KEP-2170: Deploy JobSet in kubeflow-system namespace

[andreyvelich]

Part of: #2170

Depends on: kubernetes-sigs/jobset#751, we should update the JobSet version once this fix is released.

I updated the Kubeflow Training V2 manifests to deploy JobSet in the kubeflow-system namespace. Also, given that the Training Runtimes need to be installed after the manager is deployed and is running, I think we should split out overlays between these two:

• manager
• runtimes

/cc @kannon92 @kubeflow/wg-training-leads @kubeflow/wg-manifests-leads @kubeflow/release-team @Electronic-Waste @Doris-xm @astefanutti @seanlaii @deepanker13 @saileshd1402 @vsoch

[google-oss-prow]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by:
Once this PR has been reviewed and has the lgtm label, please ask for approval from andreyvelich. 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

[google-oss-prow]

@andreyvelich: GitHub didn't allow me to request PR reviews from the following users: seanlaii, saileshd1402, astefanutti, kubeflow/release-team, doris-xm, vsoch, kannon92, kubeflow/wg-manifests-leads.

Note that only kubeflow members and repo collaborators can review this PR, and authors cannot review their own PRs.

In response to this:

Depends on: kubernetes-sigs/jobset#752.
Part of: #2170

I updated the Kubeflow Training V2 manifests to deploy JobSet in the kubeflow-system namespace.

Also, given that the Training Runtimes need to be installed after the manager is deployed and is running, I think we should split out overlays between these two:

• manager
• runtimes

Cluster operators need to install manager first, and after than deploy runtimes.

/cc @kannon92 @kubeflow/wg-training-leads @kubeflow/wg-manifests-leads @kubeflow/release-team @Electronic-Waste @Doris-xm @astefanutti @seanlaii @deepanker13 @saileshd1402 @vsoch

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/test-infra repository.

[andreyvelich]

/hold for kubernetes-sigs/jobset#751

[astefanutti]

Would that make sense to group the JobSet related resources into a dedicated directory / overlay?
It could make things more modular and accommodate environments where the JobSet API / controller is already installed.

[andreyvelich]

Hmm, we can make it similar to KFP and move it into third-party/jobset/kustomization.yaml.
So the manifests will looks like this:

/v2/base/...
/v2/overlays/...
/v2/third-party/...

@kubeflow/wg-training-leads WDYT about it ?

[Electronic-Waste]

@andreyvelich SGTM:)