From: Jongsoo kim (bstrt2002@gmail.com)
Date: Mon Apr 04 2005 - 00:38:22 GMT-3
See below.
On Apr 3, 2005 10:34 AM, Dillon Yang <gzdillon@hotmail.com> wrote:
> Hi, Jongsoo:
>
> 1. The strategy of topology may be wrong.
> I think it is better to do it ONE time not step by step as you said.
>
> 2. The redistribution may be repetitious.
The "deny" ACL(way01) vs the "distance"(way02).
>
> for instance: (This is from the course of CCNP)
>
> R1-----------ospf-------------R2
> rip...........................rip
> R3-----------rip--------------R4----vlan4
> vlan3
>
> :way01
> If you forbid the rip routes passing from R1 to R2, then vlan3 can not access vlan4 when link "34" is broken, then the redundancy is useless.
###### just note my way 01 in my checklist is to block rip routes from
R1 to R3 or from R2 to R4 in order to prevent rip routes re-entering
via OSPF.
> :way02
> It is the best way for multipoint of redistribution while it keeps the redundancy. R2 knows the rip routes from R1 all are AD=121, so it choose R4 as next-hop normally. When link "34" is broken, R2 uses the AD=121 route for the connectivity between R3 and R4.
####### Yes this is the case.
> Maybe it is too complex to realize in the real lab, right? Maybe cisco do not have the redundance requirement. But I believe you must take it into consideration when the loop link exists.
##### Oh I have seen pretty compliacted multi-exit routing in one of
my attempt. It is absolutely necessary to handle multi-exiting mutual
redistribuion.
Regards
Jongsoo from RTP
This archive was generated by hypermail 2.1.4 : Tue May 03 2005 - 07:54:52 GMT-3