From owner-freebsd-questions Sun Dec 21 06:21:15 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id GAA04060 for questions-outgoing; Sun, 21 Dec 1997 06:21:15 -0800 (PST) (envelope-from owner-freebsd-questions) Received: from ptero.ag.com.br (agsist.centroin.com.br [200.225.60.1]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id GAA04055 for ; Sun, 21 Dec 1997 06:21:09 -0800 (PST) (envelope-from ricardag@ag.com.br) Received: from server01 (du-rjo-88b.centroin.com.br [200.225.57.88]) by ptero.ag.com.br (8.8.5/8.7.3) with SMTP id MAA23209 for ; Sun, 21 Dec 1997 12:21:02 -0200 (EDT) Message-Id: <3.0.32.19971221122142.00973a70@ptero.ag.com.br> X-Sender: ricardag@ptero.ag.com.br X-Mailer: Windows Eudora Pro Version 3.0 (32) Date: Sun, 21 Dec 1997 12:21:50 -0200 To: questions@FreeBSD.ORG From: Ricardo AG Almeida Subject: pppd question Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Sender: owner-freebsd-questions@FreeBSD.ORG X-Loop: FreeBSD.org Precedence: bulk Hi, I own some FreeBSD boxes, and one of them have 3 leased lines connecting remote machines via ppp. I had set up pppd in /etc/ttys (cuaa2 "/usr/sbin/pppd -detach 57600" dial up on, for instance), and it's working fine. But now I have to set up firewall rules, to deny specifics services to some of these remote machines. I had successfully compiled a new kernel, with the firewall options, and applied the rules. That also works fine. The problem i'm facing is that when the machine boots up, the remote boxes connects into the pppN interfaces in a "first come, first served" basis. So, the first remote box that connects grabs the ppp0, the second ppp1 and so on. Clearly, that's a mess with ipfw rules like: ipfw add 1001 deny tcp from 10.0.123.0/24 to any 21 via ppp0 since I can't grant that the 10.0.123 net is always connected via ppp0. Is there any way to force pppd use a specific interface (pppN)? In other words, I wish that the cuaa2 line always uses the ppp0 interface, the cuaa3 uses the ppp1, in a way that the connect order doesn't matter. Is it possible? Best regards, Ricardo A G Almeida AG SISTEMAS http://www.ag.com.br