> On Mar 26, 2018, at 7:10 AM, Christian Amsüss <[email protected]> wrote:
> 
> Hello Chris,
> 
> On Mon, Mar 26, 2018 at 06:33:25AM -0700, Chris Wood wrote:
>>> * On the topic of DTLS protected connections: [...] so that I can
>>> issue a retransmit at a later time using the same DTLS sequence
>>> number?
>> 
>> No. Such details are hidden beneath the API. Retransmission, for
>> example, would be something you specify on a per-message basis.
> 
> Would that mean that I can tell it to send this particular message with,
> say, "retransmissit roughly after 2s, exponentialy fallin off with
> factor 2 for up to 4 transmissions", and can call off retransmission of
> that message when I've decoded a CoAP ACK?

I don’t know if the API as specified permits such fine-grained control, though 
it certainly seems within reason. 

> 
> For that would work just as well.
> 
> I'd only need to be informed whether the implementation retransmits with
> the same or a DTLS new sequence number, and whether such a retransmit
> actually happened. (The feedback to the application about the
> retransmits is not only relevant to the encrypted case, but also to tune
> the retransmission for applications as [1]).

Makes sense. Thanks for bringing this use case to the group!

Best,
Chris
_______________________________________________
Taps mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/taps

Reply via email to