Introducing OpenShift Pipelines

On May 3, 2021, Red Hat released the first public version of OpenShift Pipelines, a cloud-based continuous integration system based on an open source project. Tekton… The solution implements the Kubernetes CI / CD framework for developing and executing pipelines, in which each step runs in its own container, allowing steps to scale independently of each other. Today we will take a quick look at the key features and benefits of this solution, as well as provide a list of additional resources for further acquaintance with it and mastering it.

But before moving on to OpenShift Pipelines, let’s brush up on the core Tekton concepts.

Basic concepts of Kubernetes-native CI / CD

OpenShift Pipelines augments Kubernetes / OpenShift with appropriate CRDs (User Defined Resources) for CI / CD concepts such as pipeline, task, step. As a result, these concepts become “native” instantiable – they can be created as separate instances and, as a result, fully scaled and deployed, as well as ensure their security using Kubernetes tools.

Therefore, to begin with, let’s remember what Tekton concepts are:

Fig.  1. Tekton Concepts
Fig. 1. Tekton Concepts

Basically, Tekton’s core concepts come in two flavors: those that define the pipeline and those that start the pipeline.

Concepts that define a pipeline

  • Task – reusable and loosely coupled series of steps that perform a specific task, such as building a container image.

  • Pipeline – description of the conveyor and tasks (Task)that he must perform.

Run pipelines

  • TaskRun – launch and results of execution of an instance Task.

  • Pipelinerun – launch and results of execution of an instance of the pipeline, which includes a number of concepts TaskRun

You can read more about these concepts in official documentation

Now let’s figure out what OpenShift Pipelines is and what it is for

What’s so special about OpenShift Pipelines?

The OpenShift Container Platform is the industry-leading enterprise-class Kubernetes platform that provides developers with many features, including CI / CD.

OpenShift Pipelines is based on the Tekton open source project and extends the functionality of the OpenShift platform using standard methods, which greatly simplifies the life of developers.

Installing OpenShift Pipelines through the Operator mechanism

OpenShift Pipelines is supported at the operator engine level, so it easy to install and update, and, accordingly, is easy to administer.

OpenShift Pipelines is available on the OperatorHub site, which has over 450 different operators for the OpenShift Container Platform:

Installing OpenShift Pipelines is extremely simple, and it is immediately installed as a cluster-level operator, automatically becoming available to all projects:

OpenShift Pipelines also complements OpenShift with corresponding CRs, which will allow for even more versatility in terms of configuration and integration with the OpenShift console, etc.

When a new version of OpenShift Pipelines appears in OperatorHub, you, as an administrator, can choose to update to the next version by specifying the desired update channel.

Developed UI within the OpenShift console

Tekton also supplements the standard OpenShift delivery with CI / CD concepts, but in it, when creating and running pipelines, it is difficult to do without creating YAML code, and this code requires writing a lot, thousands of lines. Therefore, Red Hat has implemented a full-fledged UI in the OpenShift console for launching and rendering pipelines (both those that are working now and those that have already worked), as well as for graphically creating pipelines. At the same time, all the necessary YAML files are created automatically, without writing any code.

Below is a graphical representation of the pipeline, which has already been executed on the OpenShift platform, and directly from it you can access the logs and events of all tasks of this pipeline:

Fig.  2. Pipelines in the OpenShift console
Fig. 2. Pipelines in the OpenShift console

If desired, you can easily view the complete log of the selected task:

To make life even easier for developers, OpenShift Pipelines lets you draw pipelines right in the OpenShift console, so you no longer need a YAML black belt to create your first Tekton pipeline:

Fig.  3. Graphic design of the conveyor in the OpenShift console
Fig. 3. Graphic design of the conveyor in the OpenShift console

But if you, as a YAML black belt holder, want to tweak something, you can always do it directly from the OpenShift console:

Fig.  4. YAML - examples and snippets in the OpenShift console
Fig. 4. YAML – examples and snippets in the OpenShift console

Moreover, OpenShift Pipelines will come in handy even if you decide to go down the pure YAML path, and will provide you with ready-made examples and code snippets for faster development of YAML pipelines. In addition, you can integrate your corporate snippets into the system, making them available to all developers. That’s why we added a special CRD called ConsoleYAMLSamples.

Events as Triggers for Running Pipelines

Do you want to associate the launch of pipelines with some external events (in Tekton terms, this is called Trigger), such as push or pull requests from Github or Gitlab? Not a problem at all, OpenShift Pipelines has it out of the box and supports various vendors including Github, Gitlab, BitBucket, etc.

Fig.  5. Adding a trigger in the OpenShift console
Fig. 5. Adding a trigger in the OpenShift console

You just create the desired trigger in the UI, and OpenShift itself generates all the necessary concepts, such as EventListeners, TriggerTemplates (you can read more about them in official documentation).

Ready-to-use reusable tasks and customization

OpenShift Pipelines out of the box contains dozens of ready-made tasks that can be used as part of pipelines, including tasks for cloning code from a repository, building applications in various programming languages ​​such as java, dotnet core, python go, nodejs, or using build tools like maven, deployment applications, etc. The list of available tasks can be seen in the OpenShift console, ClusterTasks section, Pipelines -> Tasks menu:

Fig.  6. OpenShift Pipelines offers dozens of ready-made tasks out of the box
Fig. 6. OpenShift Pipelines offers dozens of ready-made tasks out of the box

In addition, this list can be easily expanded. To do this, just add “ClusterTasks” to the cluster, after which hundreds of additional tasks on the site will become available to you TektonHubwhich is a public repository for sharing Tekton tasks:

Fig.  7. TektonHub - public repository of reusable Tekton tasks and pipelines
Fig. 7. TektonHub – public repository of reusable Tekton tasks and pipelines

Integration with IDE

Developers using the command line and IDE can take advantage of the Tekton CLI, the Tekton extension for Visual Studio Code, and the Tekton plugin for IntelliJ to interact with pipelines right from their normal development environment and create, run, view and perform actions on the cluster directly from the command strings.

Fig.  8. VSCode Extension for OpenShift Pipelines
Fig. 8. VSCode Extension for OpenShift Pipelines

Useful Resources

In conclusion, we advise you to watch the video version of this article in English:

We also recommend the following resources (EN):

Videos in Russian:


Similar Posts

Leave a Reply

Your email address will not be published. Required fields are marked *