From: Brian Postma (brpostma@xxxxxxxxx)
Date: Fri Feb 02 2001 - 11:15:53 GMT-3
I agree that without "ip split-horizon" on S0 of R2, you will have a routing
loop. However I cannot find any evidence of it when I am running EIGRP. My
configs for R1,R2 & R5 closely mirror the provided solutions except I have not
started the BGP portion of the lab yet. My R1 routing tables only contains the
EIGRP routes from R2. R1's route to R5 E0 (11.1.0.0) is via ospf from R5. I
know there is a loop and administrative distance of EIGRP should make its route
the one that is in the table, but I do not see it.
So I converted to IGRP and the loop is instantly visible via "debug ip igrp" an
d
R1's routing table. What do I need to do to see this loop when running EIGRP?
thanks
Brian Postma
Rob Webber wrote:
> Router 2 is configured for EIGRP and is using Frame Relay without
> subinterfaces. Thus ip split-horizon is disabled by default. Thus everything
> r2 learns from R1 via EIGRP, it will turn right around and advertise back to
> R1. On a "hub" router you often need split-horizon disabled, but on a
> "spoke" router (like r2) you almost always want it enabled.
>
> In this case it could be very dangerous. If r1 redistributes OSPF into
> EIGRP, then advertises all of those EIGRP updates to r2, r2 will advertise
> them all back to r1. Since EIGRP has a lower distance than OSPF, r1 will
> prefer the EIGRP routes coming from r2 - even for networks connected to r5 -
> routing loop!
>
> Hope this helps, Rob.
>
> -----Original Message-----
> From: nobody@groupstudy.com [mailto:nobody@groupstudy.com]On Behalf Of
> Johnny Dedon
> Sent: Wednesday, January 31, 2001 8:02 PM
> To: Groupstudy
> Subject: CCIEbootcamp lab 2
>
> Can anyone help with an explanation as to why split horizon is an issue on
> router2? Its late and I'm sure I'm missing something...be gentle :)
>
> Johnny Dedon
> Senior Staff Consultant
> Exodus Professional Services
> johnny.dedon@exodus.net
> www.exodus.net
>
This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 10:28:34 GMT-3