W3C home > Mailing lists > Public > ietf-http-wg@w3.org > January to March 2013

Re: Maximum Frame Size

From: Patrick McManus <pmcmanus@mozilla.com>
Date: Mon, 4 Mar 2013 21:05:36 -0500
Message-ID: <CAOdDvNo1qx0dSYt4OzSb_p-X8-smNxoe1M59rKMR90AXDAH5uA@mail.gmail.com>
To: Martin Thomson <martin.thomson@gmail.com>
Cc: HTTP Working Group <ietf-http-wg@w3.org>
Its very tempting to just get rid of the path discovery signaling and
retry baggage and make the max frame size also the minimum supported
read frame size. When spdy had the max frame size at 31 bits that
wasn't doable - but at 16 bits I think it is. If the group thinks 64KB
is too big but 32KB is not, (for a minimum) then I'd rather reduce the
max frame size than introduce a whole bunch of mechanisms to deal with
pmtud-prime.


On Mon, Mar 4, 2013 at 6:20 PM, Martin Thomson <martin.thomson@gmail.com> wrote:
> From: https://github.com/http2/http2-spec/issues/28
>
> HTTP/2.0 requires that implementations support a minimum frame size of
> 8192 bytes. The draft does not specify how an implementation is
> expected to learn that its peer has limited frame sizes other than by
> trial and error. Using RST_STREAM causes the error to be discovered
> after the problem has been encountered.
>
> It's also not possible to use RST_STREAM to reject a too-large frame
> that is not bound to a specific stream.
>
> This could be indicated in the SETTINGS frame.
>
> --
>
> Any objections to adding a new setting for maximum frame size (minimum
> value of 8192 bytes).
>
Received on Tuesday, 5 March 2013 02:06:08 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 5 March 2013 02:06:11 GMT