Received: by taz.hyperreal.com (8.6.12/8.6.5) id UAA11837; Tue, 9 Jul 1996 20:52:22 -0700 Received: from paris.ics.uci.edu by taz.hyperreal.com (8.6.12/8.6.5) with SMTP id UAA11832; Tue, 9 Jul 1996 20:52:20 -0700 Received: from liege.ics.uci.edu by paris.ics.uci.edu id aa25160; 9 Jul 96 20:31 PDT To: new-httpd@hyperreal.com Subject: Re: Too soon for 1.2 commits ?? In-reply-to: Your message of "Tue, 09 Jul 1996 18:34:46 EDT." <199607092234.SAA30579@hershey.ai.mit.edu> Date: Tue, 09 Jul 1996 20:31:58 -0700 From: "Roy T. Fielding" Message-ID: <9607092031.aa25160@paris.ics.uci.edu> Sender: owner-new-httpd@apache.org Precedence: bulk Reply-To: new-httpd@hyperreal.com > I've got no objections to integrating the code with 1.2. FWIW, it may > be worth trying running with the code in for *all* systems in 1.2 > internal releases, to see if there are any systems which cause > problems. Yes, please do -- I couldn't see any obvious need for the NEED_LINGER once 1.1 is out, and I'd like to have that behavior on all systems in order to avoid some problems with persistent connections and pipelining. Perhaps we could set it up as #if defined(OUR_LINGER_FAILS) -- old code -- #else -- new code -- #endif and just add defines for OUR_LINGER_FAILS if it does for such a system. Either that, or write an autoconf test for it. ....Roy