In message <[email protected]>, Michael Richardson writes:
>
> {I would love a description from Apple and Microsoft of how their
> layer-2/layer-3 address discovery/configurartion systems already interact}
>
> Philip Homburg <[email protected]> wrote:
> > Does that include all edge cases and the user interface?
>
> > I.e. DHCPv4 has configured an address and by mistake a RA orders IPv4
> > to be
> > shutdown. Do you just shutdown IPv4? Does your code parse the output of
> > ifconfig to figure out if the interface was configured?
>
> No, and no.
>
> The order is not to shutdown v4. The suggestion is to stop DHCPv4, if it
> hasn't succeeded, and assume that there isn't any IPv4 for any downstream
> devices.
With the currently defined states yes you shut down everything for 1 and
3. You don't shutdown for 0 and 2. I think some here want a currently
undefined state, no dhcp but local connectivity.
I would switch 1 and 2 around so that you go from most (0) to least (3)
IPv4 support on the interface.
Currently it is:
0 full IPv4 connectivity
1 all IPv4 off on the interface
2 local (link/site not global) connectivity only on the interface
3 no IPv4 on the machine.
For 1 and 3 you remove all addressing on the interface in question.
For 2 you leave DHCP running. It would be wise to not have default
routes being announced by the DHCP server. Link-local may be in use
if there isn't a DHCP server.
I suspect that we may want a 5th state. No DHCP, local connectivity.
That is split 2 into with and without DHCP.
> Ideally the device will not announce an IPv4 default route if it hasn't got
> one itself. If the admin has manually configured IPv4, then there is no issue
> .
> What the ISP wants to prevent is barages of DHCPv4 broadcasts, and what the
> homenet situation wants is to avoid offering an IPv4 default route when it
> won't work.
>
> > In contrast a DHCPNOSERVICE message that just suspends sending
> > DHCPDISCOVER messages for a while has almost no impact. It gets
> > slightly trickier if
> > that also implies no link local, but not much.
>
> It's not that there isn't v4, it's that there isn't DHCPv4.
> I personally think that both should actually exist.
>
> --
> ] 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
> -= IPv6 IoT consulting =-
>
>
>
>
> --=-=-=
> Content-Type: application/pgp-signature
>
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.12 (GNU/Linux)
>
> iQCVAwUBU06AooqHRg3pndX9AQINOQP/VRDO9bg6WYo3GZfPFe8kcrpMQ/pgn5rx
> R3iPe0+1P/KlFFr2QSlizUjMCyn4Kuo+vweRd7G+FTgrj/+xI988uxuNZ41hjNxt
> IUuyj7e+2jBfqPgTGSY36lGGWZ+VZA36GDa4pihqz0Htybq15A3kTC5ereGvJ7Ke
> ODvLKf35i5k=
> =jnPY
> -----END PGP SIGNATURE-----
> --=-=-=--
>
>
> --===============1429064588242792841==
> Content-Type: text/plain; charset="us-ascii"
> MIME-Version: 1.0
> Content-Transfer-Encoding: 7bit
> Content-Disposition: inline
>
> _______________________________________________
> homenet mailing list
> [email protected]
> https://www.ietf.org/mailman/listinfo/homenet
>
> --===============1429064588242792841==--
>
--
Mark Andrews, ISC
1 Seymour St., Dundas Valley, NSW 2117, Australia
PHONE: +61 2 9871 4742 INTERNET: [email protected]
_______________________________________________
homenet mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/homenet
