From owner-freebsd-net@FreeBSD.ORG Mon Jan 17 14:22:48 2005 Return-Path: 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 6E47016A4CE for ; Mon, 17 Jan 2005 14:22:48 +0000 (GMT) Received: from deepblue.titoon.net (deepblue.titoon.net [62.4.22.208]) by mx1.FreeBSD.org (Postfix) with ESMTP id 05EA943D53 for ; Mon, 17 Jan 2005 14:22:48 +0000 (GMT) (envelope-from julien@deepblue.titoon.net) Received: by deepblue.titoon.net (Postfix, from userid 1000) id A16301B620; Mon, 17 Jan 2005 15:22:46 +0100 (CET) Date: Mon, 17 Jan 2005 15:22:46 +0100 From: Julien Lesaint To: Chuck Swiger Message-ID: <20050117142246.GA18374@titoon.net> References: <20050116014706.GD28728@titoon.net> <41E9D59E.2040504@mac.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <41E9D59E.2040504@mac.com> X-PGP-Key: http://www2.titoon.net/pubkey.asc User-Agent: Mutt/1.5.6+20040907i cc: freebsd-net@freebsd.org Subject: Re: ttl-exceeded sourced by arrival interface ? X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 17 Jan 2005 14:22:48 -0000 On Sat, Jan 15, 2005 at 09:46:54PM -0500, Chuck Swiger wrote: > Julien Lesaint wrote: > >Quick reminder: in the case the route to the packet's source is not the > >interface this packet arrived on, do we have a way to source ICMP errors > >(ttl-exceeded) with the original interface's IP address ? > > Yes. Use IPFW's fwd mechanism, or even set a host route for the source IP > address pointing to whichever interface you please. I don't understand how this forward feature could help... Anyway I'm not using IPFW. Adding a route for each source IP is definitely not feasible. Maybe I wasn't clear enough so I repeat: I just want the FreeBSD box to send ICMP errors with the IP address of the interface the packet which generated this error, arrived on. Just like real routers... The "problem" only occurs in an assymetric routing context, i.e. the router which is generating the ICMP packet will not reply via the interface the original packet arrived on. > That's what the routing table is supposed to do, yes. This is the normal behavior in all situations but this one, yes... -- Julien Lesaint.