Thursday, January 13, 2011

Administrative Distance Mini-Labs

Days ago a friend asked me some clarifications about administrative distance during a redistribution task.
I suggested him to take the time and test administrative distance in a simple topology, then finished to test myself too and to propose 3 mini labs to my readers.

Here is the topology: simply 4 routers with two loopbacks to redistribute as external.



Minilab 1: EIGRP and OSPF redistribution. How to obtain the optimal path to reach R4 Lo0?


Minilab 2: RIP and EIGRP redistribution. Same problem as previous, but a different configuration must be done.


Minilab 3: RIP and OSPF redistribution. Same story, again.


Ok, well, I don't provide any initial configuration here.
Try to figure out what can be the problem for each minilab, and one or more possible solutions.

This is my approach to redistribution problems: first think, make a simple diagram for redistribution, then configure and test.
In this way, you already know looking at the topology where and when there will be problems.

Problem sources during redistribution, in my opinion and experience:

-EIGRP external routes. (that damn AD 170...)
-OSPF nssa/stub areas. (have to think on it differently from the rest of ospf domain)
-filters on particular prefixes on one protocol of a single router (that prefix will be learned through other protocol on that router, loops are possible)
-OSPF transit capability (if disabled ad hoc, in a middle of one area with virtual-links, can cause loops)

Hoping that someone of my readers will try these minilabs and post his/her opinons about, I go back to my studies. (sorry but I don't have time now to write a detailed explaination... :-) )

Marco

1 comment:

Rofi said...

nothing like a good reminder for one of the most basic but important subjects :)