Building a CD process using the OpenShift Pipelines

(which generates Tekton resources behind the scenes)

Here I'll show the process of building a CD pipeline using the GUI of the OpenShift Pipelines. This pipeline will have the same tasks as the one we build previously with Tekton, when we were writing the YAML files manually.

The sample application for deployment is written in Python, the only task that will not be taken from the OpenShift Pipelines' catalog and written manually instead is the task named "nose" for unit-testing the code.

1. Preparation

To install the "nose" task, just copy the following code block for the task, save it into a file named tasks.yaml

---
apiVersion: tekton.dev/v1beta1
kind: Task
metadata:
  name: nose
spec:
  workspaces:
    - name: source
  params:
    - name: args
      description: Arguments to pass to nose
      type: string
      default: "-v"
  steps:
    - name: nosetests
      image: python:3.9-slim
      workingDir: $(workspaces.source.path)
      script: |
        #!/bin/bash
        set -e
        python -m pip install --upgrade pip wheel
        pip install -r requirements.txt
        nosetests $(params.args)

and apply the task:

kubectl apply -f tasks.yaml

If you'd like to see that it was succesfully installed, run this:

oc get tasks

2. Creating PersistentVolumeClaim

The PersistentVolumeClaim (PVC) will be use as a workspace. Use the OpenShift Administrator section to create the PVC.

In the main menu:

switch from Developer to Administrator
go to Storage -> PersistentVolumeClaims

Now, in the PersistentVolumeClaims page:

press the Create PersistentVolumeClaim button
choose the appropriate StorageClass from the dropdown
(skills-network-learner in my case)
in the PersistentVolumeClaim name type ccd-openshift-pipeline-pvc
for Access mode choose Single user (RWO) radio button
Size, set to 1 GiB
for Volume mode choose Filesystem radio button
press the Create button

3. Creating a new Pipeline

In the main menu:

switch from Administrator back to Developer
go to Pipelines

Now, in the Pipelines page:

press the Create button
(make sure that you have Pipeline and not Repository selected under the button)
((You are presented with the pipeline builder. Ensure you have Pipeline Builder selected in Configure via))
In the Name field enter cd-openshift-pipeline
Scroll towards the bottom of the page to the Parameters section
Add the name parameter with its default value by entering the following:
```
parameter.name: app-name
parameter.default: cicd-app.
```
Add the build-image parameter: replacing CONTAINER_REGISTRY_NAMESPACE with the value appropriate for your setup
((to look up my value: echo $SN_ICR_NAMESPACE))
```
parameter.name: build-image
parameter.default: image-registry.openshift-image-registry.svc:5000/CONTAINER_REGISTRY_NAMESPACE/tekton-lab:latest
```
scroll farther towards the bottom of the page to the Workspaces section
add a new workspace with the name: output (it will be used to upload the code to and work on it).
Press the Create button to save your work

4. Adding tasks to the Pipeline

Now when the groundwork on building the pipeline is done, let's add the tasks to it.

Open the pipeline in edit mode (select Pipeline from the main menu, select the pipeline name, and then go to Actions -> Edit Pipeline)

1 - Adding the cleanup task

Click Add Task in the builder UI to open the Add task ... dialog.
type cleanup to see the cleanup task by Red Hat.
press the Add button to start using the task

2 - Adding the git-clone task

Hover over the step to display the + buttons; use the + button on the right of the task to add a task
type RedHat git-clone and add the task by pressing the Add button
(the red exclaimation mark on the task means that the task should be configured)
to configure the task, press on the exclaimation mark

then fill the following fields in the open flyout:

url: https://github.com/ArthurPro123/wtecc-CICD_PracticeCode
workspace.output: output

3 - Adding the Flask8 task (to lint the source code)

use the + button on the right of the git-clone task to add this one
type Flask8 (from the community) and add the task by pressing the Add button
press on the exclaimation mark

fill the following fields to configure the Flake8 task:

image: python:3.9-slim
arg: --count
arg: --max-complexity=10
arg: --max-line-length=127
workspace.source: output

4 - Adding the nose task (for unit testing the source code)

use the + button on the BOTTOM of the Flask8 task
type nose and add the task by pressing the Add button
press on the exclaimation mark
fill the following field to configure this custom task:
```
workspace.source: output
```

5 - Adding the buildah task

use the + button on the left of the Flask8 task to add this one
type RedHat buildah (from the RedHat) and add the task by pressing the Add button
press on the exclaimation mark
fill the following fields to configure the buildah task:
```
image: $(params.build-image)
workspace.source: output
```

6 - Adding a task to deploy the image to the OpenShift cluster

(with the help of the openshift-client task to execute an inline shell script)

use the + button on the right of the git-clone task to add this one
type openshift-client (from the RedHat) and add the task by pressing the Add button
press on the exclaimation mark

fill the following fields to configure the openshift-client task:

display name: deploy
SCRIPT: oc create deployment $(params.app-name) --image=$(params.build-image) --dry-run=client -o yaml | oc apply -f -

(Save the pipeline.)

Running the pipeline

To run the pipeline manually, you should be on the pipelines page (if you're not, press the Pipelines button in the main menu. There, press on cd-openshift-pipeline pipeline, then click the Actions dropdown, and the button in it.

This will display the Start Pipeline modal window.

Make sure that the following options are chosen:

output: PersistentVolumeClaim
PVC: cd-openshift-pipeline-pvc

And finally, press the button.

Check the logs to see if there are issues with the pipeline.

((You should see the pipeline running on the next page. You can click on the task name to see the logs for a particular task. Alternatively, you can click on the Logs tab:))

Confirming the deployment of the application

Click on Topology in the main menu (available in in the Developer section). There should see the applications which are currently running.

Click on the one called cicd-app??? and click on View logs(Logs)??? in the flyout.

If everything worked well, there should be a message SERVICERUNNING displayed, this message was generated by the Python code deployed (for demonstation purposes) and has nothing to to with OpenShift.

Building a CD process using the OpenShift Pipelines.

This page was last updated on February 01, 2025