From owner-freebsd-questions Mon Dec 11 18:25: 2 2000 From owner-freebsd-questions@FreeBSD.ORG Mon Dec 11 18:24:59 2000 Return-Path: Delivered-To: freebsd-questions@freebsd.org Received: from mail.rdc1.ov.nl.home.com (ha1.rdc1.ov.nl.home.com [212.120.66.198]) by hub.freebsd.org (Postfix) with ESMTP id 4DDED37B400 for ; Mon, 11 Dec 2000 18:24:59 -0800 (PST) Received: from ricin.localnet ([212.120.85.64]) by mail.rdc1.ov.nl.home.com (InterMail vM.4.01.03.00 201-229-121) with SMTP id <20001212022444.CWRE23503.mail.rdc1.ov.nl.home.com@ricin.localnet>; Tue, 12 Dec 2000 03:24:44 +0100 From: Danny Pansters Date: Tue, 12 Dec 2000 03:24:55 +0100 X-Mailer: KMail [version 1.1.99] Content-Type: text/plain; charset="us-ascii" To: cjclark@alum.mit.edu References: <20001210150314.P96105@149.211.6.64.reflexcom.com> In-Reply-To: <20001210150314.P96105@149.211.6.64.reflexcom.com> Subject: Re: Configuring Gateway/NAT on Freebsd -- different networks? Cc: freebsd-questions@freebsd.org MIME-Version: 1.0 Message-Id: <00121203245500.28610@ricin.localnet> Content-Transfer-Encoding: 8bit Sender: owner-freebsd-questions@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG Having read the discussion so far I wonder: isn't the problem related to having your dsl link plugged into your hub. I have cable, not DSL, but in my case I definately can't plug my cable link in my hub and it would magically connect to my internally networked boxes. Now I'm not an expert on this, but my guess would be your DSL provider uses DHCP like my cable provider does and that you prefer to configure your ethernet interface statically like I do too. My next guess would be that your provider requires a box (ethernet card) not some sort of bridge bridge (your hub) to tell them its present, yell at their router and ask for name resolving etc. Your hub can't do that. It just doesn't seem logical to me that it would work the way you're trying. And why use the 172.16 range and not the 192.168 which is a c class so the netmask one would guess would actually be the right one? why complicate things. I've worked with a small ISP for over a year (they used Debian Linux) and we used the multiple IP#'s on interfaces for apache to have seperate IP numbers for our clients' websites, before we switched to virtual hosting. But in those cases all the eth0:1 .. eth0:n interfaces were on the same network as the eth0 (replace eth with xl if you like). Maybe it could work if you'd find a supernet/mask that includes both your dsl IP# and your local Ip range, if possible?? Anyway, like I said, I certainly don't know everything, and if somewhere my reasoning is way wrong, please someone let me know. It's an interesting case. Maybe I'm incorrect in assuming that what's being discussed here is not the same as the trick described in the paragraph above? I'd be interested in hearing what other people think, I've often wondered about how far one can take this "interface splitting". Best regards, Danny To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-questions" in the body of the message