Re: [dhcwg] DHCPv6 and IPv6ND

[Alexandre Petrescu <[e-mail]>]

Le 14/11/2017 à 08:38, Ted Lemon a écrit :
> On Nov 14, 2017, at 2:20 PM, Templin, Fred L <[e-mail] 
> <mailto:[e-mail]>> wrote:
>> I
>> think I have a fair question. How can we get the functionality of IPv6ND
>> RA and DHCPv6 PD in a single request/response message exchange?
> 
> Right now, you can't.   The question is, why is this a problem?
> 
> We've been around on this a lot of times.   I don't see any reason to do 
> it again.   But if you want to do it, you need to say why it's a problem 
> that you have to do router discovery, why it's better to configure the 
> DHCP server with router topology information, how you're going to get 
> the right router link-local address to send, and why all of this 
> complexity is somehow a better solution than router advertisements.

Ted,

The combined use of DHCPv6 and SLAAC is an unsolved problem in cellular 
networks.  Add to that the implications of the use of an undocumented 
link-layer setup protocol ("PDP") in the address formation.

A second aspect is that for full configuration one must use both 
DHCPv6-PD and SLAAC.

If we agree to those two points then it's easy to see the consequences 
which go along the lines that Fred proposes.

But if people dont agree DHCPv6 is needed, then it's a waste of time to 
explain.

One could as well go to 3GPP and tell them to offer not only address 
setup in PDP but also prefix delegation.  That may be less time to spend 
than do it at IETF.

Alex

> 
> It's not good enough to just say "I'd really rather use DHCPv6 for this, 
> can't we specify that?"
> 
> 
> 
> _______________________________________________
> dhcwg mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/dhcwg
>