From: Jim Jagielski <jim@jaguNET.com>
Message-Id: <199702071142.GAA26497@shado.jaguNET.com>
Subject: Re: Apache 1.2b7-dev performance
To: new-httpd@hyperreal.com
Date: Fri, 7 Feb 1997 06:42:17 -0500 (EST)
In-Reply-To: <Pine.LNX.3.95dg2.970206221352.16924C-100000@twinlark.arctic.org>
 from "Dean Gaudet" at Feb 6, 97 10:15:03 pm
Content-Type: text
Sender: new-httpd-owner@apache.org
Precedence: bulk
Reply-To: new-httpd@hyperreal.com

Dean Gaudet wrote:
> 
> Do we have a list of OSs that properly support SO_LINGER?  I'm trying to
> figure out how I could test that... 'cause I have to run with
> -DNO_LINGCLOSE on hotwired's IRIX 5.3 servers to avoid FIN_WAIT_2 death. 
> I haven't tried SO_LINGER because I'm not sure how to discern between good
> and bad SO_LINGER. 
> 

The rub is that the timer in many implementations of SO_LINGER
simply doesn't exist. So even if you set a timeout of 10 minutes
or so, this value is ignored and the process will keep trying to
send the data forever. Most BSD 4.3-based stacks have this
failing... SysV4 supposedly has the timer fixed, but just a
plain old bogus SO_LINGER setup.

I'm guessing the IRIX 5.3 has a non-functioning SO_LINGER
timer... it's of that time frame. Also sounds like it doesn't
have a FIN_WAIT_2 timeout either :/
-- 
====================================================================
      Jim Jagielski            |       jaguNET Access Services
     jim@jaguNET.com           |       http://www.jaguNET.com/
                  "Not the Craw... the CRAW!"