From owner-freebsd-stable@FreeBSD.ORG Tue Mar 16 08:04:31 2004 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 903B616A4CE for ; Tue, 16 Mar 2004 08:04:31 -0800 (PST) Received: from whizzo.transsys.com (whizzo.TransSys.COM [144.202.42.10]) by mx1.FreeBSD.org (Postfix) with ESMTP id 1BD0243D1F for ; Tue, 16 Mar 2004 08:04:31 -0800 (PST) (envelope-from louie@whizzo.transsys.com) Received: from whizzo.transsys.com (#6@localhost [127.0.0.1]) by whizzo.transsys.com (8.12.10/8.12.10) with ESMTP id i2GG4U0R061086; Tue, 16 Mar 2004 11:04:30 -0500 (EST) (envelope-from louie@whizzo.transsys.com) Message-Id: <200403161604.i2GG4U0R061086@whizzo.transsys.com> X-Mailer: exmh version 2.6.3 04/04/2003 with nmh-1.0.4 To: Andy Smith Organization: Serendipity Scheduling & Management X-Image-URL: http://www.transsys.com/louie/images/louie-mail.jpg From: "Louis A. Mamakos" References: <255A839665EA24408EB27A6AAE15518EAC1B@europa.ad.hartbrothers.com> <200403161427.i2GERR0R058377@whizzo.transsys.com> <20040316145553.GB26332@lug.org.uk> In-reply-to: Your message of "Tue, 16 Mar 2004 14:55:53 GMT." <20040316145553.GB26332@lug.org.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Tue, 16 Mar 2004 11:04:30 -0500 Sender: louie@TransSys.COM cc: freebsd-stable@freebsd.org Subject: Re: ftp.perl.org strangeness X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 16 Mar 2004 16:04:31 -0000 > On Tue, Mar 16, 2004 at 09:27:27AM -0500, Louis A. Mamakos wrote: > > > > Wouldn't it be nice to be able to configure the default behavior of > > the resolver to not query AAAA records first, before looking for A > > records? Just because my host is capable of doing IPv6 doesn't mean > > I prefer using it over IPv4. Especially since it's almost certain to > > be a lower performance path traversing various tunnels, with smaller > > MTUs, etc. > > But since I am guessing almost everyone that has IPv6 also has > IPv4[1], wouldn't that mean that IPv6 hardly ever gets used/tested? No, it simply make the abilty to test IPv6 a voluntary, deliberate act, rather than one that's forced on everyone by default. There are some platforms I'd be happy to test IPv6 on, and would really like to, and others that have a more production role that I'll get to later. > If things had been as you suggest then this problem may never have > been noticed. Then if any organisation were ever to have a "flag > day" for IPv6 it would suddenly be incredibly painful as all apps in > use throughout that organisation suddenly get affected by problems > like this that have lain dormant. I was there for the NCP -> TCP/IP flag day on the ARPANET, so you don't have to explain this. And it actually wasn't a flag day, but a transition that was preceeded by concurrent operation. The same sort of thing happens with all technology introduction. > Isn't it better that little quirks like this are found out in > advance whenever anyone starts playing with IPv6? Sure. > If you don't want to be part of the testing of IPv6, you could just > not use IPv6, right..? Please don't presume to put words in my mouth. I want a tool and mechanism, not policy shoved down my throat. I'm asking for something like /etc/host.conf that allows me to control the degree to which I want to be a pioneer, and on what platforms. This will actually encourage more people to deploy IPv6 capabile boxes and experiment with them gently, rather than dragged kicking and screaming later on. > > [1] Yes I am aware that there are plenty of networks that only have > IPv6, but for the purposes of actual working production systems > I think they will be in a minority compared to those that also > have IPv4. By definition this thread is about public resources > that have to be available by both IPv4 *and* IPv6. And what some hosts do is export an IPv6-specific domain name as a means of testing. E.g., foo6.example.com along with foo.example.com. Louis Mamakos