W3C home > Mailing lists > Public > www-amaya@w3.org > January to March 2001

Re: Amaya Crash with 100-continue

From: Jose Kahan <jose.kahan@w3.org>
Date: Thu, 29 Mar 2001 10:06:21 +0200
To: Aaron Swartz <aswartz@upclink.com>
Cc: www-amaya@w3.org
Message-ID: <20010329100621.A15864@tarawa.inrialpes.fr>
Aaron,

We need more info. Can you post the complete protocol exchange that
took place? 

Second question, why are you not using HTTP/1.1? :) Have you tried using
mod_put in apache or my own (now old) put.c program?

For every 100-continue, there's a timeout. If we don't get the reply,
we then send the body. This is done as such as we don't know beforehand
what's the version of the protocol and it's they way that the HTTP spec
precises to do it.

As for the crash, I can't reproduce it without more information from you.

-jose

On Wed, Mar 28, 2001 at 03:42:12PM -0600, Aaron Swartz wrote:
> If Amaya attempts to PUT to a server, it will send an HTTP/1.0 request with
> an "Expect: 100-continue" header. This is a 1.1 feature, and as the RFC
> states:
> 
> As the RFC you referenced states:
> 
>     An origin server [...] MUST NOT send a 100 (Continue) response if such a
>     request comes from an HTTP/1.0 (or earlier) client.
> 
> It seems odd that Amaya is expecting a response that according to the RFC I

Amaya is not an origin server, but an HTTP 1/1 client.

> am not allowed to send it. In addition, if I do not send the 100 Continue
> response (as my server does not implement it), Amaya (appropriately) sends
> the body of the message. However, it then crashes (on Win32).
> 
> I would appreciate if these two bugs (sending an inappropriate expect
> header, and crashing) could be fixed. They are serious barriers to PUT
> compatibility with Amaya.
> 
Received on Thursday, 29 March 2001 03:06:59 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 1 April 2014 11:01:35 UTC