Skip to main content
In this article, we’ll explore the key DevOps prerequisites for a software provider and demonstrate how GitLab CI/CD can address these requirements with a streamlined, scalable workflow. Dasher Technology offers a platform that connects data, applications, and devices across on-premises environments. Their R&D team is investigating a cloud migration and container strategy—starting with a Node.js project and later extending to Java and Python applications. Alice leads the new DevOps team tasked with building the CI/CD pipeline from the ground up. Their multi-cloud architecture will use Docker for containerization and Kubernetes for orchestration. Alice’s initial assessment uncovered several pain points:
  • No version control or branch management
  • Manual code integration and testing
  • Slow, unreliable test runs with low coverage
  • Infrequent merges that elevate release risk
  • Manual deployments to development, staging, and production
To overcome these challenges, the team will implement a CI/CD pipeline consisting of:
  1. Version control and collaboration on GitLab
  2. Automated unit tests with code coverage
  3. Docker image builds and registry pushes
  4. Deployment to Kubernetes clusters
  5. Automated integration tests before production rollout
The image outlines the DevOps requirements for a team, featuring tasks like code integration, collaboration, testing, and deployment, with icons representing each step. It includes a section for automated IT processes and is copyrighted by KodeKloud.

​
Evaluating CI/CD Tools

The team evaluated several popular CI/CD platforms: Alice initially chose Jenkins due to its maturity and active community. However, setting up and maintaining a Jenkins server for a single Node.js project proved time-consuming:
  • Provision virtual machines with sufficient CPU, memory, and disk
  • Install Java JDK, Jenkins, and required plugins
  • Configure firewall rules and security groups
  • Install and manage multiple Node.js versions and npm
  • Install Docker for container builds
  • Add Kubernetes tooling: kubectl, Helm
  • Integrate external testing/reporting CLIs
Extending this setup for Java or Python on AWS/Azure adds layers of complexity:
  • Java: Maven or Gradle
  • Python: virtualenv, pip
  • Cloud CLIs: AWS CLI, Azure CLI
  • DevSecOps: vulnerability scanners like Trivy and KubeSec
The image lists traditional CI/CD tools and challenges, featuring icons for Java, Maven, Python, Azure, AWS CLI, Trivy, and Kubesec.
Managing your own CI/CD infrastructure means spending more time on setup and maintenance rather than on writing pipelines.
Because many team members are new to this tool ecosystem, Alice needed a solution that would:
  • Require minimal infrastructure setup
  • Let the team focus on pipeline authoring
  • Provide built-in scalability and security
The image outlines challenges of traditional CI/CD tools, highlighting the need for simple setup without extensive installations and focusing on pipeline development without infrastructure concerns. It includes a central icon labeled "Alice."
After evaluating multiple platforms, Alice selected GitLab CI/CD. In the following sections, we’ll build and optimize a GitLab CI/CD pipeline for a real-world Node.js application—covering:
  • Version control and merge request workflows
  • Automated testing and code coverage
  • Docker image creation and registry hosting
  • Kubernetes deployments and rollbacks
  • Integration testing and monitoring
Let’s get started!

Watch Video