Re: HTTP/2 response completed before its request from Roberto Peon on 2014-07-01 (ietf-http-wg@w3.org from July to September 2014)

From: Roberto Peon <grmocg@gmail.com>
Date: Tue, 1 Jul 2014 14:15:06 -0700
To: Poul-Henning Kamp <phk@phk.freebsd.dk>
Cc: Jeff Pinner <jpinner@twitter.com>, Johnny Graettinger <jgraettinger@chromium.org>, William Chan (陈智昌) <willchan@chromium.org>, Martin Thomson <martin.thomson@gmail.com>, Patrick McManus <mcmanus@ducksong.com>, Jesse Wilson <jesse@swank.ca>, HTTP Working Group <ietf-http-wg@w3.org>
Message-ID: <CAP+FsNcL=pJXhi2nZZgVBY2M3gPaMGiVFBqRWmAS_=2Qd4KLRQ@mail.gmail.com>

Howso?

A server worried about the amount of memory it must use simply sets its
window size such that the client (assuming it was working properly) would
stop sending more data.
That is the reason for the flow control...

-=R

On Tue, Jul 1, 2014 at 2:12 PM, Poul-Henning Kamp <phk@phk.freebsd.dk>
wrote:

> In message <
> CAP+FsNdbgZLOikubd6OAoaih5uh6-JrjAMYmqYCV-A69tq8ogw@mail.gmail.com>,
> Roberto Peon writes:
>
> >An HTTP2 endpoint must always be attempting to read from the socket.
>
> ... but servers would make themselves vulnerable to a number of DoS
> attacks if they assume clients do so.
>
> --
> Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
> phk@FreeBSD.ORG         | TCP/IP since RFC 956
> FreeBSD committer       | BSD since 4.3-tahoe
> Never attribute to malice what can adequately be explained by incompetence.
>

Received on Tuesday, 1 July 2014 21:15:34 UTC