I came across this question of configuration because of fault alerts in my vCenter Operations manager. The alerts were saying that I had no network redundancy for one of my vSwitches. The vSwitch it was pointing out is my iSCSI vSwitch. The current config for the port groups and vswitch are as follows:
Port Group 1 - iSCSI1, vmk2, vmnic6 active vmnic2 unused, port group bond to iSCSI Software adapter
Port Group 2 - iSCSI2, vmk1, vmnic2 active vmnic6 unused, port group bond to iSCSI Software adapter
vSwitch- active adapters vmnic2 and vmnic6
My question is, I've seen a lot of discussion on here saying that you want to have your second vmnic set to unused but no real reason why. If both vmnics are bond to the same iSCSI initiator wouldn't it make sense to have one set to stand by in case of failure? Any explanation would be very much appreciated and also if there is good reason to keep the second vmnic in unused I know this might not be the right forum for this question but is there any way to add an exception in vCenter Operations manager or would I just have to surpress the fault, or cancel the alert? Thank you.
The reason for this is that if the NIC loses connectivity it should treated as a storage path failure not a network failure, PSA(Pluggable storage architecture) will handle this event and failover to an alternate path to stay connected. So differentiate the network failover and storage failover we use active and unused for iSCSI.
So basically PSA will understand ACTIVE/UNUSED NIC and when any active NIC goes down it will automatically select the alternate path. As you didn't have any standby NICs, you will receive "no network redundancy" warnings and you can safely ignore them.