Mike Dame

Docs PR opened at https://github.com/kubernetes/website/pull/26993, which just copies the comments from code into the docs

@ahg-g yes, I'll update the KEP with what's necessary for Beta graduation

@reylejano thanks for the update, we have a PR to update the KEP that is currently undergoing review at https://github.com/kubernetes/enhancements/pull/2691 to make the changes you have listed. This should be...

Hi @chrisnegus, I opened the placeholder PR here: https://github.com/kubernetes/website/pull/28102 Besides the status change from Alpha to Beta, I think the only other user-facing change for this release will be a...

@reylejano yes, that and the docs PR (https://github.com/kubernetes/website/pull/28102) are all that is needed for this. Thanks!

@reylejano that PR was tagged for merge this morning and should have no issues making it in by Thursday 🙂

NodeFit doesn't currently check against any topology constraints (hard or soft). This is intentional, because NodeFit is meant as a safeguard check against criteria that the strategy might not be...

It's indeed not a perfect solution in many cases. Originally, the topology constraint descheduler strategy didn't even have the option to evict based on soft constraints for this exact reason....

Hi @rittneje, while I haven't worked directly on the cluster autoscaler, I know that it relies heavily on the scheduler framework (and actually parts of the scheduler framework were written...

Hi @y0zg, are you seeing descheduler remove pods with other status codes besides `Error`, `ContainerStatusUnknown`, and `Completed`? And are the pods with these statuses at least 3600 seconds old as...