy Jenkins
• Determine the app to deploy
• Determine the values needed for Helm
• Research how to connect Jenkins to Gitea
• Plan the Jenkins pipeline
• Determine if additional Jenkins tutorials or training are needed
• Plan for accessing Jenkins (VPN, Ingress, etc.)

Step 2: Code

• Learn about Jenkins through watching YouTube content and taking Jenkins KodeKloud course (affiliate link)
• Develop a backend API using Python Github link
• Reference Jenkins’ GitHub documentation to install Helm
• Connect Jenkins to Gitea using the Gitea Jenkins plugin and an access token
• Write the pipeline code in a Jenkinsfile using either a Declarative or Scripting syntax

Below is a screenshot of Jenkins deployed via ArgoCD.

Step 3: Build

With the prereqs completed it was time to:

• Validate that Jenkins can see the Gitea repository
• Build the backend API Docker image and push it to Docker Hub using the following pipeline code snippet:

# Jenkinsfile
podTemplate(
    containers: [containerTemplate(name: 'docker', image: 'docker', command: 'sleep 1000')],
    volumes: [hostPathVolume(hostPath: '/var/run/docker.sock', mountPath: '/var/run/docker.sock')]){
    node(POD_LABEL) {
        def api 

        container('docker'){
            stage('Clone-repo-build-docker-image') {
                /* ensure repo is cloned  */          
                checkout scm

                /* This builds the actual image; synonymous to
                * docker build on the command line */
                dir("coffee_shop_api"){
                    api = docker.build("myDockerHub/coffee_shop_api")
                    }
                }

            stage('Testing'){
                    try { 
                sh 'pip install -r requirements.txt'
                sh 'pytest'
                    }
                    catch (exc) {
                echo 'Pytest tests failed'
                }                   
             }
            
            stage('Post to DockerHub') {

                docker.withRegistry('', 'dockerhub') {
                    api.push()
                    }
            }
        }
    }        
}

The following is the outcome of the successful Docker build, and the Docker image is now available for download from Dockerhub.

These steps took place over the course of a week, working on each task daily for 60-90 minutes and focusing on completing one task before moving on to the next.

Consistency and enjoyment are key to sustained learning and success.S

In the next post, I will further discuss the CI/CD process.