Hi,

We have 2 Nexus 5020 switches, a vPC is created and vPC keep-alive is in a good status. I have an iSCSI vlan 100 running on both the switches and Vlan 100 is allowed though the peer link. I have another VLAN for backup VLAN200 which is allowed also through the peer link. 

We have a couple of Win2012R2 clustered Hyper-V servers having 2 iSCSI 10GB and each of the NICs is connected to a Nexus switches. Servers are using iSCSI initiator to connect each iSCSI NIC to an iSCSI target IP, and MPIO is enabled on Win servers. The ports that connect to iSCSI on the switches are part of VLAN 100 but not a part of vPC, and they are shown as orphan ports in "Show vpc orphan"

Another set of servers have 2 NIC teamed using Windows 2012R2 "Switch independent" using vlan 200, but without using LACP at the switches. Also, those ports are shown as orphaned ports in "Show vPC orphan" on the Nexus switches. I did a lot of tests to disconnect one of the iSCSI ports at the servers, and the communication between the host and storage wasn't interrupted due to MPIO.

So, Instead of using vPC at the switch, I'm using MPIO of 2 iSCSI connection going to 2 different switches, and I'm teaming 2 NICs going to 2 different switches using win2012r2 "switch independent" teaming.

The above setup was up and running for a very long time.My question is what is the problem with my setup? What’s the advantage of adding iSCSI ports for every server to vPC at the switches? What's the advantage of  adding the ports used by the teamed NICS on Win2012R2 to vPC at the switches?

Thanks a lot!