Description

Observed Behavior:

{"level":"DEBUG","time":"2025-02-17T17:23:51.179Z","logger":"controller","caller":"disruption/controller.go:99","message":"marking consolidatable","commit":"a79abac","controller":"nodeclaim.disruption","controllerGroup":"karpenter.sh","controllerKind":"NodeClaim","NodeClaim":{"name":"default-pwzrz"},"namespace":"","name":"default-pwzrz","reconcileID":"05fd3c4d-cad7-4979-862f-de6cee15988e","Node":{"name":"ip-192-168-106-221.us-west-2.compute.internal"}}
{"level":"DEBUG","time":"2025-02-17T17:23:51.372Z","logger":"controller","caller":"disruption/controller.go:99","message":"marking consolidatable","commit":"a79abac","controller":"nodeclaim.disruption","controllerGroup":"karpenter.sh","controllerKind":"NodeClaim","NodeClaim":{"name":"default-pwzrz"},"namespace":"","name":"default-pwzrz","reconcileID":"957af1c8-0966-47ea-be7e-8ae2697ca4df","Node":{"name":"ip-192-168-106-221.us-west-2.compute.internal"}}

Karpenter will double-log that a status condition has transitioned for Consolidatable (the same probably applies to Drifted) because there is no read-after-write consistency so Karpenter needs enough time after it has mutated the object to actually see the cache update -- one way to resolve this is to do an event filter on status updates -- the other way to accomplish this is by adding an artificial sleep into the Reconcile to give the cache time to catch-up like here

Expected Behavior:

Karpenter should not produce the same log twice in a row.

• 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