>will require sending multiple packets to the server. This needs to
>be dealt with somehow.
>
>As such, I'm not convinced that sending these KE messages according
>to the NTP polling interval makes sense. I'm further concerned
>because I don't understand how the NTS KE will deal with lost
>packets.
We will add a paragraph treating the topic of lost packets.
Our intended approach is to just have the client go into a waiting state for each of its still unanswered requests.
This waiting state would either be ended successfully by a verifiable reply (what verifiable means depends on the context), or unsuccessfully by reaching a timeout value (which would be configurable in its length, but whose existence would be mandatory). In the latter case, either a re-transmission of the last message or a restart of the protocol would occur, depending on context
Does anyone here have criticism or other suggestions on this topic?
Kristof
_______________________________________________ TICTOC mailing list [email protected] https://www.ietf.org/mailman/listinfo/tictoc
