[DO NOT MERGE] Using parent devfile for Ollama

[ibuziuk]

What does this PR do?:

Using parent defile for Ollama devfile

Which issue(s) this PR fixes:

Link to github issue(s)

PR acceptance criteria:

• Contributing guide
  Have you read the devfile registry contributing guide and followed its instructions?
• Test automation
  Does this repository's tests pass with your changes?
• Documentation
  Does any documentation need to be updated with your changes?
• Check Tools Provider
  Have you tested the changes with existing tools, i.e. Odo, Che, Console? (See devfile registry contributing guide on how to test changes)

How to test changes / Special notes to the reviewer:

https://workspaces.openshift.com/#https://raw.githubusercontent.com/devfile/registry/3e49fdb71c6f77d99233ca1d1d266641f452a11b/stacks/ollama/devfile.yaml

[openshift-ci]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by: ibuziuk
Once this PR has been reviewed and has the lgtm label, please assign michael-valdron for approval. For more information see the Kubernetes Code Review Process.

The full list of commands accepted by this bot can be found here.

Needs approval from an approver in each of these files:

• OWNERS

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[ibuziuk]

@thepetk @elsony folks, any ideas why we are hitting Some Attributes are already defined in parent: controller.devfile.io/storage-type. If you want to override them, you should do it in the parent scope. - https://github.com/devfile/registry/actions/runs/9195275297/job/25290700742#step:4:891 ?

it is not possible to reference parent that is ephemeral ? cc: @AObuchow to follow up during my PTO

[thepetk]

/retest

[thepetk]

@thepetk @elsony folks, any ideas why we are hitting Some Attributes are already defined in parent: controller.devfile.io/storage-type. If you want to override them, you should do it in the parent scope. - https://github.com/devfile/registry/actions/runs/9195275297/job/25290700742#step:4:891 ?

it is not possible to reference parent that is ephemeral ? cc: @AObuchow to follow up during my PTO

I've taken a thorough look to see if there's any issue with the way we merge the flatten content and we check overrides but I haven't found anything. I would expect to see this error in case both the child & parent devfile define the same attribute (e.g. controller.devfile.io/storage-type: ephemeral).

More detailed if the parent already defines this attribute the correct way to override it in the child would be:

parent:
    attributes:
        controller.devfile.io/storage-type: ephemeral

As I see for the current state of the PR this check is successful I suspect that the mentioned check is for this commit, only because this is a case that we should raise an error that someone is trying to override an attribute already set by the parent.

[AObuchow]

I believe the check for non-terminating images is failing because the logic for getting the first container component's image is not able to fetch container components from the parent devfile. This is why we're seeing null for the component image in the command output:

Testing /home/runner/work/registry/registry/stacks/ollama/devfile.yaml
  COMMAND: minikube ssh docker pull null

[thepetk]

Testing /home/runner/work/registry/registry/stacks/ollama/devfile.yaml
COMMAND: minikube ssh docker pull null

Yes +1 I think the script should take care of the flattening of the devfile before checking for non terminating images. I've opened issue devfile/api#1568

[AObuchow]

Testing /home/runner/work/registry/registry/stacks/ollama/devfile.yaml
COMMAND: minikube ssh docker pull null

Yes +1 I think the script should take care of the flattening of the devfile before checking for non terminating images. I've opened issue devfile/api#1568

@thepetk Thanks for verifying and opening the related issue :)

[thepetk]

I think here it would be better to use a specific commit instead of the main branch. Wdyt?

[AObuchow]

This makes sense given that we are versioning the devfile (in this case, 1.0.1). We might want to instead have a branch in the parent devfile repo for each devfile version, and "release" new versions of the devfile, rather than having a rolling release. This is definitely something to gather consensus for devfile/api#1579.

[thepetk]

Yes +1 I think is important to create some guidelines regarding the versioning of every parent devfile too. This is one of the reasons the spike was opened. For using a release branch I think it also makes sense.

cc @devfile/devfile-services-team in case someone else has input for the usage of a release branch.

[openshift-ci]

@ibuziuk: The following test failed, say /retest to rerun all failed tests or /retest-required to rerun all mandatory failed tests:

Test name	Commit	Details	Required	Rerun command
ci/prow/v4.15-registry-test	a43d360	link	true	/test v4.15-registry-test

Full PR test history. Your PR dashboard.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository. I understand the commands that are listed here.

[openshift-merge-robot]

PR needs rebase.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository.