Kubeflow & ModelMesh-Serving Compatibility

[muratyarali]

Validation Checklist

• Is this a Kubeflow issue?
• Are you posting in the right repository ?
• Did you follow the Kubeflow installation guideline ?
• Is the issue report properly structured and detailed with version numbers?
• Is this for Kubeflow development ?
• Would you like to work on this issue?
• You can join the CNCF Slack and access our meetings at the Kubeflow Community website. Our channel on the CNCF Slack is here #kubeflow-platform.

Version

1.9

Describe your issue

I deployed ModelMesh-Serving in a Kubeflow K8S cluster (kubeflow version: v1.9.1-RC.2) and created an InferenceService (ISVC) with modelmesh as follows:

kubectl apply -f - <<EOF
apiVersion: serving.kserve.io/v1beta1
kind: InferenceService
metadata:
  name: example-sklearn-isvc
  namespace: modelmesh-serving
  annotations:
    serving.kserve.io/deploymentMode: ModelMesh
spec:
  predictor:
    model:
      modelFormat:
        name: sklearn
      storageUri: "gs://kfserving-examples/models/sklearn/1.0/model"
EOF

Annotation is used to select the model mesh for deploying an ISVC. When I create an ISVC with KServe (without annotation), I can view it on the KServe Endpoints tab in the Kubeflow UI, and clicking on it reveals the details.

However, when I create the ISVC with ModelMesh-Serving, I can see it listed in the KServe Endpoints tab, but clicking on it does not display the details and instead shows an error:

[404] The requested resource could not be found in the API Server https://www.example.com/kserve-endpoints/api/namespaces/modelmesh-serving/revisions/undefined

The Kubeflow-UI is looking for resources at /kserve-endpoints/api/namespaces/modelmesh-serving/revisions/ which don't exist for ModelMesh.

Can this be fixed or extended on the Kubeflow or ModelMesh-Serving side?

Steps to reproduce the issue

1. I checked this with Kubeflow
2. I checked this ModelMesh-Serving

Put here any screenshots or videos (optional)

No response

[juliusvonkohout]

Probably yes, but what we need is someone willing to create PRs to fix it.

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[github-actions]

This issue has been automatically closed because it has not had recent activity. Please comment "/reopen" to reopen it.

[juliusvonkohout]

/reopen

@Griffin-Sullivan can you help here?

[google-oss-prow]

@juliusvonkohout: Reopened this issue.

In response to this:

/reopen

@Griffin-Sullivan can you help here?

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.