I am currently experimenting with GKE auto-provisioning in order to find out if this is a viable way to go for my team's use cases.
I noticed that whenever I deploy a new workload that the cluster which results in the creation of a new node-pool, the cluster and all its workloads become unavailable for about a minute.
Is this intended? If so, is there a workaround for this issue?
We are planning to deploy and remove workloads quite regularly, so node pools would have to be added quite often. Ideally they also would not be removed in between deployments (when 0 nodes are running).