I have been struggling with configuring the network adapters on a 2-node Server 2008 R2 failover cluster with CSV, iSCSI, and Hyper-V. The attached screen prints will hopefully give you an idea of how I have configured my adapters. My network logic is as follows:
Host - usually called "Management", to be used by host fo all of it's normal networking needs
Hyper-V - 2 physical adapters teamed (for redundancy) and then presented to the hypervisor
Heartbeat - I am using the old term, I think the new way of naming this is something like "Cluster" or "Cluster Communications", if I can figure out how to do it I would like this to be the only adapter the cluster uses for cluster specific traffic (Heartbeat, CSV, etc), I have read some articles about being able to limit CSV to a percentage of this adapters max usage to prevent CSV from drowning out the Heartbeat and causing an unexpected failover event
Live Migration - the only reason I separated this adapter on it's own network was because of the numerous articles online which recommended it
iSCSI - 3 physical adapters using MPIO
Does this seem like a sound network configuration? Any suggestions?
If this is a good configuration how about some advice on how to configure the adapter properties for each physical adapter?