From: mcaplan.cs@xxxxxxxxxxxxxxx
Date: Thu Apr 19 2001 - 10:09:21 GMT-3
Peter,
>From an optimum design point of view I think it better to us point to point
sub-interfaces giving a different subnet for each spoke (or to use ip
unnumbered).
>From a lab exercise point of view, using a multipoint environment
(multipoint subint or physical) with a single subnet I would say its best to
use frame map statements unless told otherwise. This is the simplest way and
allows you to fix everything at layer 2 as opposed to layer3. Lab1 of
ccbootcamp gives a good exercise utilising policy routing to avoid frame map
statements by routing IP packets that are destined for the single subnet F/R
cloud to the hub router.
Cheers
Mark
> ----------
> From: peter brown[SMTP:pita40@hotmail.com]
> Reply To: peter brown
> Sent: Donnerstag, 19. April 2001 14:43
> To: ccielab@groupstudy.com
> Subject: Frame Relay
>
> I have seen some Frame Relay solution that rely on routing protocol to
> provide connectivity between spokes partial mesh enviroment, example ospf.
> I
> thought that it is better to use map statements or interface dlci to
> establish spoke to spoke connectivity in a partial mesh enviroment.
>
> Which is a more acceptable solution for lab exam purposes. Both solution
> works though.
This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 10:29:50 GMT-3