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

Re: http/2 initial limits - i see flow control initial limits specified, but not stream limits

From: (wrong string) 陈智昌 <willchan@chromium.org>
Date: Fri, 3 May 2013 23:04:11 -0300
Message-ID: <CAA4WUYicCQJUduw4i0G8MVpado_-m6b=jxcT=3f50d0rsTLFQQ@mail.gmail.com>
To: Roberto Peon <grmocg@gmail.com>
Cc: Martin Thomson <martin.thomson@gmail.com>, James M Snell <jasnell@gmail.com>, HTTP Working Group <ietf-http-wg@w3.org>
It sounds like I misremembered the spec. Sorry folks :) Sounds like any
change here is blocked on issue #40.


On Fri, May 3, 2013 at 10:55 PM, Roberto Peon <grmocg@gmail.com> wrote:

> Yup, the language in the specs is the same as what is currently in the
> HTTP/2 spec, but people have been interpreting them differently (or at
> least I assume so).
>
>    - 4 - SETTINGS_MAX_CONCURRENT_STREAMS allows the sender to inform the
>    remote endpoint the maximum number of concurrent streams which it will
>    allow. By default there is no limit. For implementors it is recommended
>    that this value be no smaller than 100.
>
> is the thing in question, and most implementations that I know of default
> to 100, which makes sense given that "no limit" and "unlimited" are not
> always the same.
> -=R
>
>
> On Fri, May 3, 2013 at 3:46 PM, Martin Thomson <martin.thomson@gmail.com>wrote:
>
>> On 3 May 2013 14:22, Roberto Peon <grmocg@gmail.com> wrote:
>> > This does differ from SPDY,
>>
>> This doesn't differ from the SPDY we were given.
>>
>> > but (still) seems reasonable given the number of
>> > streams created in the initial RTT shouldn't be unbounded anyway.
>> > Assuming we have persisted settings (which is in doubt, I guess), this
>> would
>> > only be a problem for the first RTT in a session where we didn't have
>> the
>> > persisted setting.
>>
>> I think that this default is under contention still:
>> https://github.com/http2/http2-spec/issues/40
>>
>> We may ultimately determine that some value between 0 (what #40
>> proposes) and infinity (status quo) is more appropriate.
>>
>
>
Received on Saturday, 4 May 2013 02:04:38 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 1 March 2016 11:11:13 UTC