BGP Route Redistribution For MPLS
Why are we using the vrf table & not the vpnv4 table OR the global routing table?
I dont understand.
router bgp 100 , address-family ipv4 vrf 1, redistribute eigrp 1 , end
I dont understand.
router bgp 100 , address-family ipv4 vrf 1, redistribute eigrp 1 , end
Comments
-
Simrid Member Posts: 327Having a full update BGP routing table can contain thousands of routes and put a lot of strain on your routers resources.
Using VRF's have many benefits other than segregation of administrative configuration. VRF's allow you to logically separate traffic (between customers or departments for example). You may find that your company is trying to keep their route table as neat and efficient as possible thus creating VRF's.
You may find in the VRF configuration itself it is importing routes from the global routing table, learnt via EIGRP.Network Engineer | London, UK | Currently working on: CCIE Routing & Switching
sriddle.co.uk
uk.linkedin.com/in/simonriddle -
dppagc Member Posts: 293I know the purpose of VRF but how does the vpnv4 address family comes in?
Why arent routes imported into it? -
Simrid Member Posts: 327Here's an in depth explanation of the address-family command:
https://supportforums.cisco.com/discussion/11476526/when-use-bgp-address-familyNetwork Engineer | London, UK | Currently working on: CCIE Routing & Switching
sriddle.co.uk
uk.linkedin.com/in/simonriddle -
reload@ Member Posts: 44 ■■□□□□□□□□Let's say I'm a service provider with Customer A connected to my PE with a VRF I named Cust-A. He is advertising 1.1.1.1/32 and 192.168.1.0/24 to me via EIGRP.
I will need to advertise those routes out to my BGP peers under the address family vpnv4 so that Customer A's remote sites can learn about them. But first I must redistribute my EIGRP learned routes into BGP. These routes learned via EIGRP are IPv4 routes and they are inside the routing table of vrf Cust-A. So under my bgp configuration, I go to address-family ipv4 vrf Cust-A, and redistribute EIGRP. Now I can advertise those routes to my BGP peers as VPN IPv4 (vpnv4) addresses which is a route plus the route-distinguisher. Let's say I used the rd of 10.1.1.1:100 for vrf Cust-A. My VPN IPv4 addresses will look like this: 10.1.1.1:100:1.1.1.1/32 and 10.1.1.1:100:192.168.1.0/24. These are what my BGP peers will get.
To summarize, I get IPv4 routes (1.1.1.1/32 and 192.168.1.0/24) from Customer A via EIGRP. I redistribute those IPv4 routes inside Cust-A into BGP. Then I will advertise those routes as VPN IPv4 addresses (10.1.1.1:100:1.1.1.1/32 and 10.1.1.1:100:192.168.1.0/24) to my BGP peers configured under the vpnv4 family.
Hopefully this makes sense.