[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 assign kimwnasptd for approval. For more information see the Kubernetes Code Review Process.

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

@christian-heusel I think the best structure is to keep everything under applications, in a folder called workspaces.

Structure should be:

applications:
  - workspaces:
       - upstream:
           - controller:
               - base, overlays, components, etc
           - backend:
               - base, overlays, components, etc
           - frontend:
               - base, overlays, components, etc
      - overlays:
           - istio:
               - xyz_patch.yaml (if needed can be included).
               - kustomization.yaml (points to the `overlays/istio` of each upstream)

Then we have a single sync target under upstream.

/hold

@christian-heusel please add a basic integration test, similar to 1. https://github.com/kubeflow/manifests/blob/master/.github/workflows/pipeline_run_from_notebook.yaml 2. https://github.com/kubeflow/manifests/blob/master/.github/workflows/notebook_controller_test.yaml 3. https://github.com/kubeflow/manifests/blob/master/.github/workflows/jupyter_web_application_test.yaml

if you want you can also kill the notebooks v1 helm charts in this PR. Since we will kill v1 there is no potential to upstream them in the future.

/retest
Maybe also @danish9039 can help.

@christian-heusel failures here are a combination of the prerelease image-tag mismatch (v2.0.0-alpha.1 in manifests https://github.com/kubeflow/manifests/pull/3430/changes#diff-2145495ad76e99d7bebb590b3ab3c4fb6dec9b87f0092b3ff4c7adf9cd36311dR20 vs 2.0.0-alpha.1 in GHCR) plus stale workflow assumptions in the new tests

global all-namespaces pod wait that times out on unrelated KinD dependencies https://github.com/kubeflow/manifests/pull/3430/changes#diff-8b3cd760feec7fb87c30efb305a6f3031f13ede999a21dcf1aa2c7bc5a88e8acR51

should wait first for cert-manager pods and then for pods in the only in kubeflow-workspaces namespace

kubectl wait --for=condition=Ready pods -n cert-manager --all --timeout=120s
        kubectl wait --for=condition=Ready pods \
          -n kubeflow-workspaces --all \
          --timeout=600s

controller test hitting the old /workspaces/api/v1/namespaces/{ns}/workspaces route https://github.com/kubeflow/manifests/pull/3430/changes#diff-8b3cd760feec7fb87c30efb305a6f3031f13ede999a21dcf1aa2c7bc5a88e8acR76 instead of /workspaces/api/v1/workspaces/{ns}

pipeline test assuming the workspace pod is always named test-0 even though the controller now generates that name dynamically .https://github.com/kubeflow/manifests/pull/3430/changes#diff-0dbda89ab22f251a228374464802a7f6b04345db1a5292dcbbb8b264b876f255R84

i have fixed and validated the changes required ( see commits on my fork danish9039@d76d396 )

@danish9039 thanks a lot for your suggestions and fixes, I have added them to the PR and made you a Co-author on the relevant commit 🤗

@christian-heusel @danish9039

The dashboard might be affected by this:

GHSA-fvcv-3m26-pcqx

i get warnings for ghcr.io/kubeflow/kubeflow/central-dashboard:v1.10.0, is that still valid for https://github.com/kubeflow/dashboard/pkgs/container/dashboard%2Fdashboard/783922119?tag=v2.0.0-rc.1 ?

@juliusvonkohout yet it seems like this (see kubeflow/dashboard#266 for the pending updates), the dashboard frontend components are mostly untouched AFAIU.

Also this is unrelated to the workspaces update 🤗