Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 19 Jul 2004 12:28:23 +0100
From:      Matthew Seaman <m.seaman@infracaninophile.co.uk>
To:        "Web Walrus (Robert Wall)" <custpriv@web-walrus.com>
Cc:        Nelis Lamprecht <nelis@8ball.co.za>
Subject:   Re: Routing issue
Message-ID:  <20040719112823.GC21175@happy-idiot-talk.infracaninophile.co.uk>
In-Reply-To: <20040719043004.T19557@iceberg.web-walrus.com>
References:  <20040719040431.V19557@iceberg.web-walrus.com> <1090230731.24867.41.camel@nelis.brabys.co.za> <20040719043004.T19557@iceberg.web-walrus.com>

next in thread | previous in thread | raw e-mail | index | archive | help

--Izn7cH1Com+I3R9J
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Mon, Jul 19, 2004 at 04:31:36AM -0500, Web Walrus (Robert Wall) wrote:
> > > ifconfig_dc0 inet 1.2.3.4 netmask 255.255.255.248
> > > ifconfig_dc0_alias0 inet 2.3.4.5 netmask 255.255.255.248
> > > defaultrouter=3D"1.2.3.1"
> >
> > You need to change your netmask for the alias to 255.255.255.255 if it's
> > on the same network.
>=20
> It's not on the same network; that's the problem.  Two complete separate
> networks, same interface card.  The issue is that one of the networks
> works, and the other doesn't, depending on what network the default router
> happens to be on.

Yes -- the OP's configuration is correct as far as it goes.  However
the problem he's facing is rather more intractable than it first
appears.

In general, you're going to need a mechanism for dynamically routing
packets in order to make this sort of setup work.  For most setups,
you'ld need the co-operation of your ISP to make things work as well.

There's two areas where you can use this dual setup profitably.

The first is failover -- should one of the connections go down, you'll
automatically switch to using the other.  About the simplest way of
doing something like that is to run a script periodically (say once
every 5 minutes) that sends a ping down the active channel, and if
there's no response, it switches the default route to the other
channel.  This means that normally all your traffic will go down one
of the connections, and there won't be any bandwidth advantages but
you will get increased resilience.

The second is 'policy based routing' -- which is a good term to google
for.  Under FreeBSD this is implemented using the ipfw(8) 'fwd'
command which lets you dynamically redirect packets down one channel
or the other.  That means you can do things like select out HTTP
traffic and send it via one channel, leaving all of the other traffic
to go by the other.  That lets you share out your bandwidth between
available channels, but doesn't give you any advantages in terms of
resilience.

	Cheers,

	Matthew

--=20
Dr Matthew J Seaman MA, D.Phil.                       26 The Paddocks
                                                      Savill Way
PGP: http://www.infracaninophile.co.uk/pgpkey         Marlow
Tel: +44 1628 476614                                  Bucks., SL7 1TH UK

--Izn7cH1Com+I3R9J
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (FreeBSD)

iD8DBQFA+7BXiD657aJF7eIRAh1HAJ0SHsJXk0gD4kNPzANUI6Z1hc56DgCgrfGn
MM9rHcSIVDEstdUEvh1JmII=
=ov7q
-----END PGP SIGNATURE-----

--Izn7cH1Com+I3R9J--



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20040719112823.GC21175>