Hello, this is my first post on this site. I am working with Service Provider providing services across Europe. The customer wants MPLS services and wants his RIP metric to be transported transparently through the MPLS network.
We will install our CPE at the customer site and this CPE will rung BGP with our PE router in the MPLS network. To our CPE, the customer will connect his own CPE on which RIP would be running.
Customer suggests that rip has to redistribute within bgp (running on our CPE)in such a way that RIP metric are transported transparently.
Can some one please help me with this?
There are multiple solutions to run RIP over a MPLS network. If the customer is using a IP VPN service, customer routing protocols can be deployed over the VRF's.
Generic Route Encapsulation can be used to tunnel the RIP messages to the neighbor accross the MPLS cloud.
If the next-hop routers is no more than 2 hops away, unicast routing can be used to statically define the RIP neighbor.
* Please rate ALL posts.
Start with the simplest solution i.e. Tunneling
Implementing Tunnels - How to Implement Tunnels
Check the Allowing Unicast Updates for RIP section
Configuring Routing Information Protocol
* Please rate ALL posts.
Use the red bgp
When RIP routes are redistributed into BGP, the route metric is stored in the BGP MED value. When BGP routes are redistributed into RIP, and the transparent keyword used, the MED value is copied back as the RIP metric. Without the transparent keyword, the metric value specified is applied to all the routes.
for the technical implementation with "metric transparent" already mentioned make sure your BGP med value is not larger than 15 in any case. Otherwise the RIP metric will assume the route to be unreachable and not announce it from PE to CE.
One way of having larger MED values is to use OSPF or EIGRP on another site, which metric will be copied to BGP MED during redistribution.
If only RIP is used with only single homed sites then this should work properly.
Another word of caution: BGP-RIP redistribution in the VPN does not prevent any routing loops per se. So with dual homed sites you will encounter loops if not carefully filtered during redistribution.
"Dual homed" includes backup scenarios with dynamic routing through RIP in place or any backdoor links the customer might install even without your knowledge.
Let me explain the scenario with the diagram which is attached with this post. This is a case of MPLS VPN.
1. RIP is configured on Rtr-B and Rtr-D with metric set as 12.
2. RIP is configured on RTR-A and Rtr-C.
3. eBGP is configured between RtrA-PE1 and RtrC-PE2
4. On Rtrs A & C, RIP is redistributed into eBGP as " redistribute rip"
5. On Rtrs A & C, BGP is redistributed into RIP as "redistribute bgp metric transparent"
Now after end to end connectivity is up, I can see that Routes of Rtr-D learnt from RIP have metric 12 using the "sh ip route" command on Rtr-C.
The same observation is there on PE1 and PE2 using the command "sh ip route vrf RIP"
Now when I see the routes for Rtr-D on Rtr-A (connected to PE1), the metric is changed back to 0. And the same routes on Rtr-B have metric 1.
BGP stores the IGP metric value into MED. But again MED is passed between the AS but it will not be propagated into a 3rd AS.
So, while passing on the routes to Rtr-A, PE1 is replacing the MED metric of 12 with the default value of 0.
How to transmit this MED value transparently into AS 100????
Note: I have used as-override command at PE1 and PE2 to avoid loops creation.