I have a lab and I am doing full mesh ibgp from PE to PE. I want to use a route reflector instead. My current setup from one of my PEs looks like this:(see attached)
Can anyone tell me how to configure a RR t? I wanted to make the route reflector, router P3. Would this work. It would obviously run BGP, but what would the sub commands look like? in production provider networks, is the RR usually a P router as well?
on the RR clients, do I need the address-family vpnv4 stuff? Or only the neighbor commands?
On the RR itself, I need both the neighbor commands AND the vpnv4 commands??
harbor235
Remember, the combination of ipv4 prefix plus a RD is a unique VPNV4 route, so if you want to pass that information between peers and ensure that it makes into the correct VRF than yes it needs to be there on both. So under IPV4 you need typical eBGP sessions with CEs and under VPNV4 you have the iBGP peers that require route information and vrf information.
As far as the RR server, remember that you are centralizing your route peering by introducing RR's, also remember that RRs will not modify the next-hop so make sure your IGP has everything you need.
AS far as RR server configuration, certain commands are available from each sub BGP config mode. So the answer is that there is several config statements in each section typically.
for example
router bgp 1
bgp log-adjacency-changes
nei 1.1.1.1 remote-as 1
nei 1.1.1.1 updaqte-source loopback 0
address-family ipv4
redistribute rip
nei 1.1.1.1 activate
exit address family
address-family vpnv4
nei 1.1.1.1 activate
nei 1.1.1.1 send-community both
exit-address-family
So each area performs a function, look through the command structure, you will get it!
router bgp 209
neighbor 20.1.1.1 remote-as 209
neighbor 20.1.1.1 update-source Loopback0
neighbor 50.1.1.1 remote-as 209
neighbor 50.1.1.1 update-source Loopback0
no auto-summary
address-family vpnv4
neighbor 20.1.1.1 activate
neighbor 20.1.1.1 send-community both
neighbor 20.1.1.1 next-hop-self
neighbor 50.1.1.1 activate
neighbor 50.1.1.1 send-community both
neighbor 50.1.1.1 next-hop-self
exit-address-family