Delete Cluster

Delete the Kubernetes cluster and clean up your environment

Prepare to delete a self-managed workload cluster

NOTE: A self-managed workload cluster cannot delete itself. If your workload cluster is self-managed, you must create a bootstrap cluster and move the cluster lifecycle services to the bootstrap cluster before deleting the workload cluster.

If you did not make your workload cluster self-managed, as described in Make New Cluster Self-Managed, see Delete the workload cluster.

  1. Create a bootstrap cluster:

    The bootstrap cluster will host the Cluster API controllers that reconcile the cluster objects marked for deletion:

    NOTE: To avoid using the wrong kubeconfig, the following steps use explicit kubeconfig paths and contexts.

    dkp create bootstrap --kubeconfig $HOME/.kube/config
    
    INFO[2021-08-25T13:43:50-32:00] Creating bootstrap cluster                    src="bootstrap/bootstrap.go:143"
    INFO[2021-08-25T13:43:52-07:00] Initializing bootstrap controllers            src="bootstrap/controllers.go:96"
    INFO[2021-08-25T13:44:29-07:00] Created bootstrap controllers                 src="bootstrap/controllers.go:101"
    INFO[2021-08-25T13:44:29-07:00] Waiting for bootstrap controllers to be ready  src="bootstrap/controllers.go:104"
    INFO[2021-08-25T13:44:39-07:00] Bootstrap controllers are ready               src="bootstrap/controllers.go:109"
    INFO[2021-08-25T13:44:39-07:00] Patching ClusterRoleBinding for CAPPP         src="bootstrap/controllers.go:112"
    INFO[2021-08-25T13:44:39-07:00] Initializing Tigera operator                  src="bootstrap/clusterresourceset.go:37"
    INFO[2021-08-25T13:44:39-07:00] Created Tigera operator                       src="bootstrap/clusterresourceset.go:42"
    INFO[2021-08-25T13:44:40-07:00] Initializing AWS EBS CSI CustomResourceSet    src="bootstrap/clusterresourceset.go:109"
    INFO[2021-08-25T13:44:40-07:00] Created AWS EBS CSI CustomResourceSet         src="bootstrap/clusterresourceset.go:114"
    INFO[2021-08-25T13:44:40-07:00] Initializing Local Volume Provisioner CustomResourceSet  src="bootstrap/clusterresourceset.go:116"
    INFO[2021-08-25T13:44:40-07:00] Created Local Volume Provisioner CustomResourceSet  src="bootstrap/clusterresourceset.go:121"
    INFO[2021-08-25T13:44:40-07:00] Initializing Cluster Autoscaler CustomResourceSet  src="bootstrap/clusterresourceset.go:181"
    INFO[2021-08-25T13:44:40-07:00] Created Cluster Autoscaler CustomResourceSet  src="bootstrap/clusterresourceset.go:186"
    INFO[2021-08-25T13:44:40-07:00] Initializing Node Feature Discovery CustomResourceSet  src="bootstrap/clusterresourceset.go:239"
    INFO[2021-08-25T13:44:40-07:00] Created Node Feature Discovery CustomResourceSet  src="bootstrap/clusterresourceset.go:244"
    INFO[2021-08-25T13:44:40-07:00] Initializing NVIDIA GPU Feature Discovery CustomResourceSet  src="bootstrap/clusterresourceset.go:297"
    INFO[2021-08-25T13:44:40-07:00] Created NVIDIA GPU Feature Discovery CustomResourceSet  src="bootstrap/clusterresourceset.go:302"
    
  2. Move the Cluster API objects from the workload to the bootstrap cluster: The cluster lifecycle services on the bootstrap cluster are ready, but the workload cluster configuration is on the workload cluster. The move command moves the configuration, which takes the form of Cluster API Custom Resource objects, from the workload to the bootstrap cluster. This process is also called a Pivot.

    dkp move \
        --from-kubeconfig ${CLUSTER_NAME}.conf \
        --from-context konvoy-${CLUSTER_NAME}-admin@konvoy-${CLUSTER_NAME} \
        --to-kubeconfig $HOME/.kube/config \
        --to-context kind-konvoy-capi-bootstrapper
    
    INFO[2021-06-09T11:47:11-07:00] Running pivot command                         fromClusterKubeconfig=eks-example.conf fromClusterContext= src="move/move.go:83" toClusterKubeconfig=/home/clusteradmin/.kube/config toClusterContext=
    INFO[2021-06-09T11:47:36-07:00] Pivot operation complete.                     src="move/move.go:108"
    INFO[2021-06-09T11:47:36-07:00] You can now view resources in the moved cluster by using the --kubeconfig flag with kubectl. For example: kubectl --kubeconfig=/home/clusteradmin/.kube/config get nodes  src="move/move.go:155"
    
  3. Use the cluster lifecycle services on the workload cluster to check the workload cluster status:

    dkp describe cluster --kubeconfig $HOME/.kube/config -c ${CLUSTER_NAME}
    
    NAME                                                               READY  SEVERITY  REASON  SINCE  MESSAGE
    /eks-example                                                       True                     9m4s
    ├─ControlPlane - AWSManagedControlPlane/eks-example-control-plane  True                     9m4s
    └─Workers
      └─MachineDeployment/eks-example                                  True                     7m39s
        └─4 Machines...
    

    NOTE: After moving the cluster lifecycle services to the workload cluster, remember to use dkp with the workload cluster kubeconfig.

    Use dkp with the bootstrap cluster to delete the workload cluster.

  4. Wait for the cluster control-plane to be ready:

    kubectl --kubeconfig $HOME/.kube/config wait --for=condition=controlplaneready "clusters/${CLUSTER_NAME}" --timeout=60m
    
    cluster.cluster.x-k8s.io/eks-example condition met
    

Delete the workload cluster

  1. Make sure your AWS credentials are up to date. Refresh the credentials using this command:

    dkp update bootstrap credentials aws --kubeconfig $HOME/.kube/config
    
  2. Delete the Kubernetes cluster and wait a few minutes:

    Before deleting the cluster, dkp deletes all Services of type LoadBalancer on the cluster. Each Service is backed by an AWS Classic ELB. Deleting the Service deletes the ELB that backs it. To skip this step, use the flag --delete-kubernetes-resources=false.

    NOTE: Do not skip this step if the VPC is managed by DKP. When dkp deletes cluster, it deletes the VPC. If the VPC has any EKS Classic ELBs, EKS does not allow the VPC to be deleted, and dkp cannot delete the cluster.

    dkp delete cluster --cluster-name=${CLUSTER_NAME} --kubeconfig $HOME/.kube/config
    
    INFO[2021-06-09T11:53:42-07:00] Running cluster delete command                clusterName=eks-example managementClusterKubeconfig= namespace=default src="cluster/delete.go:95"
    INFO[2021-06-09T11:53:42-07:00] Waiting for cluster to be fully deleted       src="cluster/delete.go:123"
    INFO[2021-06-09T12:14:03-07:00] Deleted default/eks-example cluster  src="cluster/delete.go:129"
    

    After the workload cluster is deleted, delete the bootstrap cluster.

Delete the bootstrap cluster

dkp delete bootstrap --kubeconfig $HOME/.kube/config
INFO[2021-06-09T12:15:20-07:00] Deleting bootstrap cluster                    src="bootstrap/bootstrap.go:182"

Known Limitations

NOTE: Be aware of these limitations in the current release of Konvoy.

  • The Konvoy version used to create the workload cluster must match the Konvoy version used to delete the workload cluster.