kubeflow / pipelines

Machine Learning Pipelines for Kubeflow
https://www.kubeflow.org/docs/components/pipelines/
Apache License 2.0
3.6k stars 1.62k forks source link

[frontend] <Unable to configure minio accesskey using manifests> #6493

Closed Hmr-ramzi closed 4 months ago

Hmr-ramzi commented 3 years ago

Environment

Steps to reproduce

By providing an access key to minio that is different than minio123 the frontend is unable to access minio

Expected result

Once the backend is configured with the right access key this is also assumed to be available for the frontend when we deploy kubeflow with the kubeflow manifests

Materials and Reference

https://github.com/kubeflow/pipelines/blob/cc83e1089b573256e781ed2e4ac90f604129e769/frontend/server/configs.ts

Impacted by this bug? Give it a 👍. We prioritise the issues with the most 👍.

zijianjoy commented 3 years ago

Hello @Hmr-ramzi , currently there is no good way to programmatically update minio access key in multi-user mode.

You can change minio access key using kubectl following this https://github.com/kubeflow/pipelines/blob/a35812429c9e57b29e7b559814f299af8c2fd305/manifests/kustomize/base/installs/multi-user/pipelines-profile-controller/sync.py#L346-L357, based on your user namesapce.

Welcome contribution to design and implement this for a better usage!

Hmr-ramzi commented 3 years ago

@zijianjoy Thanks for your response. Unfortunately, I changing the secret in the user namespace and also this does not get it to work. I can change the part related to the backend but the frontend according to my tests only works with accesskey minio123 and if i use a different accesskey (even updated in the secret in the user namespace) it responds with a complain from Minio that the key is not valid. From what i see here is that these credentials are hard coded in the frontend. link

marrrcin commented 3 years ago

@Hmr-ramzi the credentials are indeed hard-coded, but they can be overriden by setting appropriate environment variables. Take a look at the platform-agnostic Helm Chart for Kubeflow Pipelines: https://github.com/getindata/helm-charts/tree/main/charts/kubeflow-pipelines which sets custom MinIO accesskey and secretkey via values: https://github.com/getindata/helm-charts/blob/c17c10622ed6ba0f324d8a5432c615bf20864d1b/charts/kubeflow-pipelines/templates/pipeline.yaml#L589 in the metadata-ui Deployment and other places where it's necessary to work.

stale[bot] commented 2 years ago

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[bot] commented 4 months ago

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