Subject: Re: IPv6 routing with non-ppp interfaces
To: None <tech-net@NetBSD.org>
From: Richard Braun <syn@sceen.net>
List: tech-net
Date: 09/28/2004 17:15:13
--ZGiS0Q5IWpPtfppv
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Tue, Sep 28, 2004 at 12:03:21PM +0200, Konstantin KABASSANOV wrote:
> > -----Original Message-----
> > From: tech-net-owner@NetBSD.org [mailto:tech-net-owner@NetBSD.org] On
> > Behalf Of Richard Braun
> > Sent: mardi 28 septembre 2004 11:42
> > To: tech-net@NetBSD.org
> > Subject: Re: IPv6 routing with non-ppp interfaces
> >
> > On Tue, Sep 28, 2004 at 11:36:59AM +0200, Konstantin KABASSANOV wrote:
> > > >
> > > > /sbin/route -q add -inet6 2001:7a8:432d:1:: -prefixlen 64
> fe80::%fxp1
> > > >
> > > > I tried with -interface too, nothing new. I'm using NetBSD/i386
> > > > 1.6.2_STABLE
> > > > btw.
> > > >
> > >
> > > [KK]
> > > A local-link address like "fe80::" seems a little bit strange...
> > Supposing
> > > that it is the next hop router address over the fxp1 interface, are
> you
> > > sure the address is the correct one?
> > >
> > > Konstantin
> >
> > There is a line about the fe80::%fxp1 destination in the routing table :
> > fe80::%fxp1/64                     link#2                         UC
> > 0        0      -  fxp1
> >
> > It should be the correct address. However, I've only tested this kind of
> > addresses on PPP interfaces (including gif tunnels). You said your
> > configuration was working. How did you configure your routing then ?
> >
> > --
> > Richard Braun
> [KK]
>=20
> This entry means only that all link-local addresses on the fxp1 interface
> can be joined through this interface.
>=20
> If you you have a topology like the following:
>=20
> ---------------------RouterA-------------------------RouterB--------------
> --
> 2001:7a8:432d:2::/64        fe80::....(1) fe80::...(2)
> 2001:7a8:432d:1::/64
>=20
> any you want to route packets from 2001:7a8:432d:2::/64 to
> 2001:7a8:432d:1::/64 between RouterA and RouterB (with link-local
> addresses obly), you have to use :
>=20
> /sbin/route -q add -inet6 2001:7a8:432d:1:: -prefixlen 64
> fe80::...(2)%fxp1
> on RouterA where fe80::...(2) is the link-local address of RouterB.
>=20
> Is it what you want?
>=20
> Konstantin

No, that's what I do with my tunnels and I thought it would also work for
an ethernet interface. My topology is :

            ----- fxp0 Router fxp1 -----
                    ne0(pppoe0)
                         |
                         |

fxp0: flags=3D8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
        address: 00:90:27:71:ed:ae
        media: Ethernet autoselect (100baseTX full-duplex)
        status: active
        inet 192.168.0.1 netmask 0xffffff00 broadcast 192.168.0.255
        inet6 fe80::290:27ff:fe71:edae%fxp0 prefixlen 64 scopeid 0x1
        inet6 2001:7a8:432d::1 prefixlen 64

fxp1: flags=3D8843<UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST> mtu 1500
        address: 00:50:8b:5d:c3:09
        media: Ethernet autoselect (100baseTX full-duplex)
        status: active
        inet 192.168.1.1 netmask 0xffffff00 broadcast 192.168.1.255
        inet6 fe80::250:8bff:fe5d:c309%fxp1 prefixlen 64 scopeid 0x2

Don't care about ne0 and pppoe0. I would like to have 2001:7a8:432d:1::/64
allocated to the fxp1 network, without assigning a global address of that
block on fxp1. Is it possible, and if yes, how ?

--=20
Richard Braun

--ZGiS0Q5IWpPtfppv
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.5 (NetBSD)

iD8DBQFBWYAABlWsEPLYRi8RAkxIAKC0Y+KoiubB1g/UbYspVF6shD18kQCeIlhz
zgNxoJJwucrSb9oGBRB8cSY=
=3fHi
-----END PGP SIGNATURE-----

--ZGiS0Q5IWpPtfppv--