Re: effect of deprecated site local addresses on router renumbering (rfc 2894)

[Tim Chown <[e-mail]>]

Hi Suraj,

May I ask if you are implementing this, or are you looking from a theoretical
perspective?

Tim

On Wed, Oct 05, 2005 at 08:24:55AM +0100, Elwyn Davies wrote:
> 
> 
> Suraj wrote:
> 
> >Hi All,
> >
> >RFC 2894 ' Router Renumbering for IPv6'describes the Renumbering of
> >Prefixes using RR commands to multicast addresses. (Site local OR Link
> >local).
> >
> >Since the site local addresses are now deprecated (RFC 3879), we can
> >assume that RR is now supported only for Link local addresses (unicast
> >and multicast).
> > 
> >
> No:  The deprecation only affects site-local unicast addresses.  
> Site-scope multicast is still available.
> 
> >What is the relevance of the 'S'(site specific) flag now in the command
> >message. Should the 'S' flag be evaluated even if the scope of
> >destination address is Link local (unicast or multicast)?
> > 
> >
> Yes:  The relevance is unchanged.  If a router is at a site border and 
> is configured with some interfaces (set A) associated with one site and 
> others (set B)  associated with other site(s), then a renumbering 
> message arriving on any interface in set A (whatever the destination 
> address in the base IPv6 header) with the S flag set will be applied 
> exclusively to the interfaces in set A - those in set B will be unaffected.
> 
> >Since RFC 2894 says that the 'S' flag should be ignored unless the
> >router treats interfaces as belonging to different "sites", in this case
> >should the RR command messages be limited only to the interfaces on that
> >link OR to all interfaces on the router?
> > 
> >
> Again the type of the command message destination address has no effect. 
> Combining the words in s1, s3.1 and s4.3, the intention is that any RR 
> command with the S flag clear applies to *all* interfaces apart from 
> those that might be ruled out because they are currently shut down 
> depending on the setting of the A flag - nothing is said about altering 
> the processing depending on the type of destination address.
> 
> Regards,
> Elwyn
> 
> >Thanks and Regards,
> >Suraj.
> >
> >
> >
> >--------------------------------------------------------------------
> >IETF IPv6 working group mailing list
> >[e-mail]
> >Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6
> >--------------------------------------------------------------------
> > 
> >
> 
> --------------------------------------------------------------------
> IETF IPv6 working group mailing list
> [e-mail]
> Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6
> --------------------------------------------------------------------

-- 
Tim/::1

--------------------------------------------------------------------
IETF IPv6 working group mailing list
[e-mail]
Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------