From owner-freebsd-isp Sun Dec 9 15:38:32 2001 Delivered-To: freebsd-isp@freebsd.org Received: from femail4.sdc1.sfba.home.com (femail4.sdc1.sfba.home.com [24.0.95.84]) by hub.freebsd.org (Postfix) with ESMTP id 9054B37B417 for ; Sun, 9 Dec 2001 15:38:26 -0800 (PST) Received: from veager.jwweeks.com ([65.14.122.116]) by femail4.sdc1.sfba.home.com (InterMail vM.4.01.03.20 201-229-121-120-20010223) with ESMTP id <20011209233821.MNEW18071.femail4.sdc1.sfba.home.com@veager.jwweeks.com>; Sun, 9 Dec 2001 15:38:21 -0800 Date: Sun, 9 Dec 2001 18:38:14 -0500 (EST) From: Jim Weeks X-Sender: jim@veager.jwweeks.com To: "Forrest W. Christian" Cc: Rowan Crowe , freebsd-isp@FreeBSD.ORG Subject: Re: arplookup In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-isp@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org This is a very well thought out response, and belive I understand your example. However, this isn't representative of the problem I am having. This particular network is supposed to be set up on class c boundaries. In this case we will say that the ip of my machine is 192.168.2.100 with netmask 255.255.255.0, and my default gateway is 192.168.2.1. I have looked at tcpdumps untill I am blue in the face. While I see a lot of traffic that isn't actually on my wire, most all requests for 192.168.2.100 are tagged to tell 192.168.2.1. Here is where the problem arises. A few requests come from say, 192.168.1.5, 192.168.1.10, etc. These machines aren't on my physical wire, so when my machine answers the request I get the error. My question is this, shouldn't these machines either be asking through 192.168.2.1, or at least show the same mac address as 192.168.2.1? Thanks again, Jim On Sun, 9 Dec 2001, Forrest W. Christian wrote: > Let's say you have two nets, call them a and b. > > We'll say they're subnets of the same "classful c" such as: > > 192.168.1.64-127 and 192.168.1.128-191. Note these are both > 255.255.255.192 subnets. > > Let's say you have a router which is connected to both subnet a and b, > with an address of 192.168.1.65 and 192.168.1.129 for each subnet > respectively. > > The proper configuration would be to set up the hosts on each subnet with > an address from that subnet, and set the default router/gateway to the > respective router ip, and a netmask of 255.255.255.192. > > Each host should be able to get to every other host, regardless of the > subnet, if this is set up correctly. > > Now lets assume you have a FreeBSD host on subnet a, which has an ip > address of 192.168.1.100 and has a misconfigured netmask of > 255.255.255.0. What is going to happen is that it will be able to reach > everything on it's subnet, but not those of subnet b. > > When it tries to reach a host on subnet b, it will look at the address, > say 192.168.1.150, and then see if it is in the same subnet as the freebsd > box. Since net netmask is 255.255.255.0, it thinks that all of > 192.168.1.x is in it's subnet, and as a result, starts sending out arp > packets asking for the MAC address of 192.168.1.150. Since .150 isn't on > the same wire, NOTHING RESPONDS (unless the router is kind enough to do > proxy arp - which it shouldn't do). And then you get the errors you were > seeing. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-isp" in the body of the message