[codelab-pipelines-gis-feedback]:

[rs9899]

Reports Unsupported kubeflow version on page 3 when deploying kubeflow from https://deploy.kubeflow.cloud

All the steps previous to this was followed as in https://codelabs.developers.google.com/codelabs/cloud-kubeflow-pipelines-gis/?_ga=2.190031275.-1249852803.1588702117#2

Thanks

[issue-label-bot[bot]]

Issue-Label Bot is automatically applying the labels:

Label	Probability
kind/bug	0.73

Please mark this comment with :thumbsup: or :thumbsdown: to give our bot feedback! Links: app homepage, dashboard and code for this bot.

[jlewi]

@rs9899 Could you elaborate please? What reports "unsupported version"? Is this an error returned by click to deploy? I'm unclear whether this is referring to a version of GKE or Kubeflow?

[rs9899]

Sorry, but even I was unclear about it. It just pops as a dialogue box and that's it. Nothing is deployed. I had a report option there which resulted in an issue to this repo.

Apart from this I don't know much about it.

[jlewi]

Pops up on deploy.kubeflow.cloud? You should switch to deploying with the CLI; support for deploy.kubeflow.cloud is deprecated

[rs9899]

https://deploy.kubeflow.cloud

Arent we supposed to use the web interface for this? Thats what the docs said here

[8bitmp3]

Hi @jlewi If support for the UI (deploy.kubeflow.cloud) is deprecated, should we update the docs here:

Follow these steps to open the deployment UI and deploy Kubeflow on GCP:
1. Open https://deploy.kubeflow.cloud/ in your web browser. You should see a form like the one in the above screenshot.

I understand that on Google Cloud you go to AI Platform > Pipelines, which takes you to Kubeflow Pipelines. Then, you configure it by creating a GKE cluster and deploy the pipelines. I can assist with updating the docs. There are instructions for setting up KF Pipelines from the GCP Marketplace using the CLI here, I think.

[jlewi]

The marketplace deployment only deploys KFP nothing else in KF. So if the example is using other applications in KF that won't work.

We should probably update the docs to use the CLI.

/cc @amygdala

[amygdala]

Yes, the codelab does need to be updated to use the CLI now.

I think the example should mostly work on standalone GKE KFP aside from:

• the istio routing for the 'web app' that is deployed wouldn't work. Perhaps port-forwarding to that service would work instead.
• the notebook section would need to be overhauled (to use AI Platform Notebooks instead)

So, it might be worth creating two new variants of the example, one for KFP on AI Platform, and the other for KF with CLI installation. I'll put this on my stack :), hopefully should get to it pretty soon.

[8bitmp3]

Thanks for the clarification on standalone KFP in GCP Marketplace @jlewi.

When testing AI Pipelines using the UI, I am more keen to use the Cloud Shell on GCP. To use the full KF version you run kfctl apply -V -f {... kubeflow/... /v1.0-branch/..gcp...yaml}. Happy to see there's a separate "lighter" KF Pipelines v0.5.1 here we can configure with kfctlapply -V -f {... kubeflow/pipelines/.../kustomize/cluster-scoped-resources...}.

@amygdala if you're open to collaborate on updating the github_issue_summarization example, let me know. I've created a few tutorials based on GCP before (including with GKE and Istio mesh setups).

Thanks everyone 👍

[issue-label-bot[bot]]

Issue-Label Bot is automatically applying the labels:

Label	Probability
area/docs	0.56

Please mark this comment with :thumbsup: or :thumbsdown: to give our bot feedback! Links: app homepage, dashboard and code for this bot.

[stale[bot]]

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.

[amygdala]

/reopen

https://github.com/kubeflow/examples/pull/823

[k8s-ci-robot]

@amygdala: Reopened this issue.

In response to [this](https://github.com/kubeflow/examples/issues/800#issuecomment-697876975): >/reopen > >https://github.com/kubeflow/examples/pull/823 Instructions for interacting with me using PR comments are available [here](https://git.k8s.io/community/contributors/guide/pull-requests.md). If you have questions or suggestions related to my behavior, please file an issue against the [kubernetes/test-infra](https://github.com/kubernetes/test-infra/issues/new?title=Prow%20issue:) repository.

[stale[bot]]

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.