On Sun, 28 Oct 2001, Sven Koch wrote:
> On Fri, 26 Oct 2001, Pekka Savola wrote:
>
> > The commands above do not trigger the message in itself. You've either
> > compiled in IPv6 support, or it's loaded as the module.
> >
> > After the module is loaded and an interface is up, it takes about ~10-15
> > seconds to get this message (configurable) if there are no IPv6 routers
> > present.
>
> How can I turn off IPv6 Router-Discovery for one pcmcia-scheme?
>
Not as you mean it (except by not enabling IPv6). However..
> I need to set my ipv6-addresses by hand (no router-advertising possible
> in my local lan, and won't be possible anytime soon), and add the
> ipv6-default-route by hand too.
>
> it's very bad to do from /etc/pcmcia/network.opts
>
> /sbin/ip addr add xxxx:yyy:zzz:1001::a/64 dev $DEVICE
> /sbin/ip -f inet6 route del default 2>/dev/null
> /sbin/ip -f inet6 route add ::/0 via xxxx:yyy:zzz:1001::1
> {
> sleep 20
> /sbin/ip -f inet6 route del default dev $DEVICE 2>/dev/null
> } &
>
>
> doing
>
> for d in all default $DEVICE ; do
> echo 0 >/proc/sys/net/ipv6/conf/$d/autoconf
> echo 0 >/proc/sys/net/ipv6/conf/$d/accept_ra
> done
>
> seems not to work, the router-discovery is done, and the broken
> default-route is set after 15 to 20 seconds :(
with this, you can toggle whether autoconfiguration happens or not; the
"broken default" you speak of it irrespective of that. Please also note
that...
> > > and an extra default route for ipv6 is added by the kernel:
> > >
> > > root@wonderland:vc/2:~#ip -6 ro
> > > fe80::/10 dev eth0 proto kernel metric 256 mtu 1500
> > > ff00::/8 dev eth0 proto kernel metric 256 mtu 1500
> > > default dev eth0 proto kernel metric 256 mtu 1500
> > > unreachable default dev lo metric -1 error -101
> > > root@wonderland:vc/2:~#
> >
> > You mean the third one? It is required by the specification; if no better
> > route is known, all destinations are considered to be on-link.
>
> but when I set a valid ipv6-defaultroute before router-discovery finishes,
> a better route is known, and the broken defaultroute should imho not be
> set.
... the broken default route is set with metric 256; any routes you add
will override it automatically. So it doesn't need to be removed.
You _may_ be a seeing a known phenomenon that default route will be
ignored for forwarded packets if forwarding is enabled.
--
Pekka Savola "Tell me of difficulties surmounted,
Netcore Oy not those you stumble over and fall"
Systems. Networks. Security. -- Robert Jordan: A Crown of Swords
|