From owner-freebsd-net Sun Dec 17 3:31:41 2000 From owner-freebsd-net@FreeBSD.ORG Sun Dec 17 03:31:35 2000 Return-Path: Delivered-To: freebsd-net@freebsd.org Received: from mail.interware.hu (mail.interware.hu [195.70.32.130]) by hub.freebsd.org (Postfix) with ESMTP id 226FF37B400; Sun, 17 Dec 2000 03:31:34 -0800 (PST) Received: from gaborone-07.budapest.interware.hu ([195.70.52.135] helo=elischer.org) by mail.interware.hu with esmtp (Exim 3.16 #1 (Debian)) id 147c2N-0000nU-00; Sun, 17 Dec 2000 12:31:32 +0100 Sender: julian@FreeBSD.ORG Message-ID: <3A3CA2DE.28AB4780@elischer.org> Date: Sun, 17 Dec 2000 03:26:22 -0800 From: Julian Elischer X-Mailer: Mozilla 4.7 [en] (X11; U; FreeBSD 5.0-CURRENT i386) X-Accept-Language: en, hu MIME-Version: 1.0 To: Nils Bokermann Cc: net@freebsd.org, questions@freebsd.org Subject: Re: PPPoE and MTU problem References: <20001217113357.A3485@atlas.internal.bermuda.de> Content-Type: text/plain; charset=iso-8859-15 Content-Transfer-Encoding: 7bit Sender: owner-freebsd-net@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org Nils Bokermann wrote: > > Hi! > > The following problem: I have a DSL-Connection to the internet, and a > few client behind the BSD-Box (see picture). I can do connections to the > internet, nearly everything works. But I DO have 2 systems (which I > desperatly need), which do NOT fragment the packets. When using a Linux > Router (don't blame me for that) everything just works fine. Is there a > problem with my configuration, or is that a BSD-specific-Problem. > > client <-- ethernet (MTU=1500) --> FreeBSD-Box <-- DSL > (MTU=1492)-->Provider(german Telekom) > > unakruemel# uname -a > FreeBSD kruemel.monster.sensibelchen.org 4.2-BETA FreeBSD 4.2-BETA #4: > Fri Dec 8 15:27:31 CET 2000 root@kruemel.monster.sensibelchen.org:/usr/src/sys/compile/KRUEMEL > i386 > > kruemel# ipfw list > 00100 allow ip from any to any via lo0 > 00200 allow ip from any to any via rl0 > 00300 allow tcp from any to any out xmit tun0 setup > 00400 allow tcp from any to any via tun0 established > 00500 allow udp from any to any via tun0 > 00600 allow log logamount 100 tcp from any to any 80 setup > 00700 allow log logamount 100 tcp from any to any 22 setup > 00800 allow log logamount 100 tcp from any to any 20 setup > 00900 allow log logamount 100 tcp from any to any 25 setup > 01000 reset log logamount 100 tcp from any to any 113 in recv tun0 > 01100 allow udp from any to any 53 out xmit tun0 > 01200 allow udp from any 53 to any in recv tun0 > 65435 allow icmp from any to any > 65435 deny log logamount 100 ip from any to any > 65535 deny ip from any to any >ng thatyou are using ipfw to filter rather than the it's intersting thatyou are using ipfw to filter rather than the filtering in ppp. but no matter it should still work. Theoretically, the BSD machine should fragment any 'too large' packets coming through, and the router at the other end should too, though the standard behaviour of clients should be to use a smaller size by default for routes that are not local. In any case sometimes this doesn't work right. One answer is to use the newest version of ppp (apply to brian@freebsd for a 4.2 version) which fiddles with the negotiations as they go past and fool both sides to use smaller packets. check recent pppoe sidcussion in the mailing lists, as this problem often shows up with pppoe. -- __--_|\ Julian Elischer / \ julian@elischer.org ( OZ ) World tour 2000 ---> X_.---._/ presently in: Budapest v To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-net" in the body of the message