This is an absolute stinker of a problem that I can't solve.

I have 2 Solaris 11 boxes, A and B. They have each 7 physical interfaces.....net0,1,2,3,4,5,8. net8 is a usb thingy and net5 is unknown. net2 is the backup interface that I use to connect to the machines remotely (ssh).

I've created an aggregate, aggr0 and within it I have created 2 vnics, prod0 (on the 10.129.10 vlan) and admin0 (on the 10.130.10 vlan) on both machines and given them IP addresses and added them to the aggregation, aggr0. The aggregation was created and the 4 interface net0, net1, net3 and net4 were added.


When all 4 netX's are in the aggregation each machine can only ping the other machines 10.130.10 address but not its 10.129.10 address. When I remove net4 from the aggregate both machines can ping each others 10.129.10 and 10.130.10 ip's.



When just net4 is in the aggregation neither machine can ping the aforementioned IP's.


All the cables are connected and the network guys say the ports on the switch are all up (I don't really know much about switches so can't give anything more than that)

Anyway it's baffling.

Also when I remove net0, net1 or net3 from the aggregation and execute a dladm show-link is shows those injterfaces as up but if net4 is removed it is shown as "unknown".



root@ggadmin1:~# dladm show-link
LINK CLASS MTU STATE OVER
net3 phys 1500 up --
net0 phys 1500 up --
net1 phys 1500 up --
net2 phys 1500 up --
net8 phys 1500 up --
net4 phys 1500 unknown --
net5 phys 1500 unknown --
aggr0 aggr 1500 up net1 net3
admin0 vnic 1500 up aggr0
prod0 vnic 1500 up aggr0




Could the type of device have a bearing on this:

root@ggadmin1:~# dladm show-phys
LINK MEDIA STATE SPEED DUPLEX DEVICE
net3 Ethernet up 1000 full i40e3
net0 Ethernet up 1000 full i40e0
net1 Ethernet up 1000 full i40e1
net2 Ethernet up 1000 full i40e2
net8 Ethernet up 10 full usbecm2
net4 Ethernet up 1000 full ixgbe0
net5 Ethernet unknown 0 unknown ixgbe1



Thanks in advance for any help or feedback.