Kubeflow Notebooks: Cannot select `Allow access to Kubeflow Pipelines` from Configurations Dropdown

[ncbkr]

Summary

I want to give my notebook server access to Kubeflow Pipelines. I cannot select it from the dropdown. When I try to connect to the Pipelines API Server, I get status 500 and error message: "Internal error: Unauthenticated: Request header error: there is no user identity header".

Reproduction

in Kubeflow, go to the notebooks webapp and try to check the "Allow access to Kubeflow Pipelines" option in the Configurations Dropdown

Dagit UI/UX Issue Screenshots

Additional Info about Your Environment

Plural managed EKS cluster with Kubeflow

---

Message from the maintainers:

Impacted by this bug? Give it a 👍. We factor engagement into prioritization.

[ncbkr]

The PodDefault resources in the user namespaces are missing. That's why it is not possible to select "Allow access to Kubeflow Pipelines"

apiVersion: kubeflow.org/v1alpha1
kind: PodDefault
metadata:
  name: access-ml-pipeline
  namespace: <user-namespace>
spec:
  desc: Allow access to Kubeflow Pipelines
  env:
  - name: KF_PIPELINES_SA_TOKEN_PATH
    value: /var/run/secrets/kubeflow/pipelines/token
  selector:
    matchLabels:
      access-ml-pipeline: "true"
  volumeMounts:
  - mountPath: /var/run/secrets/kubeflow/pipelines
    name: volume-kf-pipeline-token
    readOnly: true
  volumes:
  - name: volume-kf-pipeline-token
    projected:
      sources:
      - serviceAccountToken:
          audience: pipelines.kubeflow.org
          expirationSeconds: 7200
          path: token

[ncbkr]

i assume 8a758fc670dd2a735007ab175c0e825dd3046921 allows the access token to be mounted, if it was selected?

What would be the right place to add the above PodDefault? It should be available as soon as the namespace is created, correct?