From owner-freebsd-net Mon Dec 2 2:23:24 2002 Delivered-To: freebsd-net@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 42C6E37B401 for ; Mon, 2 Dec 2002 02:23:23 -0800 (PST) Received: from itesec.hsc.fr (itesec.hsc.fr [192.70.106.33]) by mx1.FreeBSD.org (Postfix) with ESMTP id 4A40943EAF for ; Mon, 2 Dec 2002 02:23:22 -0800 (PST) (envelope-from yb@sainte-barbe.org) Received: from taz.hsc.fr (ogoun.hsc.fr [192.70.106.75]) (using TLSv1 with cipher EDH-RSA-DES-CBC3-SHA (168/168 bits)) (Client CN "taz.hsc.fr", Issuer "HSC CA" (verified OK)) by itesec.hsc.fr (Postfix) with ESMTP id 1B1BE20FE6 for ; Mon, 2 Dec 2002 11:23:21 +0100 (CET) Received: by taz.hsc.fr (Postfix, from userid 1000) id 24D9BFF; Mon, 2 Dec 2002 11:23:01 +0100 (CET) Date: Mon, 2 Dec 2002 11:23:01 +0100 From: Yann Berthier To: freebsd-net@FreeBSD.ORG Subject: Re: Multihoming - implementing RFC 1122 Message-ID: <20021202102301.GA63681@hsc.fr> Mail-Followup-To: freebsd-net@FreeBSD.ORG References: <200211282148.gASLmpas025733@sep.oldach.net> <3DEB30D9.A290D483@it.uc3m.es> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <3DEB30D9.A290D483@it.uc3m.es> X-Organization: Herve Schauer Consultants X-Web: http://www.hsc.fr/ X-Operating-System: FreeBSD 5.0-CURRENT User-Agent: Mutt/1.5.1i Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On Mon, 02 Dec 2002, Juan Francisco Rodriguez Hervella wrote: > Helge Oldach escribió: > > > > All, > > > > I wonder whether there are plans to complete implementation of the > > "strong ES" model as described in RFC 1122 for multihoming hosts on > > FreeBSD. Essentially this would assure that a multihomed host would > > send and receive IP packets through the "correct" interface (that is, > > the physical interface that is configured with the IP address used in > > the packets). > > > > I don't like the strong ES model. IMHO, with weak ES model we can > obtain the best of multihoming benefits. the strong ES model > makes use of source routing, which might forbid the communication > where it could be possible. > > I don't see why a host should discard a packet for him, only > because it has arrived at other interface. I see good reasons indeed. For example when I bind a daemon on a given IP, I have sometimes good reasons to have the multihomed host discard packets coming from the 'wrong' interface. As long as the model (strong vs weak) is tunable via a sysctl (as it is already with net.inet.ip.check_interface), I think that this is the perfect behavior: I can select whatever fits best my needs in a given situation. Cheers, - yann To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message