Kubernetes Cluster Autoscaler

• https://github.com/kubernetes/autoscaler

k8s.io/cluster-autoscaler/

kubectl get pods -A | grep autoscaler
cluster-autoscaler              cluster-autoscaler-86cff49xx4-6tcht                        1/1     Running   1 
(109d ago)     123d

helm list -A | grep autoscaler
cluster-autoscaler            	cluster-autoscaler           	1       	2022-06-06 11:11:42.560486 +0200 CEST  	deployed	cluster-autoscaler-9.16.2         	1.23.0

 autoScaler:
   enabled: true
   min: 2
   max: 20

kubectl get pods -n cluster-autoscaler -o yaml cluster-autoscaler-0987654321

kubectl get events -o yaml | grep message
 message: '0/2 nodes are available: 2 Insufficient cpu.'
 message: 'pod didnt trigger scale-up: '

Related terms

• Horizontal Pod Autoscaler (HPA)
• kubectl scale deployment
• Karpenter
• autoScaler:
• FailedScheduling
• Control Plane
• Karpenter

Activities

• Read https://docs.aws.amazon.com/eks/latest/userguide/autoscaling.html#cluster-autoscaler

See also

• Horizontal Pod Autoscaler (HPA), Horizontal cluster-proportional-autoscaler container, Kubernetes cluster-proportional-vertical-autoscaler, Kubernetes Horizontal cluster-proportional-autoscaler container, FailedGetResourceMetric
• Kubernetes autoscaling, HPA, VPA, Karpenter, Kubernetes Autoscaler, kubectl autoscale, Kubernetes Metrics Server (KMS)