How to bring down kubernetes cluster?

 Bringing down a Kubernetes cluster depends on how it was set up and the tools used to manage it. Below are general steps for different deployment types:

 

1. Using kubeadm

If the cluster was set up using kubeadm:

On the Control Plane Node:

1. Drain the node to safely evict workloads: 

2. kubectl drain <node-name> --ignore-daemonsets --delete-emptydir-data

3. Delete the control plane node: 

4. kubectl delete node <node-name>

5. Reset the cluster: 

6. sudo kubeadm reset

7. Clean up any residual configuration: 

8. sudo rm -rf /etc/cni/net.d

9. sudo rm -rf ~/.kube

On Worker Nodes:

1. Drain the node: 

2. kubectl drain <node-name> --ignore-daemonsets --delete-emptydir-data

3. Delete the node: 

4. kubectl delete node <node-name>

5. Reset the node: 

6. sudo kubeadm reset

7. Clean up configurations (similar to the control plane node): 

8. sudo rm -rf /etc/cni/net.d

9. sudo rm -rf ~/.kube

 

2. Managed Kubernetes Services

If you're using a managed service like GKE, EKS, or AKS:

• Use the provider's CLI or web console to delete the cluster.

Example for GKE:

gcloud container clusters delete <cluster-name> --zone <zone>

Example for EKS:

aws eks delete-cluster --name <cluster-name>

Example for AKS:

az aks delete --name <cluster-name> --resource-group <resource-group>

 

3. Minikube

If you're using Minikube (local Kubernetes):

1. Stop the cluster: 

2. minikube stop

3. Delete the cluster: 

4. minikube delete

 

4. Kind (Kubernetes IN Docker)

If you're using Kind:

1. List Kind clusters: 

2. kind get clusters

3. Delete the cluster: 

4. kind delete cluster --name <cluster-name>

 

5. Custom Deployment

For custom deployments (e.g., using scripts or bare metal):

1. Identify and stop all Kubernetes processes: 

2. sudo systemctl stop kubelet

3. Remove configuration and data: 

4. sudo rm -rf /var/lib/kubelet /var/lib/etcd /etc/kubernetes

5. Clean up networking (if using tools like Flannel or Calico): 

6. sudo ip link delete cni0

7. sudo ip link delete flannel.1

 

Post Shutdown

• Clean Resources: If the cluster managed cloud resources (e.g., load balancers, persistent volumes), ensure they're cleaned up.

• Backup First: Before shutting down, ensure you've backed up any critical data or configurations.