From owner-freebsd-net@FreeBSD.ORG Tue Sep 30 00:27:37 2003 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 6B2E816A4B3 for ; Tue, 30 Sep 2003 00:27:37 -0700 (PDT) Received: from smtp.uc3m.es (smtp03.uc3m.es [163.117.136.123]) by mx1.FreeBSD.org (Postfix) with ESMTP id C774444032 for ; Tue, 30 Sep 2003 00:27:20 -0700 (PDT) (envelope-from jrh@it.uc3m.es) Received: from smtp03.uc3m.es (localhost [127.0.0.1]) by smtp.uc3m.es (Postfix) with ESMTP id B781D43354; Tue, 30 Sep 2003 09:27:19 +0200 (CEST) Received: from cimborrio.it.uc3m.es (cimborrio.it.uc3m.es [163.117.139.95]) by smtp03.uc3m.es (Postfix) with ESMTP id 0686B2B674; Tue, 30 Sep 2003 09:27:19 +0200 (CEST) From: Juan Rodriguez Hervella Organization: UC3M To: JINMEI Tatuya / =?utf-8?q?=E7=A5=9E=E6=98=8E=E9=81=94=E5=93=89?= Date: Tue, 30 Sep 2003 09:27:17 +0200 User-Agent: KMail/1.5 References: <200309291859.09355.jrh@it.uc3m.es> In-Reply-To: MIME-Version: 1.0 Content-Type: Multipart/Mixed; boundary="Boundary-00=_WBTe/pGb0IuMFI3" Message-Id: <200309300927.18850.jrh@it.uc3m.es> cc: freebsd-net@freebsd.org cc: snap-users@kame.net Subject: Re: IPv6 autoconfiguration on a multihomed site 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: Tue, 30 Sep 2003 07:27:37 -0000 --Boundary-00=_WBTe/pGb0IuMFI3 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 8bit Content-Disposition: inline On Monday 29 September 2003 19:22, JINMEI Tatuya / 神明達哉 wrote: > >>>>> On Mon, 29 Sep 2003 18:59:08 +0200, > >>>>> Juan Rodriguez Hervella said: > > > > I've got two routers sending RAs on the same link, so the > > host is configured with two IPv6 prefix on the same NIC. > > > > Watching the routing table I only see one default IPv6 route. > > > > When I try to ping the global IPv6 address of one router, if > > I'm lucky and the router is the same router which I've got > > installed on the default route, it will answer. > > > > The problem is that the host doesn't seem to realize that > > it is on two links at the same time, so it always sends packets to the > > default router.... > > > > For example, if there's only one RA, the autoconfiguration process > > ends up with a link-route like this: > > > > 2001:720:410:1001::/64 link#2 UC > > fxp0 > > > > On the other hand, when there are 2 RAs on the same link, this > > route doesn't exist ! > > I don't quite understand the situation...could you depict the network > configuration with concrete prefixes/addresses, and describe the > concrete problem (if it's a problem) with exact output of some > commands (such as netstat, ifconfig, etc)? > Well, attached I depict the topology plus "ifconfig" and "netstat -rn" outputs. [ time to see the topology ] After seeing the topology (uh Im a bit asleep)... the situation is quite simple, 2 routers (site exit routers), giving access to the internet using different ISPs. There is only 1 link, so the host only have 1 NIC. The problem is that I can ping one of the routers, though both the host and the router belong to the same network. I guess the packets are always forwarded to the default router, and I think that's not the expected behaviour, right ? Cheers. PS: I really love FreeBSD :) -- JFRH --Boundary-00=_WBTe/pGb0IuMFI3 Content-Type: text/plain; charset="iso-8859-1"; name="topology.txt" Content-Transfer-Encoding: 7bit Content-Disposition: attachment; filename="topology.txt" (Im really keen on ASCI art) (of course, Im kidding) Host X X XXXXXXXXXXXXXXXXXXXXXXX <-- wired link X X Switch 3Com X X X X X X X X Router1 Router2 X X X X X X ISP1 ISP2 ISP1 delegated prefix: 2001:720:410:1000::/60 ISP2 delegated prefix: 2001:800:40:2470::/60 RA of Router1: 2001:720:410:100b::/64 RA of Router2: 2001:800:40:2471::/64 =>>>>> Output of "ifconfig" jrh@pepitogrillo:~$ ifconfig rl0: flags=8843 mtu 1500 inet 163.117.140.44 netmask 0xffffff00 broadcast 163.117.140.255 inet6 fe80::2c0:26ff:fea3:5df6%rl0 prefixlen 64 scopeid 0x1 ether 00:c0:26:a3:5d:f6 media: Ethernet autoselect (100baseTX ) status: active rl1: flags=8843 mtu 1500 inet6 fe80::2c0:26ff:fea3:7d19%rl1 prefixlen 64 scopeid 0x2 inet6 2001:800:40:2471:2c0:26ff:fea3:7d19 prefixlen 64 autoconf inet6 2001:720:410:100b:2c0:26ff:fea3:7d19 prefixlen 64 autoconf ether 00:c0:26:a3:7d:19 media: Ethernet autoselect (100baseTX) status: active [snipped] The NIC I'm talking about is rl1, I use rl0 (IPv4) for configuration purposes but I plan to remove IPv4 on the short term. =>>>>> Output of "netstat -rn": jrh@pepitogrillo:~$ netstat -rn Routing tables Internet: Destination Gateway Flags Refs Use Netif Expire default 163.117.140.2 UGSc 1 2 rl0 127.0.0.1 127.0.0.1 UH 0 220 lo0 163.117.140/24 link#1 UC 2 0 rl0 163.117.140.2 00:02:1b:f4:01:13 UHLW 2 0 rl0 1200 163.117.140.166 00:90:27:86:09:3d UHLW 0 377 rl0 1060 Internet6: Destination Gateway Flags Netif Expire default fe80::20b:5fff:fe77:dcc0%rl1 UGc rl1 ::1 ::1 UH lo0 2001:720:410:100b:2c0:26ff:fea3:7d19 00:c0:26:a3:7d:19 UHL lo0 2001:800:40:2471:2c0:26ff:fea3:7d19 00:c0:26:a3:7d:19 UHL lo0 fe80::%rl0/64 link#1 UC rl0 fe80::2c0:26ff:fea3:5df6%rl0 00:c0:26:a3:5d:f6 UHL lo0 fe80::%rl1/64 link#2 UC rl1 fe80::20b:5fff:fe77:dcc0%rl1 00:0b:5f:77:dc:c0 UHLW rl1 fe80::20b:5fff:fe77:de61%rl1 00:0b:5f:77:de:61 UHLW rl1 fe80::2c0:26ff:fea3:7d19%rl1 00:c0:26:a3:7d:19 UHL lo0 fe80::%lo0/64 fe80::1%lo0 Uc lo0 fe80::1%lo0 link#8 UHL lo0 ff01::%rl0/32 link#1 UC rl0 ff01::%rl1/32 link#2 UC rl1 ff01::%lo0/32 ::1 UC lo0 ff02::%rl0/32 link#1 UC rl0 ff02::%rl1/32 link#2 UC rl1 ff02::%lo0/32 ::1 UC lo0 jrh@pepitogrillo:~$ --Boundary-00=_WBTe/pGb0IuMFI3--