*BSD News Article 53443


Return to BSD News archive

Path: euryale.cc.adfa.oz.au!newshost.anu.edu.au!harbinger.cc.monash.edu.au!news.uwa.edu.au!classic.iinet.com.au!swing.iinet.net.au!news.uoregon.edu!news.dacom.co.kr!newsfeed.internetmci.com!in2.uu.net!news.net99.net!cooking.extremes.net!usenet
From: Jason Fordham <jclf@extremes.net>
Newsgroups: comp.unix.bsd.freebsd.misc
Subject: Re: Is Multi-Link PPP available?
Date: 27 Oct 1995 13:39:28 GMT
Organization: Digital Extremes, Inc.
Lines: 41
Message-ID: <46qneg$6j3@cooking.extremes.net>
References: <45n7ht$81a@rob.inetdirect.net> <3085E6AA.41C67EA6@FreeBSD.org> <465qeh$dmf@news.bu.edu> <46ah2v$g2h@uriah.heep.sax.de>
NNTP-Posting-Host: steaming.extremes.net
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 1.12 (X11; I; BSD/386 uname failed)
To: j@uriah.heep.sax.de
X-URL: news:46ah2v$g2h@uriah.heep.sax.de

j@uriah.heep.sax.de (J Wunsch) wrote:
>Mikhail Teterin <mi@bu.edu> wrote:
>
>>Wait, what is that
>>	pseudo ppp	2
>>			~
>>for? Never had two modems, but what if I just start two pppd with different
>>settings? I thought It would give me two network interfaces...
>
>Of course, but you cannot route the traffic intented for just a single
>remote peer over two different IP interfaces.

That said, you should be able to route outgoing traffic over one interface and
incoming traffic over the other. 

For domain home.net, with routing host interfaces gate-ppp0.home.net,
gate-ppp1.home.net, gate-eth0.home.net; net.home.net set to the appropriate net
(like "net.home.net. in a a.b.c.0", in your home.net DNS file), and interfaces
gate-ppp0.remote.net, gate-ppp1.remote.net, routing can be configured as
follows:

  -  on home.net routing host:

	route add default gate-ppp0
	route add net.home.net -interface eth0

  -  on remote.net router:

	route add net.home.net gate-ppp1.far.net

Assuming that gate-ppp0.home.net and gate-ppp0.remote.net are connected, and
gate-ppp1.home.net and gate-ppp1.remote.net are connected.

Of course, I could be wrong! We use a similar method for pairs of T1's, which
definitely seems to work.

-- 
Jason Fordham
Vice President, Operations
Digital Extremes