Hi,
Wondering if anyone has any info on configuring VPLS with Cisco but not using the Cisco style VPLS feature, material I can find on this all seems to be either old pre-dating Cisco adding BGP signaling or using old hardware with IOS. I ask as Cisco do support BGP or LDP discovery and signaling on some models. e.g. ASR1000 series see: https://www.cisco.com/c/en/us/td/docs/ios-xml/ios/mp_l2_vpns/configuration/xe-16/mp-l2-vpns-xe-16-book/vpls-autodiscovery-bgp-based.html#GUID-E197082D-0DDE-412B-893B-539765E1EDA2
Enables BGP signaling and discovery or LDP signaling and enters L2VPN VFI autodiscovery configuration mode.
Note
For the VPLS BGP Signaling feature use the autodiscovery bgp signaling bgp command.
Has anyone done this or has it not been tested? I have an ASR I am about to setup with a CCR to test this but wanted to see if anyone else has. If we introduce some Cisco hardware I would prefer not to have to re-do all our VPLS setup in our network.
as an update:
I have built a lab in CML2 using the Cisco CSR1000 as it runs the same IOS-XE as the ASR routers.
I have VPLS working using BGP advertisements and BGP signaling I am now going through more extensive testing to check traffic etc.
to get it working on the cisco side all I had to do was:
Cisco enable LDP
mpls label protocol ldp
mpls ldp advertise-labels interface GigabitEthernet2
!
...
interface GigabitEthernet2
mtu 1600
ip address 192.168.0.1 255.255.255.252
mpls ip
mpls label protocol ldp
Cisco side VPLS
l2vpn vfi context customer
vpn id 100 <--- give the connection an ID
autodiscovery bgp signaling bgp <--- enable BGP discovery and signaling (not LDP signaling or Cisco style VPLS)
ve id 1 <--- set the site ID this one is 1 while the MT was set to 3
rd 1:50 <--- set the route distinguisher if not defaults to <BGP AS>:<VPN ID> so 65530:100
route-target export 1:50 <--- set your import and export targets
route-target import 1:50
control-word include <--- include control word if set MT side (is by default)
BGP peering to a route reflector
note the address-family and the suppression of the LDP signaling (not sure the LDP suppression is required but added as per best practice)
router bgp 65530
bgp router-id 10.255.254.1
bgp log-neighbor-changes
neighbor 10.255.254.4 remote-as 65530
neighbor 10.255.254.4 update-source Loopback0
!
address-family ipv4
no neighbor 10.255.254.4 activate <--- doing IP routing via OSPF so not required in BGP
exit-address-family
!
address-family l2vpn vpls
neighbor 10.255.254.4 activate
neighbor 10.255.254.4 send-community extended
neighbor 10.255.254.4 suppress-signaling-protocol ldp <--- suppress LDP signaling
exit-address-family
Cisco setup a VRF for customer traffic and add interfaces etc.,
create a bridge-domain add the VPLS to it and add it to the VRF. Add local edge ports to the VRF for upstream routing
ip vrf customer
!
bridge-domain 100
member BD-VIF 100
member vfi customer
!
interface BD-VIF100
ip vrf forwarding customer
ip address 172.16.0.1 255.255.255.0
!
interface GigabitEthernet4
ip vrf forwarding customer
ip address 172.16.1.1 255.255.255.0
On the MT side it was basically enable MPLS/OSPF/BGP/BGP VPLS as per defaults/wiki example.