Chapter 1. Red Hat OpenShift Pipelines release notes

With this release, you can configure resolution timeout settings for pipeline resolvers to gain greater flexibility and control when running a pipeline.

apiVersion: operator.tekton.dev/v1alpha1
kind: TektonConfig
metadata:
  name: config
spec:
# ...
  pipeline:
    options:
      configMaps:
        config-defaults:
          data:
            default-maximum-resolution-timeout: 5m 1
        bundleresolver-config:
          data:
            fetch-timeout: 1m 2
# ...

With this release, the OpenShift Pipelines now support community tasks.

The following community tasks are installed by default in the openshift-pipelines namespace:

With this release, OpenShift Pipelines supports the deployment of new containers through the TektonConfig CR.

apiVersion: operator.tekton.dev/v1alpha1
kind: TektonConfig
metadata:
  name: config
# ...
spec:
  result:
    options:
      deployments:
        tekton-results-watcher:
          spec:
            template:
              spec:
                containers:
                - name: kube-rbac-proxy
                  args:
                    - --secure-listen-address=0.0.0.0:8443
                    - --upstream=http://127.0.0.1:9090/
                    - --logtostderr=true
                  image: registry.redhat.io/openshift4/ose-kube-rbac-proxy:v4.12
# ...

With this release, the EventListener object for OpenShift Pipelines triggers now includes the ImagePullSecrets field to specify a secret that the listener uses to pull images from private registries.

apiVersion: triggers.tekton.dev/v1beta1
kind: EventListener
metadata:
  name: imagepullsecrets-example
# ...
spec:
  serviceAccountName: triggers-example
  resources:
    kubernetesResource:
      spec:
        template:
          spec:
            imagePullSecrets:
              - name: docker-login
# ...

With this release, the opc command line utility is now shipped with the following components:

With this release, a PipelineRun resource can be triggered based on which files have changed by using the on-path-change and on-path-change-ignore annotations, which simplifies the process by not requiring writing complex CEL expressions.

apiVersion: tekton.dev/v1beta1
kind: PipelineRun
metadata:
  name: "on-path-change"
  annotations:
    pipelinesascode.tekton.dev/on-path-change: ["docs/**"] 1
    pipelinesascode.tekton.dev/on-path-change-ignore: [".github/**"] 2
spec:
# ...

With this release, a PipelineRun resource can now be triggered by pushed commits on GitHub by using the on-comment annotation. This feature gives you more control over when a PipelineRun is triggered, enabling you to match it based on specific comments.

You must configure a pipeline run to enable the triggering with the on-comment annotation:

apiVersion: tekton.dev/v1beta1
kind: PipelineRun
metadata:
  name: "on-comment-pr"
  annotations:
    pipelinesascode.tekton.dev/on-comment: "^/hello-world" 1
spec:
# ...

With this release, you can use the tkn pac info globbing [-s | -d] “<pattern>” command to test patterns with sample input data or in the working directory where the command is run. The command tests if the patterns are working properly in the PipelineRun definition.

For example, you can test the following on-target-branch annotation to ensure that it matches the main branch in the git event payload:

apiVersion: tekton.dev/v1beta1
kind: PipelineRun
metadata:
  name: "on-target-branch"
  annotations:
    pipelinesascode.tekton.dev/on-target-branch: "[refs/heads/*]"
spec:
# ...

$ tkn pac info globbing -s "refs/heads/main" "refs/heads/*"

With this release, Pipelines as Code supports commas within annotations by using the , HTML entity. Now, you can use comma-separated values together with commas used directly in the values.

For example, to match two branches called main and branchWith,comma, specify the annotation as follows:

apiVersion: tekton.dev/v1beta1
kind: PipelineRun
metadata:
  name: "comma-annotation"
  annotations:
    pipelinesascode.tekton.dev/on-target-branch: "[main, branchWith,comma]"
spec:
# ...

With this release, after a pull request is closed or merged, all associated PipelineRun resources that have the cancel-in-progress: true annotation are automatically canceled. To enable this feature, specify the cancel-in-progress annotation in your PipelineRun definition:

apiVersion: tekton.dev/v1beta1
kind: PipelineRun
metadata:
  name: "cancel-in-progress"
  annotations:
    pipelinesascode.tekton.dev/cancel-in-progress: "true"
spec:
# ...

With this release, an active PipelineRun resource that has the cancel-in-progress: true annotation is automatically canceled when a new PipelineRun resource with the same name is triggered by Pipelines as Code. Now, after a pull request is raised and triggers a PipelineRun, and subsequent commits to the same pull request trigger a new PipelineRun, the older obsolete PipelineRun is canceled to save resources.

To enable the feature, set the pipelinesascode.tekton.dev/cancel-in-progress annotation to true in the PipelineRun resource:

apiVersion: tekton.dev/v1beta1
kind: PipelineRun
metadata:
  name: "cancel-in-progress"
  annotations:
    pipelinesascode.tekton.dev/cancel-in-progress: "true"
spec:
# ...

With this release, you can match pipeline runs to pull requests by using labels. To match a pipeline run to a label, set the pipelinesascode.tekton.dev/on-label annotation in PipelineRun resources. Adding the label to a pull request or push event immediately triggers the pipeline run. If a push request or pull event still has the label and is updated with a commit, the pipeline run is triggered again.

This functionality is supported on GitHub, GitLab, and Gitea repository providers. It is not supported on Bitbucket Cloud and Bitbucket Data Center.

Use the cache-upload and cache-fetch step actions to preserve the cache directory where a build process keeps its dependencies, storing it in an S3 bucket, Google Cloud Services (GCS) bucket, or an OCI repository.

The cache-fetch and cache-upload step actions are installed by default in the openshift-pipelines namespace.