Re: [dhcwg] about v4configuration and co (1)

Hi Ole,

Inline

Cheers,
Ian

On 6 Aug 2013, at 15:28, Ole Troan <otroan@employees.org> wrote:

> Ian,
> 
>> [ian] In the introduction of the v4config draft, it says:
>> "Although softwire mechanisms are currently the only use-case for DHCP
>> based configuration of IPv4 parameters in IPv6 only networks, a suitable
>> approach must not be limited to only supporting softwire configuration."
>> 
>> Putting it over the IPv4 data path is going to make this a pretty softwire
>> dependant solution.
> 
> it would be dependent on a broadcast capable link-layer supporting IPv4,
> that isn't softwire specific as such.

[ian] Couple of points arising from this:

1, This also describes a basic Ethernet interface. The behaviour for an Ethernet based DHCP client will be different (regular port 68 communication, can't use shared IPv4 addresses). So, there needs to be DHCP client logic to work out the right behaviour for the particular properties of the link it's running on.
2, Are translation (4-6, such as MAP-T) broadcast capable (I don't know the answer to this one, I've really never considered putting broadcasts through a NAT)?

> 
> you aren't suggesting that there is a need to provision IPv4 on a node that has no way to send IPv4 packets?

[ian] Not at all. I believe the concept of separating data and configuration planes is well understood:-)

> 
> cheers,
> Ole
> _______________________________________________
> dhcwg mailing list
> dhcwg@ietf.org
> https://www.ietf.org/mailman/listinfo/dhcwg