KinD is a solution that allows to quickly create local Kubernetes clusters, ideal for development or testing tasks. However, once these clusters are created, KinD does not have built-in features to modify the cluster configuration by adding or removing nodes, whether they are control-planes or workers, and the entire cluster would need to be regenerated from scratch.
KindScaler comes to facilitate this task. After dissecting how KinD creates and adds different nodes and roles to the cluster, this bash script has been created and allows to add both workers and control planes.
./kindscaler.sh <clustername> -r <control-plane|worker> -c <count>
For example, adding 3 workers to cluster kind:
$kind create cluster --config cluster.yaml
Creating cluster "kind" ...
✓ Ensuring node image (kindest/node:v1.28.0) 🖼
✓ Preparing nodes 📦 📦 📦
✓ Writing configuration 📜
✓ Starting control-plane 🕹
✓ Installing CNI 🔌
✓ Installing StorageClass 💾
✓ Joining worker nodes 🚜
Set kubectl context to "kind-kind"
You can now use your cluster with:
kubectl cluster-info --context kind-kind
Not sure what to do next? 😅 Check out https://kind.sigs.k8s.io/docs/user/quick-start/
$kubectl get nodes
NAME STATUS ROLES AGE VERSION
kind-control-plane Ready control-plane 34s v1.28.0
kind-worker Ready <none> 13s v1.28.0
kind-worker2 Ready <none> 8s v1.28.0
$./kindscaler.sh kind -r worker -c 3
Adding kind-worker3 node to kind cluster... Done!
Adding kind-worker4 node to kind cluster... Done!
Adding kind-worker5 node to kind cluster... Done!
$kubectl get nodes
NAME STATUS ROLES AGE VERSION
kind-control-plane Ready control-plane 109s v1.28.0
kind-worker Ready <none> 88s v1.28.0
kind-worker2 Ready <none> 83s v1.28.0
kind-worker3 Ready <none> 49s v1.28.0
kind-worker4 Ready <none> 36s v1.28.0
kind-worker5 Ready <none> 7s v1.28.0
kubectl delete node <nodename>
docker stop <container>
docker container rm <container>