I have several servers with local SATA/SAS RAID arrays. I would like to run there virtual machines with some level of automated failover - sync VM to secondary node in 1 minute interval and automated start when primary node died at least.
I've already created failover cluster with fileshare witness on another machine. In Failover Manager I'm not able to add any Disk, says "No suitable disk". I tried to add virtual iSCSI disk via MS iSCSI Target but it didn't appear either. Hyper-V Replica Broker can't be enabled without shared stored.
How to achieve failover with current hardware on Hyper-V? If not is there any other virtualization platform for my needs? Based on my search with VMWare I would need HA + vSAN which is quite expensive, Google's Ganeti doesn't work well with Windows guests.
You have two options in order to create a Hyper-V Failover Cluster using your current hardware:
1) Windows Server 2016 with Storage Replica http://www.tech-coffee.net/storage-replica/ that assumes a manual failover in case one of the servers fails or Storage Spaces Direct http://windowsitpro.com/windows-server/what-storage-spaces-direct that need at least 4 hosts https://slog.starwindsoftware.com/microsoft-storage-spaces-direct-4-node-setup/ but provides you automated failover and so on.
2) 3rd party software like StarWind https://www.starwindsoftware.com/starwind-virtual-san
that can take your local disks and mirror them between hosts providing those highly available storage instances as shared storage for your Hyper-V cluster. In case you would like to create an SMB 3.0 based failover cluster using dedicated servers for storage you can use their free version https://www.starwindsoftware.com/starwind-virtual-san-free
I don't have a solution for you but I kinda felt this deserves at least some sort of answer:
To state the obvious, a machine is highly dependent on its storage. In a highly-available environment - where all VMs can run on any node - the nodes must all be able to see the storage for a VM otherwise they cannot be a host for that VM. So clustered shared volumes can only be created if all nodes can see the storage.
With two nodes and no shared storage, I think the best you can hope for - without going 3rd party - is to go without the failover cluster and use Hyper-v replica.
In 2016 we get Storage Replica and I think you can use it to create a stretch cluster with asynchronous replication.
I'd love to see some solutions here if they exist.