Looking for design assistance.
Typically, I use a 2 node SAN disk cluster running CentOS and DRBD for high availability. I will then connect the hypervisor(s) via iSCSI; sometimes KVM/QEMU, sometimes Hyper-V Failover Cluster.
I am trying to translate this to a fully Microsoft solution. In total I have 5 physical servers; 3 for the Failover Cluster, and 2 for the Disk Cluster. Each server has 4 NICs, but I don't want to confuse the issue right now with NIC Teaming or anything like that.
At first, I started making a "Disk Cluster" using the 2 servers running Windows Server 2019 and Scale-Out File Services (SOFS). Each server has 2 local disks; one for the OS, and the other (I was thinking) to be made available as storage for the Failover Cluster. I have read a few articles, and watched a few videos, but I think I am missing something. I cannot add disks to the SOFS, because there are none to add.
I do not have the budget for Windows Server 2019 Datacenter, so I cannot use Storage Spaces Direct.
Does anyone have any advise or articles to share for this kind of build? A lot of articles reference StarWind Virtual SAN; maybe that is the route I should take.
I hope I am explaining myself properly. Thanks for any assistance.