Skip to content

Latest commit

 

History

History

karpenter

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
 
 
 
 
 
 

Testing the karpenter CRD syncing

Prerequisites

  • This example assumes that you have a an AWS EKS cluster configured with Karpeneter to provision nodes.
  • Most of the examples listed here also create the related ClusterRoles and/or Roles needed for the relevant tool/framework.
  1. Once you've followed the installation instructions for setting up and installing karpenter on the host cluster and verified that the installation works, proceed to the next step of creating the vcluster
  2. Create a vcluster with the above config as a values file
vcluster create vcluster -f config.yaml
  1. Connect to the vcluster and apply the crds. Update lines 7 and 9 with the EKS cluster name for auto-discovery features to work
kubectl apply -f ./karpenterCrds/provisioner-node.yaml
  1. Create a workload and set the nodeselector to match the labels set within the Provisioner. If not, the node will not be scheduled by karpenter.
nodeSelector:
  project: vcluster