Re: [quicwg/base-drafts] Conflicting error codes (#4087)

[Marten Seemann <[e-mail]>]

@huitema There's another condition where KEY_UPDATE_ERROR might occur: If you acknowledge a packet in key phase N+1 using key phase N. It's just a MAY in the specification though (I don't know why, this should be trivial to implement):
> An endpoint that receives an acknowledgement that is carried in a packet protected with old keys where any acknowledged packet was protected with newer keys MAY treat that as a connection error of type KEY_UPDATE_ERROR. This indicates that a peer has received and acknowledged a packet that initiates a key update, but has not updated keys in response.

@huitema 
> I can't wait to see the first error report in which some poor dude tries to transmit more than 50GB over a QUIC connection and hits a bug in the implementation of key update...

That's why we recently added a key update test to the interop runner. Unfortunately, it looks like a number of implementations (even those that already boasted about rolling out QUIC into production) treat key updates as some kind of optional feature, which it isn't. This is bad, as this now prevents others from rolling out QUIC implementations that initiate a key update early in the connection to exercise the mechanism.

-- 
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/4087#issuecomment-692399479