03-31-2007 10:53 AM
Hi,
Could anyone show how to configure a Per-vrf traffic engineering in MPLS Backbone?
Thanks,
Mel Sebastian
04-02-2007 07:43 AM
Hi,
Dunno actually if the feature exists in IOS now. If you have a link about this please post it here.
Maybe you should try to play with different Loopback addresses. For instance create two Loopbacks on the tail-end, then setup two TE-tunnels from head-end. Be sure to advertise VRFs with different next-hop (pointing to Lo1 and Lo2 accordingly) via MP-BGP. But i'm not sure that there's no "hidden obstacles" with this scenario.
04-02-2007 03:14 PM
Mel,
One way to do it is
- Setup a tunnel (no autoroute announce) on the ingress PE
- configure a static route through that tunnel interface as follow:
ip route x.x.x.x 255.255.255.255 tu0
- And use this same address as the bgp next-hop on the egress PE using the following command:
ip vrf xxx
bgp next-hop loxx
This way only the traffic destined to this address will traverse this specific tunnel.
Hope this helps,
04-02-2007 07:36 PM
Hi,
I have plotted below a similar configuration to my customer setup for the per-vrf TE which you suggest.
PE2 is a central vrf and has extranet to customer 1 and 3 vrf?s. Since there will be a defined bgp nex-hop on the cust_1 vrf pointing to PE2, and while central vrf will also have bgp next-hop pointing to PE1, what will happen to the PE2 central vrf going to PE3 cust_3 vrf traffic?
==============================================
PE1 :
ip vrf cust_1
rd 1:1
route-target export 1:1
route-target import 1:1
bgp next-hop Loopback1
!
interface Loopback1
description NEXT_HOP VRF cust_1
ip address 192.168.1.211 255.255.255.255
!
ip route 192.168.1.221 255.255.255.255 Tunnel1
PE2 :
ip vrf central
rd 1:2
route-target export 1:2
route-target import 1:2
route-target export 1:1
route-target import 1:1
route-target export 1:3
route-target import 1:3
bgp next-hop Loopback1
!
interface Loopback1
description NEXT_HOP VRF cust_1
ip address 192.168.1.221 255.255.255.255
!
ip route 192.168.1.211 255.255.255.255 Tunnel1
!
!
PE3 :
ip vrf cust_3
rd 1:3
route-target export 1:3
route-target import 1:3
!
==============================================
Thanks,
Mel
04-03-2007 05:05 AM
Traffic from PE2 to PE3 and vice versa will continue to be routed through the LDP signaled LSP and therefore following the IGP route.
Hope this helps,
04-03-2007 06:39 AM
Hi,
Thanks for your help and time...
Mel
Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: