To demonstrate the restore scenario of etcd for an existing AKS Engine cluster, we first need to modify some Kubernetes objects to later prove that the backup restore has worked. Please note that all the commands shown in this section assume that you are running AKS Engine where an external etcd topology is used with etcd members running on the same machines that host Kubernetes master components. For other clusters, such as an on-premise kubeadm setup, the structure of the directories will be different.
First, let's introduce some changes to the cluster state. For example, if you have our Voting application running, delete the associated Deployment object using the following command:
kubectl delete deployment -n dev-helm voting-application
After a while, all the pods will be terminated—let's assume that this was our disaster...
