Open eugenea opened 1 week ago
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.
maybe instead of or in addition to minTTL there should be "stabilization window" for scaling down, to match the same functionality as HPA provides. StabilizationWindow would reduce amount of cluster changes where clusters are being used for spiky/batch workloads.
Description
What problem are you trying to solve?
We observe that karpenter sometimes disrupts node less than 1 min after it got launched. This creates several problems in the cluster:
How important is this feature to you?
It is pretty important because it is challenging to find workaround for some situations listed above.