Azure/AKS

AKS node pool Autoscaling not working properly

Closed this issue · 1 comments

shawnz88 commented

Describe the bug
Default autoscaling profile for node pool not scaling down nodes properly.

I have a v1.30.3 AKS cluster with a single node pool that's configured for autoscaling with default profile. Min is set to 2 and max is set to 10 for number of nodes. Currently 6 nodes are running.

The CPU utilization for each node is equal or less than 11%. Memory utilization for each node is: 60%, 58%, 48%, 41%, 33%, 28%.

Expected behavior
The default scale-down utilization is 50%. So node3, node4, node5 and node6 should have been scaled down and the cluster should have 2 or 3 nodes running instead of 6.

Screenshots
image
image

Environment (please complete the following information):

  • Kubernetes version [1.30.3]
Sagarsinghinfra360 commented

Hi @shawnz88 were you able to determine the cause of this issue? I am experiencing the same problem with my AKS node pools and would appreciate any insights you might have.