-
Notifications
You must be signed in to change notification settings - Fork 38.7k
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
Resolution of preemption_victims metric is limiting #112912
Comments
/sig scheduling @logicalhan is the proposal backwards compatible? The guide is not very clear about changing the buckets, but I seem to remember it's ok to change them |
Changing buckets is fine actually. |
Can I work for this issue?If there is a better option, feel free to unassign. @alculquicondor |
/assign |
/triage accepted |
/close |
@lengrongfu: You can't close an active issue/PR unless you authored it or you are a collaborator. In response to this:
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. |
/close If you set |
@pacoxu: Closing this issue. In response to this:
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. |
What would you like to be added?
Have better resolution of the metric. The range is probably good (0 to 50), but the resolution is not very useful. Maybe it should be an exponential scale instead of linear
kubernetes/pkg/scheduler/metrics/metrics.go
Line 81 in e11e226
Why is this needed?
We should have more visibility around average number of preemptions. It's not possible to see if pods require less than 5 preemptions to inform how expensive preemption is for a certain QPS.
The text was updated successfully, but these errors were encountered: