I have stuck with this already few days trying to figure out how to properly set tagged and untagged vlans on RB1100Ahx2.
the principle diagram is here as an example: (pink is for testing, both cable ar not connected at the same cable)
Cisco 1/0/3 UNtagged is connected to ether1 - port doesn’t get up running, stays orange on cisco
re-plugging the same cable to:
Cisco 1/0/4 tagged —> ether5 - ports works and everything is ok.
in production I have in place of Cisco an unmanaged HP switch, connecting RB1100Ahx2 hangs up all network…
interface range GigabitEthernet1/0/1-3
switchport access vlan 2
switchport mode access
!
interface GigabitEthernet1/0/4
switchport trunk allowed vlan 2,10
switchport trunk native vlan 999
switchport mode trunk
!
----------------------------------------------------------------------------------------------
# jan/03/1970 01:48:04 by RouterOS 6.45.9
# model = 1100AHx2
/interface bridge
add name=bridge1
/interface vlan
add interface=bridge1 name=vlan2 vlan-id=2
/interface ethernet switch port
# ether1,ether2,ether3,ether4,ether5,switch2-cpu
set 0 default-vlan-id=2 vlan-header=always-strip vlan-mode=fallback # error here?
set 1 vlan-mode=secure
set 2 vlan-mode=secure
set 3 vlan-mode=secure
set 4 vlan-mode=secure
set 11 vlan-mode=secure
/interface bridge port
add bridge=bridge1 interface=ether1
add bridge=bridge1 interface=ether2
add bridge=bridge1 interface=ether3
add bridge=bridge1 interface=ether4
add bridge=bridge1 interface=ether5
/interface ethernet switch vlan
add independent-learning=yes ports=switch2-cpu,ether1,ether2,ether3,ether4,ether5 switch=switch2 vlan-id=2 # error here?
/ip address
add address=xxxxxxx interface=vlan2
/system package update
set channel=long-term
/system routerboard settings
set auto-upgrade=yes
Note: QCA8337 and Atheros8327 switch chips ignore the vlan-header property and uses the default-vlan-id property to determine which ports are access ports. The vlan-header is set to leave-as-is and cannot be changed while the default-vlan-id property should only be used on access ports to tag all ingress traffic.
I’m assuming error is in one of those of two rows setting vlans, but have tried different settings, either doesn’t work IP network from Mikrotik, either cisco blocks port
You don't need a default-vlan-id on a trunk port, only on access port. And switch2 is usually ether6-10 (it is on an RB2011 anyway, which I do have), whereas ether1-5 is switch1.
on RB1100 it’s opposite, switch2 is 1-5, switch1 is 6-10…
ether1 is UNtagged port
ether5 is tagged
so unfortunately your provided code won’t help. That’s the main problem, that everything work with tagged port, but can’t get it working with the untagged port
On my RB951G, which features (single) AR8327, the following sets ether2 as hybrid port (trunk with native) … VLAN 40 is untagged (native) and VLAN 3999 is tagged:
i think that one works for hybrid port fine, haven’t tested it though, but as the tagged port is working fine, it might also work fine with hybrid port.
but I think there is kind of issue how Mikrotik ROS or precisely this switch works with VLAN’s.
I checked with RB850 (ether1, ether2 put together with bridge, no vlans, no tagging etc., just plain bridge) Ether1 comes from unmanaged HP, ether2 goes to RB1100. When I connect RB1100 , then they see each other, but traffic from uplink HP on ether1 not work any more, although it doesn’t hang up the HP switch and other hosts
will try it tomorrow with Cisco, to see the behaviour.
Otherwise will have to make network setup in different way, using other MT or somthn…
Any log on Cisco to see why it deems port/connection faulty?
One thing, just to make sure: when you try the untagged connection between cisco and RB, you have to disconnect the trunk connection first. Or else there will be loop on VLAN 2 and without MTSP cisco would panick and disable port …
As there are VLANs in play, it should be MSTP rather than RSTP.
And it doesn’t matter how many switches are in chain, it only depends on interconnection … if there’s a loop in network infrastructure, switches around that loop need some kind of xSTP running (the same variant all of them) to avoid connectikn loops.
