From: Pavel Bykov (slidersv@gmail.com)
Date: Sun Sep 28 2008 - 18:46:23 ART
If every RR peers to all other iBGP neighbors and regards them as clients,
there is no need to peer them together for the purpose of redundancy. What
you should do, is assign them to the same cluster ID to reduce problems with
redundant routes - so both of them recognize that there is another RR in the
same "cluster".
On Sun, Sep 28, 2008 at 10:20 PM, Roger RPF <rpf@bluemail.ch> wrote:
> Hi group,
>
> Assume that we have an iBGP cloud with a couple of clients and two
> route-reflectors, connecting to each client. Is it correct that in order to
> get full redundancy for all "cases" the RR's should also peer together.
>
> I'm asking because I'm currently working on a LAB where the two RR's do not
> peer together and to my understanding (and after reading in "Odom"), I was
> thinking that the RR's in such a scenario should always have a peering
> session together...
>
> regards
>
> Roger
>
>
> Blogs and organic groups at http://www.ccie.net
>
> _______________________________________________________________________
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html
>
>
>
>
>
>
>
>
-- Pavel Bykov ------------------------------------------------- Stop the braindumps! http://www.stopbraindumps.com/Blogs and organic groups at http://www.ccie.net
This archive was generated by hypermail 2.1.4 : Sat Oct 04 2008 - 09:26:20 ART