- From: Willy Tarreau <w@1wt.eu>
- Date: Tue, 19 Apr 2016 18:45:52 +0200
- To: Daniel Stenberg <daniel@haxx.se>
- Cc: Cory Benfield <cory@lukasa.co.uk>, Lucas Pardue <Lucas.Pardue@bbc.co.uk>, "ietf-http-wg@w3.org" <ietf-http-wg@w3.org>
On Tue, Apr 19, 2016 at 06:00:36PM +0200, Daniel Stenberg wrote: > On Tue, 19 Apr 2016, Cory Benfield wrote: > > >>Stefan and Daniel point out that the server uses the Upgrade header to > >>"advertise support" for h2. > > > >I don???t think that???s really a good way to read this section of RFC 7230. > > I'd like to point out that my argument in that report was when the client in > question originally took the Upgrade: response header as some sort of > instruction of what to do next, while I'm saying Upgrade: in a response > header is only an advertisement for support - not an instruction to do > anything. I said that disregarding the actual protocol it mentioned. Absolutely, "101 Switching Protocol" would be needed here for an action to take place. > I think Upgrade: should probably say h2c as a h2 in there begs the question > what that really means... Well, we knew that having two names for the same thing would cause trouble :-) Willy
Received on Tuesday, 19 April 2016 16:46:20 UTC