Re: Working Group Last Call on "Multi-Topology (MT) Routing in OSPF" (draft-ietf-ospf-mt-03.txt) - Corrected

[Tony Przygienda <prz at XEBEO.COM>]

> > So, why computing different paths for different classes of service
> > be any different ?
> >  
> Hi Yakov,
>
> In addition to the differences between the two protocol mechanisms, 
> we've evolved
> technologically since RFC 1583. It is now fairly straight-forward to 
> support multiple
> topologies (protocol signaling, SPF computations, and logically 
> separate RIBs/FIBs).
> As for requirements, the example most often given is when a SP wants 
> to provide
> separate topologies for unicast and multicast services.
>
> Note that there is a similar proposal in the ISIS WG so I'd venture to 
> say that there
> is some agreement on the requirement - 
> draft-ietf-isis-wg-multi-topology-09.txt.


we could always just force diffserv to disavow the TOS bits or, even 
better, make network
operators deploy carefull 'precedence designation remarking nodes' and 
set metrics
for certain TOSes on certain links to infinity and then we would almost 
have the same

   good idea, NOT

   -- tony