Support ML module Kibana type#152
Conversation
💚 Build Succeeded
Expand to view the summary
Build stats
Trends 🧪 |
|
Hi @jgowdyelastic, it's not obvious, but updating the changelog requires another |
ruflin
left a comment
There was a problem hiding this comment.
LGTM. But would be good to have also input from @mtojek and @ycombinator on this.
@jgowdyelastic One thing that would be useful is having just a quick note in the PR description on how this assets should be installed / removed / upgrade. I know you already have an open PR against Kibana that describes this in code. But someone coming back to the package spec change and wants to know how this assets work, it would be useful to have it here.
| @@ -0,0 +1 @@ | |||
| {} No newline at end of file | |||
There was a problem hiding this comment.
Would be nice to have some example content, mostly to see how it will look and as an example.
There was a problem hiding this comment.
I've added the NGINX module saved object in 05395ed
| pattern: '^.+\.json$' | ||
| - description: Folder containing ML module assets | ||
| type: folder | ||
| name: ml_module |
There was a problem hiding this comment.
Looking at the test package, I assume you want the folder under kibana/ to be named ml/?
| name: ml_module | |
| name: ml |
There was a problem hiding this comment.
I think the folder should be under kibana but named ml_module. There might be other ML assets in the future.
There was a problem hiding this comment.
Yes this is a mistake. The changes I had made to test elastic/kibana#94950 use ml_module
Updated in 92fab43
|
@jgowdyelastic Is this ready for review? |
@ruflin It is now! You were all so keen to look at it I hadn't had time to update the description or test it before taking it out of draft. |
|
@elasticmachine merge upstream |
Adds support for ML module asset types.
ML Module's are now supported in the form of saved objects in kibana. This PR adds the support for installing the modules via Fleet's integration packages.
Once a Module has been installed via Fleet, it is available for use in ML. This involved calling ML's "setup" endpoint which will create the contained Jobs, with optional overrides, and then optionally starting the Jobs.
An ML Module installed via Fleet should be free to be deleted or upgraded with all other assets in the package.
Deleting an ML Module will not remove any previously created Jobs.
Upgrading the ML Module will not update or affect in any way any previously created Jobs.
test/packagesthat prove my change is effective.versions/N/changelog.yml.Related issues
Related to:
#148
elastic/kibana#92855
elastic/kibana#94950