Re: [DNA] IMPORTANT: Combined DNA Solution draft

[Brett Pentland <brett.pentland@eng.monash.edu.au>]

Hi Christian,

Christian Vogt wrote:
>>>>[...]
>>>>
>>>>  - If the RS has a landmark and the router recognizes it
>>>>    a short RA is possible.
>>>>  - If the RS has a landmark but the router doesn't
>>>>    understand it, the landmark is ignored and a complete RA is sent
>>>>    that the host can use to make an immediate decision.
>>>>  - If the RS has no landmark, then the router sends a
>>>>    complete RA that the host can again make use of immediately.
>>>
>>>
>>>Its agreed that if complete RAs can be used then there's
>>>really no issue (other than perhaps loss of a little bit of
>>>bandwidth, but I wouldn't worry about that). However, in
>>>your second case above, what happens if the RS has a landmark,
>>>the router doesn't implement them, and it can't fit all the
>>>prefixes in the complete RA? It seems that the router will
>>>make a "random" decision of what prefixes to include, and
>>>since it doesn't recognize the landmark option, it may not
>>>include the host's prefix. Will this cause a problem?
>>
>>
>>I don't think so.  The choice of prefixes to include is not quite
>>random, because the router is required to include the "LinkID" prefix.
>>That may not be the prefix that the host picked as a landmark, but
>>it will have been received by the host in any earlier RAs if movement
>>has not taken place.  Note that I removed the word optional in the
>>suggested text for the previous comment because the the language in
>>section 6 does not make the LinkID inclusion optional.
> 
> 
> 
> I thought you agreed earlier during your discussion that Landmark and
> LinkID are optional for hosts only, but both mandatory for routers.

I think we agreed that LinkID was mandatory for routers.
> 
> If you change the draft in this regard, a router not understanding the
> Landmark option would consequently not support the DNA Protocol 3 at
> all.  But such a router would not support LinkID either, and the host
> would end up having to apply CPL logic anyway to arrive at a movement
> decision.
> 
> The host recognizes that it must use CPL because it sent a Landmark
> option with the RS, but the solicited RA neither includes a Landmark
> option nor has the C flag set.

Yes, but the response for a DNA router that does not (for reasons of its
own) wish to send a landmark response, *is* to send a complete RA, which
will allow the host to correctly make a decision.
> 
> This said, I don't see a reason why "case 2" (i.e., the RS has a
> landmark but the router doesn't understand it) need to be considered for
> this protocol at all.
> 
> Am I missing something?

The point was that not including a landmark response does not actaully
break anything.  We can make its support a MUST if that's what people
prefer, but I didn't make a MUST so far because I thought MUSTs were
only supposed to be included where failing to take the recommended
action would break the protocol.

Cheers,
Brett.