Re: [quicwg/base-drafts] Should we allow ACK-only packets to be declared lost? (#3451)

[ianswett <[e-mail]>]

> I do not think that there is any burden on even the simplest implementation. If the simplest implementation detects a gap, it can send a packet with an ACK. If that gap was detected due to a receipt of an non-ack-eliciting packet, a PING frame could be included with the resulting ACK (to satisfy "An endpoint MUST NOT send a non-ack-eliciting packet in response to a non-ack-eliciting packet, even if there are packet gaps which precede the received packet."

Your solution very much goes against what I intended for that text, so I filed #3480 to clarify what I meant and add some more text about what can be expected of ACKs.  I think it makes sense to continue discussion on that issue, if there is further discussion, because this issue is about whether ACK-only packets can be declared lost, not about acknowledgement reliability.

In particular, acknowledgements are not guaranteed to be received.  Almost all packets get acknowledged, but a sender can never rely on all packets being acknowledged, even of ACK-eliciting packets.

-- 
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/quicwg/base-drafts/issues/3451#issuecomment-590360731