Let's say you have 4 virtual machines each should have 2 interfaces. Each VM should have at least 1 interface that private, and reachable to only the other virtual machines. This could be a private bridge or if your hypervisor has a setting for isolated network use that. Here is an example of mine using home network.
eno1 - Management interface, has additional static IP aliases for cluster traffic
eno2 - Private interface for cluster traffic only
nas01
eno1 - 192.168.1.1/24
eno2 - 192.168.100.1/24
nas02
eno1 - 192.168.1.2/24
eno2 - 192.168.100.2/24
nas03
eno1 - 192.168.1.3/24
eno2 - 192.168.100.3/24
nas04
eno1 - 192.168.1.4/24
eno2 - 192.168.100.4/24
When you complete the first part of the clustering wizard if you select 192.168.100.1, 192.168.100.2, 192.168.100.3, 192.168.100.4 then eno2 is not going to show as an interface during the second part of the wizard where you will need to provide additional IP's that will be used to access the SMB shares.
You will need 4 additional address during 2nd part of the wizard to use for the SMB shares. Lets say eno1 is used for your management subnet you can provide 192.168.1.5/24, 192.168.1.6/24, 192.168.1.7/24, 192.168.1.8/24 assuming these are outside of your DHCP range.
There is also an open bug ticket where even a hostname like tn-kvm01-nodea can be seen as too long 15 characters and will prevent the wizard from completing. A shorter hostname like nas01 will work. A fix will be coming in TrueCommand 2.2.1 that uses the cluster name to generate the NETBIOS computer object for the fused AD join that all hosts in the cluster share, and ensures proper validation. Another issue to be aware of is that creating a dispersed volume in the 3rd part of the wizard may fail with 22.02.2.1 but creating a replicated volume should be worked fine. That issue will be fixed in 22.02.3. Hope this helps.