Re: [DNA] No FastRA unless Landmark?(draft-pentland-dna-protocol-00)

[Sathya Narayanan <sathya@research.panasonic.com>]

Greg -

> 
> The question wasn't about completeRA, it was about FastRA.
Yes - I understand (understood) this, but I was thinking without a completeRA the hosts will not be able to really answer the 'Link change question' - please see below.

> Non-DNA hosts aren't going to benefit from the RA's DNA content,
> but they are going to benefit from a fast response.
I guess, here is the problem in my understanding - do the non-DNA hosts  really benefit from receiving fastRA that is NOT complete. What would you like to include in these RA messages? 

Including all PIOs may help - atleast if there was no link change, because the host will see a prefix it is using in one of the RA messages. If there was a link change, the non-DNA host will still have to wait for MAX_RA_DELAY_TIME before they decide that they have moved, which is still better than waiting for the multicast. This is a cost-benefit question, is it worth the extra messages to support non-DNA hosts? I don't have an opinion on this.

<snip> 
> (Now visiting the CompleteRA without Landmark solicitation issue).
<snip>
> 
> Why doesn't it know? because without the landmark you can't tell that it's DNA capable.
> 
> Fortunately, there are plenty of bits left in the RS header (32)
> which could be used to identify a DNA solicitation without divulging
> the last attached prefix of the host in a landmark option.

Yeah - we can do this. We just need to be aware that sending unicast-completeRA at fastRA rate from all routers for every received RS message makes the amplification effect more deadly.

> 
> Please be aware that I personally consider location privacy to be a 'big issue'.

I don't understand this concern. When a IPv6 hosts transmits IPv6 packets on the network wouldn't its IPv6 address become public immediately - or when it responds to NS message for that matter - what information are you trying to protect by not including a landmark option in the RS message. Please explain.

- Sathya