From: Matt Smith (matt-n-donna@xxxxxxxxxxxxxx)
Date: Sun Nov 04 2001 - 22:22:07 GMT-3
Check out the OSPF to BGP redistribution on R2 I think (going from memory).
Keep in mind that EBGP learned routes are AD20 which would be less than
IGRP. You may want to take a look at BGP backdoor routes to fix this but
oddly enough the routing should function just fine in this scenario. Since
you are redistributing OSPF into BGP at R2 the Serial interface that R8
peers with will not be injected into BGP and therefore wont be learned by R8
via EBGP. Since the EBGP routes that are learned by R8 will all point to
R2s Serial (assumming you peered R8 to this interface) then when R8 does the
recersive lookup to see how to get to the BGP next hop it will still follow
the normal routing path. In this scenario the routing table looks odd but
functionally it seems to work fine. The biggest problem I had with this lab
was catching the virtual link authentication issues;-)
Hope this helps
Matt
----- Original Message -----
From: "A.Strobel" <anja10@usa.net>
To: <ccielab@groupstudy.com>
Sent: Thursday, November 01, 2001 3:52 PM
Subject: lab 19
> Anyone currently working on lab19, BGP-OSPF redistribution section?
> I have problem keeping R8's ip routing table stable.
> It flips between B(BGP) and I(IGRP).
> What cuases that?
> The tunnel is NOT the culprit.
>
> Ideas?
>
> A. Strobel
>
This archive was generated by hypermail 2.1.4 : Fri Jun 21 2002 - 06:45:03 GMT-3