From owner-freebsd-hubs@FreeBSD.ORG Wed Jul 16 07:18:59 2003 Return-Path: Delivered-To: freebsd-hubs@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 5E1FC37B401 for ; Wed, 16 Jul 2003 07:18:58 -0700 (PDT) Received: from coe.ufrj.br (roma.coe.ufrj.br [146.164.53.65]) by mx1.FreeBSD.org (Postfix) with ESMTP id 8861643F3F for ; Wed, 16 Jul 2003 07:18:57 -0700 (PDT) (envelope-from jonny@jonny.eng.br) Received: by coe.ufrj.br (Postfix, from userid 2000) id CF6C4D9034; Wed, 16 Jul 2003 11:18:55 -0300 (BRT) Date: Wed, 16 Jul 2003 11:18:55 -0300 To: "Daniel C. Sobral" Message-ID: <20030716141855.GB35857@roma.coe.ufrj.br> References: <20030712173332.GB14686@electra.cse.Buffalo.EDU> <3F14147F.2010706@jonny.eng.br> <20030715203333.GA25714@roma.coe.ufrj.br> <20030715215327.GB23859@electra.cse.Buffalo.EDU> <3F154C4A.5080807@tcoip.com.br> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <3F154C4A.5080807@tcoip.com.br> User-Agent: Mutt/1.4i X-Quote: What are you looking for in my mail headers ? X-Operating-System: FreeBSD X-URL: http://www.jonny.eng.br From: jonny@jonny.eng.br (Joao Carlos Mendes Luis) cc: jason andrade cc: freebsd-hubs@freebsd.org cc: Ken Smith Subject: Re: Mirror Site Requirements... X-BeenThere: freebsd-hubs@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: FreeBSD Distributions Hubs: mail sup ftp List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 16 Jul 2003 14:18:59 -0000 Somebody told me that Daniel C. Sobral said: > jason andrade wrote: > >it is *much* better for mirrors to maintain a base level of consistency as > >our research into mirror use here has shown that users are more likely to > >use mirrors when they have stability in the process. any loss of stability > >almost certainly leads to the majority defaulting back to their perceived > >master site: ftp.freebsd.org and it takes a while for them to switch back > >again (e.g build trust in the process again). Very good point! And considering the stupid^H^H^H^H^H^Haverage user, the "nearest server" routine would be handy, altough I confess I do not have a perfect idea of how to implement this. It must not be only a matter of RTT, but also of bandwidth and maybe link costs. > >so to address the point - if rules come about they are being developed in > >consultation with the community and should be consistently applied to all > >mirrors. > > OTOH, we'd better do a consultation to find out how many mirrors won't > be able to keep up with the standards we set upon, because if lose too > many mirrors, maybe it's better to rethink the standards. Risking to loose my mirror status, I have to agree here... BTW: Could we devise some means to REQUIRE subscription of mirror managers to some list, and remove mirrors who do not answer after some time? Jonny -- João Carlos Mendes Luís jonny@jonny.eng.br Networking Engineer jonny@coe.ufrj.br