Re: [v6ops] How do you solve 3GPP issue if neither operator nor handset supports PD?

[[e-mail]]

Philip,

>>> I found that many implementations just don't react
>>> to an NS or send any themselves.
>> 
>> That is correct for any implementation I have written.
> 
> It seems to me that this is broken. Nowhere in RFC 4861 does it say that if
> you are on a point-to-point link you are free to drop an incoming NS.

Neither do any implementation I'm aware do that.
To be clear. If a link-type has no L2 address / is point to point, then no implementation I've touched does ND address resolution.
It would be happy to respond to a NUD message.

>> And my point was that this is the desired behaviour in this case.
>> At least we should explore the consequences of not treating the
>> nodes on each end as having a directly connected shared subnet
>> (apart from fe80::/10) and what that does for address assignment/pd.
> 
> ND is currently link agnostic. Having a different meaning for the L and A 
> bits depending on the link type strikes me as a bad idea.

And I'm not proposing that.

The current behaviour of 64share does that (and gets the hack label for that reason).
What I'm saying is that 64share (and what I propose in p2p ethernet) is a lot closer to PD than it is to address assignment.
As soon as you stop thinking that a p2p link has a shared subnet, that's where you end up.

Best regards,
Ole