Return-Path: Delivered-To: apmail-httpd-dev-archive@httpd.apache.org Received: (qmail 11822 invoked by uid 500); 9 Nov 2001 02:27:19 -0000 Mailing-List: contact dev-help@httpd.apache.org; run by ezmlm Precedence: bulk Reply-To: dev@httpd.apache.org list-help: list-unsubscribe: list-post: Delivered-To: mailing list dev@httpd.apache.org Received: (qmail 11805 invoked from network); 9 Nov 2001 02:27:19 -0000 Date: Thu, 8 Nov 2001 18:23:11 -0800 From: "Roy T. Fielding" To: dev@httpd.apache.org Subject: Re: Too-few data from client Message-ID: <20011108182311.A2197@waka.ebuilt.net> References: <3BEAE804.90A15C51@Golux.Com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <3BEAE804.90A15C51@Golux.Com> User-Agent: Mutt/1.3.20i X-AntiVirus: scanned for viruses by AMaViS 0.2.1-pre3 (http://amavis.org/) X-Spam-Rating: daedalus.apache.org 1.6.2 0/1000/N On Thu, Nov 08, 2001 at 03:16:04PM -0500, Rodent of Unusual Size wrote: > More experimenting with boundary conditions (1.3).. another one for > Roy. :-) > > If I send a request header such as > > --- > POST / HTTP/1.0 > Content-type: text/plain > Content-length: 1024 > > less than 1K of data > --- > > Apache 1.3 eventually responds by closing the connexion, having > sent no response at all. > > Is this appropriate? Or should we be returning a 408 Request Timeout > instead? We should be returning a 408, but doing so is difficult because the 1.3 timeouts are based on a signal handler which probably doesn't have the context necessary to know why the timeout occurred. ....Roy