Hi Tim,

Tim Hartrick wrote:
Greg,


Backward compatability shouldn't really be a problem.
Hosts which are doing RFC2461 Router Discovery
will understand RAs with options or bits in them
indicating solicitation or completeness, but just not
be able to access the improved function.



If a host that understands the new "bit" arrives on a network that has
routers that don't speak the new "bit" but do send abbreviated router
advertisements, then the host will need to take some new unspecified
protocol action to force the routers to send unabbreviated router
advertisements.  Otherwise the host will not detect movement as quickly
as it could when arriving on a network with routers that speak the new
"bit".  I agree that the other combinations of old and new functionality
are not likely to have problems.

This is not a new function. Hosts already have to deal with abbreviated router advertisements if they comply with RFC-2461.

Section 6.2.3, Page 45:

"A router MAY choose not to include some or all options
when sending unsolicited Router Advertisements..."

The defined mechanism for discovering all options
is to send an RS.  The Router should (not must...?)
send all options in that case.

You're right that the behaviour for movement detection
is sub optimal though, in that a message has to be sent.

Of course I don't really care for using router advertisements for movement
detection, but if we are going to do it, getting in all the corners is
important.
>

I don't think it's that hard, but whether it goes into
RFC2461bis is another matter.



I know it isn't hard.  However, in my opinion, trying to roll it into
RFC2461bis is on the edge what should be going in what was advertised as
a "bug fix" update.

I think that the solution may not actually be a "complete" bit, if we can achieve what we want to do with another mechanism. The fact that there are multiple possible solutions and no clear winner is an indication that it shouldn't go directly into RFC-2461bis.

That said, I don't think it unimportant to consider.

Greg.


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

Reply via email to