Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[backend] metadata-writer cannot save metadata of S3 artifacts with argo v3.1+ #8283

Closed
tktest1234 opened this issue Sep 20, 2022 · 7 comments

Comments

@tktest1234
Copy link

Environment

Steps to reproduce

  1. setting to save artifacts to S3 (change configmap for workflow-controller)
  2. Run pipeline: [Tutorial] Data passing in python components

metadata-writer output log

Kubernetes Pod event:  ADDED file-passing-pipelines-txmqj-2254123803 14598940
Traceback (most recent call last):
  File "/kfp/metadata_writer/metadata_writer.py", line 238, in <module>
    artifact_uri = argo_artifact_to_uri(argo_artifact)
  File "/kfp/metadata_writer/metadata_writer.py", line 106, in argo_artifact_to_uri
    provider=get_object_store_provider(s3_artifact['endpoint']),
KeyError: 'endpoint'

result:

  • Artifact in metadb is empty

Expected result

metadata for the artifacts is logged into metadb

Materials and Reference

I think it is caused by changes for argo Key-Only Artifacts.

In https://github.com/kubeflow/pipelines/blob/1.7.0/backend/metadata_writer/src/metadata_writer.py#L318, metadata-writer read "workflows.argoproj.io/outputs", but endpoint and bucket are not passed by argo3.1+. So https://github.com/kubeflow/pipelines/blob/1.7.0/backend/metadata_writer/src/metadata_writer.py#L105 cannot set the correct URL.

Related frontend issue: #5930


Impacted by this bug? Give it a 👍.

@tktest1234 tktest1234 changed the title [backend] metadata-writer cannot save metadata of artifacts to S3 with argo v3.1+ [backend] metadata-writer cannot save metadata of S3 artifacts with argo v3.1+ Sep 20, 2022
@chensun chensun assigned chensun and unassigned zijianjoy Sep 22, 2022
@surajkota
Copy link
Contributor

@tktest1234 Are you running on AWS and want to use KFP with S3 as storage backend?

@tktest1234
Copy link
Author

tktest1234 commented Oct 16, 2022

Yes, I install kubeflow1.4 on AWS(EKS) and want to use KFP with S3 as storage backend.
Isn't it expected?

@surajkota
Copy link
Contributor

surajkota commented Oct 20, 2022

Hey @tktest1234 please follow the instructions on AWS distribution of Kubeflow to install KFP with S3 as artifact storage https://awslabs.github.io/kubeflow-manifests/docs/deployment/rds-s3/

Please create an issue on the awslabs repository if you face any issues

@tktest1234
Copy link
Author

Thank you, but awslabs's installation doesn't solve the problem.

#5829 is exactly the PR for this issue. I hope it to progress.

Copy link

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 github-actions bot added the lifecycle/stale The issue / pull request is stale, any activities remove this label. label Feb 27, 2024
@rimolive
Copy link
Member

rimolive commented Apr 3, 2024

Closing this issue. No activity for more than a year.

/close

@stale stale bot removed the lifecycle/stale The issue / pull request is stale, any activities remove this label. label Apr 3, 2024
Copy link

@rimolive: Closing this issue.

In response to this:

Closing this issue. No activity for more than a year.

/close

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.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
Status: Closed
Development

No branches or pull requests

5 participants