This may be extremely basic HyperV cluster 101, but I need a quick double check on my understanding here...
We've got three identical hosts connected to a SAN, the plan being to create a cluster and doll out no more VMs than can comfortably fit on two hosts so that in case of all out failure of one of them, the VMs in question fail over to the third.
Then we realized that a cluster needs an AD Controller. Now we could add a simple physical server, but that would create a serious single point of failure would it not? IE, if it goes down the whole cluster is shot. I suppose we could add
TWO small physical servers, one primary controller and one secondary, but that would add considerable expense considering the austere nature of this layout.
Only alternative I see is to create the AD Controller as a VM on one of the hosts and a secondary controller as a VM on another, such that in an all out host failure, there is still a domain controller for the cluster to function.
Do I have that right?