Re: [quicwg/base-drafts] recovery: clarification on TLP when sender is outpacing the timer (#1718)

[ianswett <[e-mail]>]

To address your first question, PR #1652 adds some more clarity to transport on the idle timeout and makes it clear that if you're sending into a void without receiving any ACKs, you should not keep extending your idle timeout.  The recovery doc doesn't describe anything relating to the idle timer, FWIW.

The intent is for the RTO timer to only go off after TLP.  There is text about setting the TLP to the min(TLP timeout, RTO timeout) in 4.3.2(https://tools.ietf.org/html/draft-ietf-quic-recovery-14#section-4.3.2).  If you want to discuss whether that should be what this document tries to specify, please open a separate issue about whether RTO should ever fire before any TLPs are sent?

Dead peer detection(#1638) is an entirely different problem, and there's no effort to describe that in the recovery draft at the moment.  It would be nice to talk about that somewhere, but I'm not sure it's known what best practices are?

-- 
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/1718#issuecomment-419145274