Hi!

We've a 2 node DSS cluster with each 6 NICs:

1 Management 192.168.0.220
1 Aux + iSCSI vIP #1 (MPIO #1) 10.0.1.220 (vIP 10.0.11.230)
1 Aux + iSCSI vIP #2 (MPIO #2) 10.0.2.220 (vIP 10.0.12.230)
2 on Bond for Replication (10.0.10.220 / mirror IP)
(1 free at the moment)

The ping node is in the management subnet 192.168.0.0 (e.g. a switch or a DC ...).

If I try to use 10.0.1.220 for iSCSI failover + 192.168.0.x for ping node, we catch a broadcast error.

Is it neccessary to have the ping node in the same subnet as aux/failover connection? If so, is it a problem to use the management subnet for failover connections (performance, security, ...)??

I would like to strictly seperate LAN/managment from iSCSI/SAN/Replication/Failover-Traffic ...