Ray Hunter <[email protected]> wrote: > ALL configuration for ALL homenet devices, including end nodes.
> Which is why I think we need to look at a layered set of solutions,
> with each layer almost certainly consisting of a number of protocols.
> There's no point in having a nice standardised routing protocol if the
> end nodes still can't resolve names.
> On the other hand, trying to shoe horn all of DHCPv6 and all it's
> future (vendor) extensions into a routing protocol clearly ain't going
> to work either.
so we have stateful configuration of IPv6.
Let me recap, stating the obvious, so that I know that I am on the same page
as others.
We have the M and O bits in the RA.
I think that we will typically have M=0 and O=1.
The client then asks for more information via DHCPv6, which goes to a
link-local multicast address. (I'm unclear when the site-scoped address is
used).
I don't think that *anything* can or *should* change here(*)
(What and how the client does DNS vs mDNS vs sdnsext is a higher-level issue
for the moment)
The question is: how does the non-ISP-connected router get the right
information to use to answer the DHCPv6 requests it sees and/or realize that
it shouldn't run full DHCPv6, but rather just be a DHCPv6 relay.
So I think that the set of information that the *routers* need to exchange is
really not that large, if we assume that the "DHCPv6 server is HERE" is one
of those pieces of information.
(*) THERE IS STILL an issue of what do hosts and routers do if there are
two ISP connected routers, and potentially two completely different
sources of DHCPv6 information. Even in the simplest two ISP case,
hosts can see both of these DHCPv6 servers directly.
--
] Never tell me the odds! | ipv6 mesh networks [
] Michael Richardson, Sandelman Software Works | network architect [
] [email protected] http://www.sandelman.ca/ | ruby on rails [
--
Michael Richardson <[email protected]>, Sandelman Software Works
pgpFxR_e0yUSQ.pgp
Description: PGP signature
_______________________________________________ homenet mailing list [email protected] https://www.ietf.org/mailman/listinfo/homenet
