Re: [DNA-BOF] Do we need a separate DAD Problem Statement? (was Re:About Charter, DAD, and SEND)

[Bernard Aboba <aboba@internaut.com>]

> I agree that here may be a problem.  However, I also think that we
> will need to change the TCP algorithms somewhat due to intermittend
> connectivity anyway.

This is not an intermittent connectivity problem -- the link is up, it is
just that the host might be using the wrong address.

One way to deal with this might be to clear the ARP/ND cache on a "linkup"
event.  That way, the linklocal packets would not be sent unless the
destination were onlink.  Packets destined to the default gateway would
not be sent unless the host were on the same subnet. However, this is not
"optimistic" in that it will only allow sending of traffic after
reachability can be demonstrated -- phase 1 of DNA.  If reachability
cannot be demonstrated, then there is little point in sending traffic --
the host knows it is on the wrong subnet and needs to obtain a CoA.