Skip to content

Autoscaling to 0 #443

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

Closed
nickwalton opened this issue Sep 9, 2019 · 5 comments
Closed

Autoscaling to 0 #443

nickwalton opened this issue Sep 9, 2019 · 5 comments
Labels
question Further information is requested

Comments

@nickwalton
Copy link

Does Cortex support auto scaling down to 0 if there is no current traffic? Couldn't find documentation about auto scaling.

@nickwalton nickwalton added the question Further information is requested label Sep 9, 2019
@ospillinger
Copy link
Member

ospillinger commented Sep 9, 2019

Yes, you would need to set min_replicas on each API to 0. I'll clarify this in the docs.

@nickwalton
Copy link
Author

Awesome thanks!

@ospillinger
Copy link
Member

@deliahu corrected me that we can only autoscale down to 1 replica right now because the autoscaler is based on CPU which is undefined when there are no replicas. I'm sorry about the miscommunication and we'll look into how we can support autoscaling to 0.

@nickwalton
Copy link
Author

Gotcha. Well if you are able to enable it then that would be a great feature. I'll add it as a feature request #445

@deliahu
Copy link
Member

deliahu commented Sep 10, 2019

Thanks @nickwalton, I agree and we'll definitely look into it! According to kubernetes/kubernetes#69687 it should be possible if we use custom metrics for autoscaling instead of CPU utilization.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
question Further information is requested
Projects
None yet
Development

No branches or pull requests

3 participants