3.5 - BGP Reachability Alternative

I know this probably violates " Do not change or add any IP addresses...", but I'd like to know if this would be an alternate solution for 3.5


When R4 link to R3 goes down, R3 can still reach R4 loopback, but R5 (intermediate Router) drops/blackholes the traffic.

What I did was create a tunnel interface with a new subnet (150.10.3.x) - src/dst IP to respective loopbacks - advertised new subnet into OSPF. Pings from R3 to routes advertised from AS 54 to Router 4 are successful.

Does anyone see any issues/caveats with this (besides that I violated Rule #1)?




  • Without having the lab in front of me - I can say that it sounds like you did come up with a creative solution. But, as you pointed out, it is going to be typical that you cannot add new address space. 


    Cisco realizes that you can fix just about anything with static routes, Policy Based Routing, and/or the use of tunnels. That is why they typically remove these as options for solving a problem. 


    I always try and find solutions that do not involve tunnels if at all possible to ensure I do not violate any rules and also to keep things as simple as possible. I have found in the lab when the solution requires a tunnel, it is pretty obvious. 

  • I went with a method similiar to SG with some basic modifications, but I think I would have approached the protocor.  I used the route tags from the BGP AS to pull in my routes, but I would be curious to get some feedback on how this task affects 3.3.  The whole reason was to reduce the routing table, but now the same routes exist via IGP. 

    I assume based on the SG's approach that a protocor would just view the BGP table and see the single default and call it a day.  I guess I need to take a minute and learn how to properly grade my own labs.  Isn't there a link to that somewhere on the site?



Sign In or Register to comment.