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

[feature] delete MinIO artifacts on pipeline run deletion #10816

Open
tmvfb opened this issue May 10, 2024 · 1 comment
Open

[feature] delete MinIO artifacts on pipeline run deletion #10816

tmvfb opened this issue May 10, 2024 · 1 comment

Comments

@tmvfb
Copy link
Contributor

tmvfb commented May 10, 2024

Feature Area

/area frontend
/area backend
/area components

What feature would you like to see?

When pipeline run gets deleted, remove associated artifacts from MinIO storage and from artifacts list/DB.

What is the use case or pain point?

Consistently running kubeflow pipelines on a production cluster produces a lot of artifacts that use up a lot of storage resources. That leads to a necessity of periodical storage clean up. This process could be made much easier if the artifacts got automatically deleted on run deletion.

I found some issues that seem to address either this or similar matter, namely #8104, #5509, #5667, #5783 (related to DB entries), #7939 (might be related).

Comment #8104 (comment) describes exactly the behaviour I would expect from kubeflow pipelines as an unexperienced user.

Is there a workaround currently?

Our solution is manual cleanup which is surely not effective. If there is currently a way to setup an automated cleanup, we'd be happy to know!


Love this idea? Give it a 👍.

@HumairAK
Copy link
Contributor

So artifacts aren't necessarily tied to a specific pipeline run, for instance:

Pipeline Run can produce Artifact X

But Artifact X can go on to later be used by a different pipeline run (cached).

I'm not sure the details of imported artifacts, but I think you could also technically import Artifact X in a completely separate pipeline run.

So we would need to consider how to handle these cases.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

2 participants