Conclusion and next steps

Congratulations!

You have completed the Building Pipelines using Tekton on OpenShift workshop.

Coolstuff Store came in with manual deployments, slow release cycles, and no visibility into build status. You’ve left them with a fully automated CI/CD system: reusable Tasks, an end-to-end Pipeline pulling from a Git repository, monitoring tools for diagnosing failures, and Pipelines as Code automatically triggering runs on every push — all without manual cluster intervention.

What you’ve learned

Throughout this workshop, you gained hands-on experience with:

  • How Tekton Tasks encapsulate reusable CI/CD steps and execute as Kubernetes pods, with TaskRuns providing each execution record

  • How multiple steps within a Task run sequentially and share data through a Workspace, with Parameters making the same Task reusable across applications and versions

  • How Pipelines chain Tasks using runAfter without tightly coupling them, with Pipeline Parameters and Workspace mappings flowing values down to each stage

  • How the OpenShift console visual pipeline graph and the tkn CLI give operators and developers instant visibility into pipeline status at every stage

  • How Pipelines as Code stores pipeline definitions in the .tekton/ directory alongside application code, triggers runs automatically on Git events, and gives developers GitOps PR commands to control execution without touching the cluster

  • How to diagnose failed pipeline runs using tkn pipelinerun describe and step-level logs, and how to recover by fixing the underlying cause and re-running

You now have the skills to design, build, run, and maintain CI/CD pipelines on Red Hat OpenShift Pipelines for any application.

Key takeaways

The most important concepts to carry forward:

  1. Tasks are the atomic unit — A Task defines what to do; a TaskRun executes it. Parameters make Tasks reusable; Workspaces let steps share data. Design Tasks to do 1 thing well and they become building blocks for any pipeline.

  2. Pipelines orchestrate without coupling — runAfter defines execution order without creating direct dependencies between Task definitions. The same Task can appear in multiple Pipelines, and changing 1 Task does not require updating all the Pipelines that use it.

  3. The diagnostic pattern is consistent — Whether a TaskRun or a PipelineRun fails, the workflow is the same: list to find the failure, describe to identify which task and why, logs to read the step-level output. This pattern applies to every Tekton resource.

  4. Storage is separate from logic — Workspaces declared on Tasks and Pipelines are bound to actual storage at runtime by TaskRuns and PipelineRuns. The same Task works with a PVC in staging and an emptyDir in a quick CI run. Never hardcode storage into Task definitions.

  5. Pipelines as Code closes the loop — Storing .tekton/ templates in Git makes CI/CD changes reviewable, versionable, and auditable. Webhook-triggered runs and GitOps commands (/retest, /cancel) mean developers interact with CI/CD entirely within their Git workflow.

Next steps

Additional learning paths

Talk to your instructors if you want to explore further topics like:

  • Intermediate: OpenShift GitOps with ArgoCD - basic or advanced workshops. Integrating pipelines with GitOps practices. Pair your CI pipeline with a GitOps-based CD workflow that automatically syncs Kubernetes resources from Git to the cluster. CI builds the artifact; Argo CD deploys it.

  • Advanced: Secure software supply chain with Advanced Developer Suite workshop. Secure your supply chain by adding signature steps and SBOM generation. Automatically sign pipeline artifacts and generate Software Bill of Materials (SBOM) entries for every build. Built into Red Hat OpenShift Pipelines or similar CI engines.

  • Advanced: Platform Engineering for success - with Red Hat Developer Hub. Enable your development teams by building an Internal Developer Portal with Developer Hub and Backstage.

Other recommended topics to explore

Build on what you’ve learned with these related areas:

  • Container image builds with Buildah — Extend your pipeline with a build stage that uses the Buildah ClusterTask (or Tekton Hub buildah Task) to build and push a container image to a registry.

  • OpenShift Pipelines Triggers — Before Pipelines as Code, Tekton Triggers handled webhook-to-PipelineRun conversion. Understanding Triggers helps when reading existing cluster configurations and when PAC is not available.

Practice projects

Put your new skills to work with these hands-on projects:

  1. Add a container image build stage — Extend the coolstuff-build-pipeline from Module 3 with a fourth Task that builds a container image using Buildah and pushes it to the OpenShift internal registry. Use a Workspace to pass the cloned source from clone-source through to the build stage.

  2. Add a deploy stage — Create a deploy-app Task that runs oc rollout restart deployment/coolstuff-app after a successful image build. Chain it as the final stage in the pipeline so a passing build automatically updates the running application.

  3. Create a pull request pipeline — Add a second file to .tekton/, named pr-pipeline.yaml, that triggers only on pull_request events and runs only the validate and test stages (skipping package and deploy). This gives developers fast feedback on pull requests without triggering a full deployment.

  4. Add pipeline notifications — Create a notify-slack or notify-email Task that runs in the finally clause of a Pipeline to report the pipeline result, regardless of whether it succeeded or failed. Tekton’s finally tasks always run even when the pipeline fails.

Certification paths

References

All references used across this workshop, organized by category.

Official Red Hat documentation

Tekton upstream documentation

  • Tekton documentation — Upstream project documentation covering core concepts: Tasks, Pipelines, TaskRuns, PipelineRuns, Workspaces, and Parameters. Used in: Modules 1, 2.

  • OpenShift Pipelines as Code operator documentation — Configuration reference for the OpenShiftPipelinesAsCode custom resource and PAC controller settings. Used in: Module 5.

  • Tekton project home — Overview of the Tekton ecosystem: Pipelines, Triggers, Chains, Dashboard, and Hub. Used in: Module 1.

Pipelines as Code

  • Pipelines as Code project — Official documentation for Pipelines as Code: getting started, Repository CRD, .tekton/ authoring, supported Git providers, and configuration. Used in: Module 5.

  • Pipelines as Code GitOps commands reference — Full reference for /retest, /test, /cancel, and /ok-to-test PR comment commands including syntax, permissions, and advanced options. Used in: Module 5.

Red Hat Developer resources

  • Tekton Hub — Community catalog of reusable Tasks including git-clone, buildah, kaniko, and many more. Referenced in: Workshop details, Module 3.

  • Red Hat Developer: CI/CD — Learning resources, blog posts, and tutorials on CI/CD with Red Hat technologies. Referenced in: Workshop details.

Authors and contributors

Primary Author: Red Hat Sweden, OpenShift SSA-team

Last Updated: 2026-03-20

Workshop Version: 1.0

Contact information:

  • Technical questions: Contact your Red Hat account team

  • Content updates: Submit a pull request to the workshop repository

Help improve this workshop for the next team:

  • What exercise was most useful for your work?

  • Were there any steps that were unclear or took significantly longer than expected?

  • What topics would you like to see added or expanded in a future version?

Submit feedback via your lab portal or contact your Red Hat account team.

Thank you

Thank you for completing this workshop. Coolstuff Store’s team — and yours — now has a solid foundation for building and maintaining CI/CD pipelines on Red Hat OpenShift Pipelines.

Workshop: Building Pipelines using Tekton on OpenShift
Platform: Red Hat OpenShift Pipelines 1.21 on OpenShift 4.20
Platform: Red Hat Showroom