Re: [quicwg/base-drafts] Receiver's behavior on key update (#2791)

[Kazuho Oku <[e-mail]>]

@DavidSchinazi Thank you for the comments.

> Perhaps we should keep `and one set for protecting packets`

I've updated the text to clarify that only one send keys is needed at a time (it's a fact, not something we need to enforce or encourage), and have changed "MAY limit to two" to "SHOULD retain at least two receive keys." Technically this changes the RFC 2119 keyword, but I think that the intent of the text is better clarified with the change.

> and add details about what `the endpoint installs the updated keys` means - I think you mean the endpoint installs the updated receive keys and if the packet decrypts successfully also updates sending keys.

I've changed the text to "the endpoint installs the updated _receive_ keys." What needs to be done when the receive key is updated is clarified later on, in the paragraph that starts with "If the packet can be decrypted and authenticated using the updated key and IV...". I think that the text flows better now.

-- 
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/2791#issuecomment-502547320