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

[ML] Allow force delete for a data frame transform #43961

Closed
sophiec20 opened this issue Jul 4, 2019 · 2 comments · Fixed by #44590
Closed

[ML] Allow force delete for a data frame transform #43961

sophiec20 opened this issue Jul 4, 2019 · 2 comments · Fixed by #44590
Assignees
Labels
:ml/Transform Transform :ml Machine learning

Comments

@sophiec20
Copy link
Contributor

sophiec20 commented Jul 4, 2019

Found in 7.3.0-SNAPSHOT 03-Jul-2019

In order to delete a failed transform, the user must first _stop?force it and then DELETE.

This is a theoretical two step process but in reality they will probably try DELETE and see a failure, then _stop and see a failure, then _stop?force and then the final DELETE.

We should allow the user to delete a failed transform without directing them through this convoluted path. By having a force-delete, we can force-stop in the background thus removing the persistent task and then delete the data frame transform.

This can be used by the UI in order to delete failed transforms.

For the context of the UI experience, note that the error in the transform (the reason for the failure) should be visible in the messages tab in the UI and therefore we do not need to also show it to the user when they are trying to delete it. UI ref elastic/kibana#40298

@sophiec20 sophiec20 added :ml Machine learning :ml/Transform Transform labels Jul 4, 2019
@elasticmachine
Copy link
Collaborator

Pinging @elastic/ml-core

@walterra walterra assigned walterra and unassigned walterra Jul 4, 2019
@benwtrent benwtrent self-assigned this Jul 18, 2019
@droberts195
Copy link
Contributor

To clarify, force delete should not just unconditionally delete failed transforms, but should unconditionally delete transforms whatever their current state. As noted, the envisaged use case is the UI, where the current state is visible and an "are you sure" confirmation can be displayed if we want to protect users from deleting transforms that are progressing successfully.

(This is consistent with anomaly detectors where force delete will delete jobs whether they are open, closed or failed.)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
:ml/Transform Transform :ml Machine learning
Projects
None yet
Development

Successfully merging a pull request may close this issue.

5 participants