Re: help with bgp

From: Daniel Valle (danielfrvalle@gmail.com)
Date: Sat Mar 29 2008 - 00:03:30 ART


Chris,

The requirement he said in his question is to not touch R3. So that woul'd
be an option...

On 3/29/08, Chris McGuire <cmcguire@firstdigital.com> wrote:
>
> The best way to satisfy your requirement would be with changing the Weight
> on R3 inbound from the external AS to a higher priority. This will cause
> R3
> and R3 alone to prefer the route out. Other routers in their AS will
> continue to use the Local preference assigned by R1.
>
> I understand the logic of EBGP (pref 20) over Ibgp (pref 200). However,
> BGP
> will not look at this until way down the line. It will first look at the
> Weight, then the local preference.... And actually 9th preference prefers
> EBGP over IBGP. So using the Weight would be optimal for your requirement.
>
>
> http://www.cisco.com/en/US/docs/ios/12_2/ip/configuration/guide/1cfbgp.html#
> wp1000898
>
> HTH
> -chris
>
>
> On 3/28/08 2:01 PM, "Athaide, Dwayne" <DAthaide@eprod.com> wrote:
>
> > (R1-R2-R3)
> >
> > Is there a way to stop the propagation of a modified local preference
> > value within an AS without making any configuration to the router that
> > needs to keep its default value. In this case R3. For example R1 and R3
> > are connected to the same external AS and learning routes from the AS.
> > R1 has modified the routes learnt via the external AS by using local
> > preference. R2 now gets those routes and passes it on to R3. Now R3
> > uses R1 as an exit point. Anyway to have R3 ignore the higher local
> > preference it learns on R3 and have it exit directly by not making any
> > changes on R3.
> >
> > Thanks
> >
> > _______________________________________________________________________
> > Subscription information may be found at:
> > http://www.groupstudy.com/list/CCIELab.html
>
> _______________________________________________________________________
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html



This archive was generated by hypermail 2.1.4 : Tue Apr 01 2008 - 07:53:54 ART