Re: [quicwg/base-drafts] Request to Retire Locally Issued CIDs (#2769)

[Nick Banks <[e-mail]>]

Thanks @ianswett @martinthomson @kazuho for your feedback. I have made some changes, making the Retire Prior To field required, so we only have one frame again, added more 2119 language, cleaned up the text on ignoring reductions in the Retire Prior To field, and made it a protocol violation to have Retire Prior To greater than or equal to Sequence Number.

@ianswett on your request to require the difference between the Retire Prior To field and Sequence Number field to be no greater than active_connection_id_limit, that would mean the sender has to keep track of both what's the highest retired sequence number retired by the peer and what it wants locally, and the max of those two values in this field. I'm not sure the extra complexity for keep track of that is worth the cost. What benefit do you see in that requirement?

@martinthomson on the topic of the handshake and preferred address CIDs, I thought we already had text that implicitly assigned them sequence numbers. Is that not the case? What text beyond that would you like me to add?

-- 
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/pull/2769#issuecomment-499109697