W3C home > Mailing lists > Public > ietf-http-wg@w3.org > October to December 2015

Re: http/2 and non-safe http methods

From: James M Snell <jasnell@gmail.com>
Date: Tue, 3 Nov 2015 16:36:10 -0800
Message-ID: <CABP7Rbc5QXw8Zm5ru+sOy1R=wQSqYQc5O3+v1WLhsOC8DcvoYg@mail.gmail.com>
To: Patrick McManus <mcmanus@ducksong.com>
Cc: "ietf-http-wg@w3.org" <ietf-http-wg@w3.org>
Ok. Will need to put together a few tests to verify if/how the
existing h2 clients are handling this case then.

Would I be correct in saying that, on the server side, a server
receiving intermingled safe and unsafe requests for the same resource
on the same connection likely ought to ensure that the handling of
those are either serialized or rejected? The best practice of using
Etags and conditionals should also be emphasized further.

On Tue, Nov 3, 2015 at 4:14 PM, Patrick McManus <mcmanus@ducksong.com> wrote:
> Hey James,
>
> In this respect h2 isn't a lot different than h1 with parallel connections
> (which doesn't have any specified rules either). At least in my experience,
> the application needs to figure out above http whether the two requests need
> to be serialized (and hold back, if so).
>
> On Wed, Nov 4, 2015 at 12:36 AM, James M Snell <jasnell@gmail.com> wrote:
>>
>> Quick non-trolling question: the HTTP/2 RFC currently deals with
>> replay of idempotent requests that likely have not yet been handled
>> but I note that the issue over multiplexing non-safe methods was never
>> addressed. Has a recommended best practice emerged among
>> implementations for ensuring that unsafe requests are not sent or
>> handled out of order? Or is it considered to simply be up to
>> individual implementations to figure out?
>>
>
Received on Wednesday, 4 November 2015 00:36:58 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 1 March 2016 11:11:40 UTC