From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: Routing over multiple interfaces Date: Wed, 10 Nov 2010 16:08:24 +0100 Message-ID: <1289401704.2860.182.camel@edumazet-laptop> References: <1288645922.5977.41.camel@macbook.infradead.org> <20101101.141638.116372747.davem@davemloft.net> <1288647330.2660.116.camel@edumazet-laptop> <1289400654.2721.7.camel@macbook.infradead.org> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: David Miller , netdev@vger.kernel.org, uweber@astaro.com To: David Woodhouse Return-path: Received: from mail-ey0-f174.google.com ([209.85.215.174]:44373 "EHLO mail-ey0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756475Ab0KJPIb (ORCPT ); Wed, 10 Nov 2010 10:08:31 -0500 Received: by eye27 with SMTP id 27so339770eye.19 for ; Wed, 10 Nov 2010 07:08:30 -0800 (PST) In-Reply-To: <1289400654.2721.7.camel@macbook.infradead.org> Sender: netdev-owner@vger.kernel.org List-ID: Le mercredi 10 novembre 2010 =C3=A0 14:50 +0000, David Woodhouse a =C3=A9= crit : > On Mon, 2010-11-01 at 22:35 +0100, Eric Dumazet wrote: > >=20 > > David W. probably wants to use teql or some bonding ? > >=20 > > # tc qdisc add dev ppp0 root teql0 > > # tc qdisc add dev ppp1 root teql0 > > # ip link set dev teql0 up > > # ip route add default src 90.155.92.214 dev teql0 >=20 > That works; thanks. Or it should do... there's a slight complication > right now in that one of my lines is actually routed through a separa= te > ADSL<->Ethernet box, because one of the ports on my Solos card is a b= it > shagged and syncs a lot slower than it should. So it's actually eth1 = and > ppp1 that I'm sharing. That should be fixed relatively soon as they'r= e > sending me a new box. >=20 > There's a minor issue with this setup though =E2=80=94 ideally, any g= iven TCP > connection would tend to use the *same* interface as much as possible= , > and only 'overflow' onto the other interface if it's actually saturat= ing > the uplink on the first. That would tend to reduce the amount of pack= et > re-ordering. At the moment it will split outbound packets over both > lines even when they're relatively idle, introducing packet re-orderi= ng > where it wasn't really necessary. >=20 > That said, the world is supposed to cope with packet re-ordering, so = I'm > not going to lose a lot of sleep over it. And I have a feeling the > downstream link from the ISP does it exactly the same. >=20 Its a bit problematic, as you said you want to use both links to do one upload ;) Maybe it is possible to instruct teql to try to not change links unless one link is full...