Choosing when to run certain Integration Tests

Integration test scenarios can be configured to run in only certain cases (referred to as contexts in further text). Examples include running the integration test pipelines only in case of Pull Requests, or running them only for builds of specific components. The full list of supported contexts is:

"application" - runs the integration test in all cases - this is the default state
"component" - runs the integration tests only in case of component builds
- This context is run for builds of all components, but not for special cases like manually created Snapshots
“component_COMPONENT” - runs the integration test only for a build of a specific component
- For a component sample-component, this context would need to be component_sample-component
"pull_request" - runs the integration test in case of the Snapshot being created for a pull request event
"push" - runs the integration test in case of the Snapshot being created for a push event
"override" - runs the integration test for an override Snapshot
- See more in Creating an override snapshot
"disabled" - disables the execution of the given integration test if it’s the only context that’s defined
- This can be useful if the scenario in question needs to only be executed manually, see more about running integration tests manually in Rerunning an integration test

"group" - runs the integration test for a group Snapshot

See more in Creating a group snapshot

When multiple contexts are defined, the IntegrationTestScenario will execute if any of the specified contexts match the snapshot. This approach ensures that the test runs as long as at least one context applies, without requiring all defined contexts to be met.

Prerequisites

You have CLI access to the specific OpenShift cluster. For information on obtaining CLI access, refer Getting started with the CLI.

Procedure

Choose the IntegrationTestScenario that you want to run only in certain cases.

Edit the chosen IntegrationTestScenario using kubectl.

$ kubectl edit integrationtestscenario [integrationtestscenario name]

Modify the contexts field by setting the list of contexts you wish the IntegrationTestScenario.

The name field of every entry needs to be set to the chosen context name from the above list
The description can be used to note down why this context is set for the given IntegrationTestScenario

Example integrationTestScenario configuration with the single component context for the Component named sample-component:

apiVersion: appstudio.redhat.com/v1beta1
kind: IntegrationTestScenario
metadata:
  name: example-pass
  namespace: default
spec:
  application: application-sample
  contexts:
    - description: Single component testing
      name: component_sample-component
  resolverRef:
    resolver: git
    params:
      - name: url
        value: https://github.com/konflux-ci/integration-examples
      - name: revision
        value: main
      - name: pathInRepo
        value: pipelines/integration_pipeline_pass.yaml

Verification

To verify the configuration, run a component build in the desired context (e.g. as part of a Pull Request) and verify if the integration test pipeline was executed according to your expectations.