Expert Redistribution 2

Posted by January 19, 2013 in Redistribution

Scenario:

You’ve been practicing your redistribution scenarios but you are not able to procure enough routers to make a big network. Instead, you only have (4) routers but want to make a scenario that is as complex as possible …

Goal:

  • Nothing has been preconfigured for you!
  • In order to keep the diagram clean, IP addresses and interfaces have been excluded.
  • Select whichever interfaces you like, but be sure they are all of the same speed and default bandwidth for EIGRP/OSPF metric calculation.
  • Use 192.168.xy.0/24 for each network where x and y are router numbers.
  • Example: 192.168.12.1 and 192.168.12.2 for the link between R1 and R2.
  • If an IP address is in a colored circle, that means it belongs to that protocol and should be advertised by that protocol.
  • No static, policy, or default routing allowed.
  • You may modify any metric, administrative distance, or route tags anywhere to solve this lab.

TASKS

  • Mutually redistribute between OSPF and RIP on R1.
  • Mutually redistribute between EIGRP and RIP on R2.
  • Mutually redistribute between EIGRP and OSPF on R3.
  • Mutually redistribute between OSPF and RIP on R4.
  • Mutually redistribute between EIGRP and RIP on R4.
  • Mutually redistribute between EIGRP and OSPF on R4.
  • Create a new loopback interface on R4: 4.4.4.4/32.
  • Redistribute this new loopback into EIGRP (don’t use the network command).
  • Redistribute this new loopback into OSPF (don’t use the network command).
  • Redistribute this new loopback into RIP (don’t use the network command).
  • To complete this lab, the route from any two loopback addresses must be one hop. Perform this test on every router with a traceroute sourced from the local loopback destined for the other (3) loopbacks.
  • Example: R1#traceroute 1.1.1.1 source loopback0 should only have one hop
  • Hints:
    • Routers will automatically rate-limit ICMP unreachables which makes traceroute slower. “no ip icmp rate-limit unreachable”.
    • You will need to use the word “redistribute” (15) times to solve this lab.
    • “debug ip routing” is your friend.

IOS:

c3640-jk9o3s-mz.124-16.bin

Topology:

redis

Opt In Image
Do you want your CCNA or CCNP Certificate?

The How to Master series helps you to understand complex topics like spanning-tree, VLANs, trunks, OSPF, EIGRP, BGP and more.

Written by René Molenaar - CCIE #41726

Master CCNA and CCNP Today!

You May Also Like

No Picture

RIP to EIGRP redistribution

No Picture

RIP to OSPF redistribution

About the Author: Nicholas Russo

24 Comments

  1. Tag all routes with routemaps (and re-tag them when redistributing them in)
    Do not allow routes with the tag of your IGP to come back.
    I would also suggest distance ospf external 180 just in case…

    That’s all

    1. Nope. There is more to it than that. Your solution will prevent loops but does not completely solve the task.

      You have to pay very close to metrics to avoid suboptimal routing. In one case, you MUST make a manual AD change for a single route as well. It’s tougher than you think!

    2. Hi Pavel.

      Can you please provide a password to download the file?

      Thanks.

  4. Can anyone post the solution for this? I have redistribution working well, but am dumbfounded as to how to fix the suboptimal routing issues. Everytime I change the AD and cost and get it working for 1 router, it causes suboptimal routing in the others. Kind of like when you get 1 side of a Rubik’s cube set and you try the 2nd side and mess up the first, and are right back where you started.

    1. That is the same problem I had towards the end. You’ll need to use a per-route AD-adjustment feature to make this work. I believe my issue was that R1 was preferring RIP (R2) to reach 3.3.3.3, when it should have preferred R3 via OSPF external.

  6. There is something confusing me, the task says to create a loopback 4.4.4.4 on R4 and redistribute it to all routing protocols in which R4 participate, but them says that we have to test trace to others loopback (3), it means that we have to create loopback on the other routers and do the same we did on R4?

    1. I wrote this lab a very long time ago. I believe the intent is to create a loopback per router (1.1.1.1, 2.2.2.2, etc) and ensure optimal routing to each based on the lab guidelines.

  7. Can anyone post the solution for this? or email me:::murshid.cse@gmail.com

    1. I will talk to Rene and see if we can get it uploaded. Perhaps he forgot to move it when the site changed.

  10. So far, it looks to me like you can’t do this without some level of policy based routing.

  13. For me the solution was too easy, I just play with metric, prefix-list, distribute list and changing administrative distance in some of them, but was too easy ! 😀 thanks ! i think the hard was the first redistribution expert, oh men, i spent 4 days thinking for a solution, but i finally did it and now thanks to you I know many things about redistribution, thanks again men 😀

    1. This was really a lab to test your knowledge of “redistribute connected”. Most people mess that up.

  14. To meet the objectives of this lab you have to do alot of filtering in addition to redistribution to meet the objectives. I personally used distance, prefix lists, distribute lists, offset-lists to achieve the desired result. Also make sure the links are of same speed and if they are still double check the OSPF cost as I have seen on One FastEthernet it was cost of 1 and on another it was 10…

  15. Very Very interesting, after 3 days and a lot of reading documentation about redistribution, i done this lab. Thanks a lot

Comments are closed.