Log when drift is not possible due to scheduling constrains

[oscarboher]

Description

What problem are you trying to solve?
When a nodeclaims are drifted, karpenter immediately marks the nodes as drifted. But if the scheduling simulation does not find capacity for the pods the nodes are never changed. This is OK as behaviour, but It would be interesting to have a log indicating that, as by default you expect the nodes to drift, even if it is only with log level debug.

For example, having a single nodepool where workloads are running, and adding a taint to the nodepool will mark the nodes as drifted, but if there are pods running in the nodes that do not have the required tolerations the nodes won't be drifted.

This behaviour raises a question, if there is nodegroups that are not managed by karpenter (aws managed nodegroups), are this considered during the simulation? Or that capacity won't be taken into account?.

How important is this feature to you?
Nice to have

• Please vote on this issue by adding a 👍 reaction to the original issue to help the community and maintainers prioritize this request
• Please do not leave "+1" or "me too" comments, they generate extra noise for issue followers and do not help prioritize the request
• If you are interested in working on this issue or have submitted a pull request, please leave a comment

I'll try to see if I can do a contribution, if you have any guidance on where to start it would be welcome.

[k8s-ci-robot]

This issue is currently awaiting triage.

If Karpenter contributors determines this is a relevant issue, they will accept it by applying the triage/accepted label and provide further guidance.

The triage/accepted label can be added by org members by writing /triage accepted in a comment.

Details

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-sigs/prow repository.