[DNA-BOF] Re: Sending while DNA is in progress...

[Pekka Nikander <pekka.nikander@nomadiclab.com>]

Bernard Aboba wrote:
>> Even if the ND cache or transmission state is latent, packets could
>> still be sent for DNA purposes of course, but we'd have to examine
>> whether other traffic would be paused, sent or discarded (for example,
>> if it takes a long time to do DNA, then VoIP packets are better
>> dropped).
> 
> It has been pointed out to me that packets already in the queue aren't
> salvagable if the host has changed subnets.  Typically the choice is
> either to send them or discard them -- you can't go back and change the
> headers after DNA completes, since that might require resegmentation.

As a description of the situation today, I concur.  However, if we
add mobility management to the picture, it may be possible to change
the the source address of the packet without affecting the rest of
the packet, depending on the situation.  On the other hand, dispatching
packets with a wrong source address might make sense in the future,
e.g., with HIP, where the source address doesn't matter at all.

The problem here is that the link layer address of the router is not
known.  If it is known, then it may make sense to dispatch packets
even if they are malformed.

> If the host remains on the same subnet, the packets can and should be
> sent. If the host has changed subnets, then no packets in the queue can be
> delivered successfully.

What comes to the same subnet case, I concur.  What comes to the
changed subnet case, if the host has the new router's link layer
address, it sometimes may make sense to dispatch the packets, as
I argued above.

> As a result, continuing to send enqueued packets under the assumption
> that the subnet remains the same has no downside, though it might make
> sense to suspend the addition of packets to the send queue, pending completion
> of DNA, so that they could be sure to acquire the correct headers.
> Latency sensitive applications such as VOIP might be an exception to this.

Do we need to consider the excess traffic in the case of dispatching
packets that cannot be delivered succesfully?  If not, I think we agree.

--Pekka Nikander