Switches, Hubs, and Modems
cancel
Showing results for 
Search instead for 
Did you mean: 

Problems with distributed-trunks

mkont
Occasional Visitor

Problems with distributed-trunks

I'm having some issues with the following setup.

Two 5400 switches each with latest firmware (K.15.03.0007) connected via switch-interconnect for allowing servers to be connected redundant.

In this example the server (RHEL6) with bonding (802.3ad) eth8/eth9 and tagged vlans (vid 100 and 108) is connected to both switches on port b20 which is configured as trk55 with dt-lacp.
After "failing" switch0 by disabling port b20 only the ip on vlan 108 is reachable. Trying to fail switch1 both vlans are reachable on the server.
I noticed also, that in regular operation only one server interface is used for communication. I'm used to see traffic beeing splitted over all interfaces using all slave interfaces.
What's the mistake?

switch0:
switch-interconnect Trk1
trunk A23-A24,B1-B4 Trk1 LACP
trunk B20 Trk55 dt-lacp
spanning-tree
spanning-tree instance 1 vlan 100-200 301-400
spanning-tree instance 1 priority 0
spanning-tree instance 2 vlan 1-99 201-300
spanning-tree instance 2 priority 1
spanning-tree priority 0
primary-vlan 100
management-vlan 108

switch1:
switch-interconnect Trk1
trunk A23-A24,B1-B4 Trk1 LACP
trunk B20 Trk55 dt-lacp
spanning-tree
spanning-tree instance 1 vlan 100-200 301-400
spanning-tree instance 1 priority 1
spanning-tree instance 2 vlan 1-99 201-300
spanning-tree instance 2 priority 0
spanning-tree priority 1
primary-vlan 100
management-vlan 108

Thanks and best regards