There would not be an IGP running between F and D or F and E. F only has a
neighbor statement to D to allow it to establish a peering relationship with
it. The neighbor 192.168.12.1 ebgp-multihop statement in F's BGP routing
process allows this to work. There would not be a 192.168.12.0 network in
F's routing table unless D has the statement 'network 192.168.12.0' in its
own BGP routing process allowing it to advertise that network to other As'.
----- Original Message -----
From: Bradley J. Wilson <[EMAIL PROTECTED]>
To: cisco <[EMAIL PROTECTED]>; Chris Williams
<[EMAIL PROTECTED]>; Neil Schneider <[EMAIL PROTECTED]>;
<[EMAIL PROTECTED]>; Jim Coyne <[EMAIL PROTECTED]>; Victor Alba
<[EMAIL PROTECTED]>; Jeff Assarian <[EMAIL PROTECTED]>; Phil Heller
<[EMAIL PROTECTED]>
Sent: Sunday, February 25, 2001 10:40 AM
Subject: EBGP multihop question
> Take a look at Halabi (First Edition) p. 300, and riddle me this regarding
> Figure 10-1:
>
> How does network 192.68.12.0 get into RTF's routing table? The EBGP
session
> needs to be established before RTF can accept any routes from RTD. But
how
> can the session be established before RTF knows how to route packets to
RTD?
>
> The only thing I can think of would be a static route. There really is no
> feasible way to run an IGP between RTF and that network. However, Halabi
> doesn't include a static route in his configuration.
>
> (And if someone wants to post which page this is in the Second Edition,
> please do. I'm thinking about buying it, but I'm too cheap at this point.
> ;-)
>
> BJ
>
>
>
>
> _________________________________
> FAQ, list archives, and subscription info:
http://www.groupstudy.com/list/cisco.html
> Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
>
_________________________________
FAQ, list archives, and subscription info: http://www.groupstudy.com/list/cisco.html
Report misconduct and Nondisclosure violations to [EMAIL PROTECTED]
