Re: [quicwg/base-drafts] Spin per peer (#1982)

[MikkelFJ <[e-mail]>]

mikkelfj commented on this pull request.



> -
-When the spin bit is disabled, endpoints SHOULD set the spin bit value to zero,
-regardless of the values received from their peer. Addendums or revisions to
+fraction of connections.
+
+The selection process SHOULD be designed such that
+on average the spin bit is disabled for at least one eighth of network paths.
+The random choice SHOULD be dependent
+on the source and destination addresses of the path,
+so that the spin bit is consistently enabled or
+disabled for repeated use of the same path.
+
+When the spin bit is disabled, endpoints SHOULD set the spin bit value to
+a constant value randomly chosen to be 0 or 1,
+regardless of the values received from their peer.  Alternatively, endpoints MAY
+change this value when changing connection ID.  Addendums or revisions to

> To do what? 
Leaking that the endpoint is not spinning

But if you don't care about the endpoint finding out, why try to hide it. Maybe masking the bit requires more coordination than worthwhile - I am not too keen on this either, but alternatives are not really a lot better. Setting random per packet before AEAD would make it random on wire and would not require endpoint coordination, but it is yet another thing to deal with in header protection (tracking PRNG state to set a bit per packet). Leaving it undefined and unmasked would make the pattern arbitrarily detectable by middleboxes - fingerprinting as @huitema says - this is bad.

> I'm trying to avoid a constant 0 or 1 signal when we've elsewhere tried not to do that.

That is mostly on wire, no? TP values have lots of 0 signals by default.

-- 
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/1982#discussion_r236257596