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

[Security Solution][Detections] Ensure ML Detection Rules/Jobs function with introduction of Space Aware ML Jobs #82553

Closed
spong opened this issue Nov 4, 2020 · 5 comments
Assignees
Labels
chore Feature:Detection Rules Security Solution rules and Detection Engine Team:Detections and Resp Security Detection Response Team Team: SecuritySolution Security Solutions Team working on SIEM, Endpoint, Timeline, Resolver, etc. v7.11.0

Comments

@spong
Copy link
Member

spong commented Nov 4, 2020

With the introduction of Space Aware ML Jobs (#77916), there will be a few tweaks we'll need to do on the Detections side to ensure everything continues to function as intended. Since new jobs will be scoped to the current space when created, we'll need to update these requests to specify that the new jobs be scoped to the * space to maintain the current behavior of ML Rules/Jobs between spaces.

Additionally, since we're sticking to global ML Jobs for this release, ML may potentially add a warning to the ML Job/Spaces UI to warn users that they may break existing Detection Rules when changing the configured space for ML Jobs with the security group.

@spong spong added chore Feature:Detection Rules Security Solution rules and Detection Engine v7.11.0 Team:Detections and Resp Security Detection Response Team Team: SecuritySolution Security Solutions Team working on SIEM, Endpoint, Timeline, Resolver, etc. labels Nov 4, 2020
@rylnd rylnd self-assigned this Dec 21, 2020
@rylnd
Copy link
Contributor

rylnd commented Jan 6, 2021

A few updates here:

TL;DR: this all works, but I don't believe there's any UI to distinguish space-aware from -agnostic (aka global, '*' space) jobs, which is potentially confusing to the user.

  • The ML popover correctly creates global jobs. That change was introduced in this commit
    • These jobs are visible in any space via the ML API: in the ML popover, Anomalies tabs, etc.
  • If desired, users can also create custom jobs that are space-aware.
    • This is the default when creating new jobs via the ML UI, and I did not see any way to override that behavior.
  • In a given space, ML Rules can be created for both global jobs and/or space-specific jobs (assuming the job is visible in that space)

Potential UX improvements

  • Adding a note to the ML Popover that jobs will be created in the * space
  • Distinguishing the space of a given job in the ML Popover
  • Distinguishing the space within the ML Jobs table

A global and a space-specific job in the Jobs table (expanded views are similarly identical):

Anomaly_Detection_-_Machine_Learning_-_Elastic

A global and a space-specific job in the ML Popover:

Detections_-_Kibana

@rylnd
Copy link
Contributor

rylnd commented Jan 6, 2021

@jgowdyelastic RE the above: is there a way to determine the space of the job in the ML UI that I've overlooked?

@jgowdyelastic
Copy link
Member

@rylnd a job's spaces can be viewed in Stack Management, under the Machine Learning Jobs section.
Spaces can be assigned and removed from jobs.
We do not list a job's spaces in the main Machine Learning plugin.
Space management was added in this PR #83320

@rylnd
Copy link
Contributor

rylnd commented Jan 6, 2021

Thanks @jgowdyelastic, I knew I'd seen that data somewhere! @spong here's what that looks like:

Elastic

@spong
Copy link
Member Author

spong commented Jun 21, 2021

Closing this issue as this functionality has been shipped and verified.

@spong spong closed this as completed Jun 21, 2021
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
chore Feature:Detection Rules Security Solution rules and Detection Engine Team:Detections and Resp Security Detection Response Team Team: SecuritySolution Security Solutions Team working on SIEM, Endpoint, Timeline, Resolver, etc. v7.11.0
Projects
None yet
Development

No branches or pull requests

3 participants