On Mon, 13 May 2002, [iso-8859-2] Dawid Szyma=F1ski wrote:
> On Sun, May 12, 2002 at 05:20:04PM -0500, Lista de NetBSD Users wrote:
> =3D> Hello
> =3D> zebra=3DYES  zebra_flags=3D"-d"
> =3D> bgpd=3DYES   bgpd_flags=3D"-d"
>
> but did you run zebra and bgpd ?
> afair we _do not_ have any rc.d scripts for it so we need to run it by ha=
nd
> ie from /etc/rc.local

I copied /etc/rc.d/somescript to /etc/rc.d/zebra also bgpd
and I changed the values in the file:
----------------------------------------------------------
#!/bin/sh
#
# $NetBSD: zebra,v 1.3.4.1 2000/10/02 01:08:55 lukem Exp $
#

# PROVIDE: zebra
# REQUIRE: rtadvd

=2E /etc/rc.subr

name=3D"zebra"
rcvar=3D$name
command=3D"/usr/pkg/sbin/${name}"
pidfile=3D"/var/run/${name}.pid"
required_files=3D"/etc/zebra/zebra.conf"

load_rc_config $name
run_rc_command "$1"
------------------------------------------------
#!/bin/sh
#
# $NetBSD: bgpd,v 1.3.4.1 2000/10/02 01:08:55 lukem Exp $
#

# PROVIDE: bgpd
# REQUIRE: zebra

=2E /etc/rc.subr

name=3D"bgpd"
rcvar=3D$name
command=3D"/usr/pkg/sbin/${name}"
pidfile=3D"/var/run/${name}.pid"
required_files=3D"/etc/zebra/bgpd.conf"

load_rc_config $name
run_rc_command "$1"
------------------------------------------------

"/etc/rc.d/bgpd start" doesn't works all times,
then I need check with "ps -ax" and if bgpd is not
working then start it by hand (/usr/pkg/sbin/bgpd -d).

>
> are you sure that your zebra/bgpd are running?
>
> telnet localhost 2601
> telnet localhost 2605
>
> and check it

Yes, both are running... I can do
"telnet localhost zebra" and "telnet localhost bgpd"
and I can see when routers learn new networks via
the other routers. Links between NetBSD machines
are working right.

The problem is with the connection between cisco and
NetBSD via tunnel gif0

Look at this ping6 (It have DUPs!)

router1# ping6 -n ff02::1%gif0
PING6(56=3D40+8+8 bytes) fe80::250:daff:fe69:47fd%gif0 --> ff02::1%gif0
16 bytes from fe80::250:daff:fe69:47fd%lo0, icmp_seq=3D0 hlim=3D64 time=3D0=
=2E227 ms
16 bytes from fe80::aabb:ccdd%gif0, icmp_seq=3D0 hlim=3D64 time=3D34.618 ms=
(DUP!)
16 bytes from fe80::250:daff:fe69:47fd%lo0, icmp_seq=3D1 hlim=3D64 time=3D0=
=2E183 ms
16 bytes from fe80::aabb:ccdd%gif0, icmp_seq=3D1 hlim=3D64 time=3D35.759 ms=
(DUP!)
16 bytes from fe80::250:daff:fe69:47fd%lo0, icmp_seq=3D2 hlim=3D64 time=3D0=
=2E178 ms
16 bytes from fe80::aabb:ccdd%gif0, icmp_seq=3D2 hlim=3D64 time=3D34.61 ms(=
DUP!)
^C
--- ff02::1%gif0 ping6 statistics ---
3 packets transmitted, 3 packets received, +3 duplicates, 0% packet loss
round-trip min/avg/max/std-dev =3D 0.178/17.596/35.759/17.404 ms

Question:

This results of ping6 is enough evidence to say "tunnel is working"???

If it is truth, then I will discard the tunnel and I will
look for filters (routers across the route) or problems
in the sintax of bgp in cisco and zebra.

Thanks a lot

Heron Gallegos