From owner-freebsd-net@FreeBSD.ORG Tue Dec 13 22:39:38 2011 Return-Path: Delivered-To: freebsd-net@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id D18591065670 for ; Tue, 13 Dec 2011 22:39:38 +0000 (UTC) (envelope-from mrossi@swin.edu.au) Received: from outbound.icp-qv1-irony-out5.iinet.net.au (outbound.icp-qv1-irony-out5.iinet.net.au [203.59.1.105]) by mx1.freebsd.org (Postfix) with ESMTP id 5BDC38FC13 for ; Tue, 13 Dec 2011 22:39:37 +0000 (UTC) X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: ApMBAJbM5058qjZb/2dsb2JhbAAMN64WAQEBBDhAARALDQsJFg8JAwIBAgFFBg0BBQIBAbUiiQWLYwSUcYVtjFA X-IronPort-AV: E=Sophos;i="4.71,348,1320595200"; d="scan'208";a="315027808" Received: from unknown (HELO [192.168.15.65]) ([124.170.54.91]) by outbound.icp-qv1-irony-out5.iinet.net.au with ESMTP/TLS/DHE-RSA-CAMELLIA256-SHA; 14 Dec 2011 06:11:41 +0800 Message-ID: <4EE7CDBE.1090605@swin.edu.au> Date: Wed, 14 Dec 2011 09:12:14 +1100 From: Mattia Rossi Organization: Swinburne University of Technology User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:8.0) Gecko/20111105 Thunderbird/8.0 MIME-Version: 1.0 To: Hajimu UMEMOTO References: <2CECE1B6-98B6-4219-BDD7-220F83CAEC36@gsoft.com.au> <4F9821A6-673B-4DE6-A543-5F37BDD3F9B7@gsoft.com.au> <58FFF22D-6578-447D-AAC0-9673057DAD84@gsoft.com.au> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: Daniel O'Connor , freebsd-net@freebsd.org Subject: Re: FreeBSD 8 as an IPv6 router X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list Reply-To: mrossi@swin.edu.au List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 13 Dec 2011 22:39:38 -0000 On 14/12/11 04:07, Hajimu UMEMOTO wrote: > Hi, > >>>>>> On Tue, 13 Dec 2011 20:11:45 +1030 >>>>>> "Daniel O'Connor" said: > > doconnor> On 13/12/2011, at 19:54, Hajimu UMEMOTO wrote: >> doconnor> Is there a way to tweak it to do the right thing? >> >> Perhaps, sla-len should be 8. > > doconnor> Ahh many thanks, that seems work work. > > You are welcome. > > doconnor> Such are the risk when you copy things off the internet :) > > Yes, the sla-len depends on the prefixlen of the delegated prefix, and > the length depends on your ISP. > The DHCPv6 server announces the prefixlen, and the dhcp6c can know it. > However, the dhcp6c doesn't assume that the prefixlen of the > prefix-interface is 64. So, you still need to specify an appropriate > sla-len value. It seems inconvenient to me. So, I applied the > attached patch to calculate the sla-len automatically with the > assumption that the prefix-interface is 64, personally. > Ok, this is something I always get a bit confused with. I understand that it's the right clean thing to set up a /64 on the interface which sends router advertisements, but I also would expect by nature, that whatever prefixlength you chose on the interface, rtadvd would simply grab the lowest /64 prefix out of the configured one to send router advertisements out. The idea there is, that you might use this router for multiple subnets, and have a single default route. Now of course to do that you'd need to configure rtadvd.conf, so I guess the whole thing missing here is a bit of documentation which says, that if you don't configure rtadvd via rtadvd.conf you're not allowed to be lazy and configure any prefix on the interface and expect rtadvd to do the right thing. It seems to me, that a lot of people (including me) would expect that, so maybe some info about that wouldn't be to bad. Just my 2c. Mat