From owner-freebsd-net@FreeBSD.ORG Mon Mar 6 20:06:17 2006 Return-Path: X-Original-To: freebsd-net@freebsd.org Delivered-To: freebsd-net@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 0F9ED16A422 for ; Mon, 6 Mar 2006 20:06:17 +0000 (GMT) (envelope-from ml.diespammer@netfence.it) Received: from parrot.aev.net (parrot.aev.net [212.31.247.179]) by mx1.FreeBSD.org (Postfix) with ESMTP id 0ADA443D5D for ; Mon, 6 Mar 2006 20:06:00 +0000 (GMT) (envelope-from ml.diespammer@netfence.it) Received: from soth.ventu (adsl-ull-53-236.51-151.net24.it [151.51.236.53]) (authenticated bits=128) by parrot.aev.net (8.13.5/8.13.5) with ESMTP id k26KIKnI044249 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK) for ; Mon, 6 Mar 2006 21:18:28 +0100 (CET) (envelope-from ml.diespammer@netfence.it) Received: from [10.1.2.18] (alamar.ventu [10.1.2.18]) by soth.ventu (8.13.5/8.13.3) with ESMTP id k26K5jZm029492 for ; Mon, 6 Mar 2006 21:05:45 +0100 (CET) (envelope-from ml.diespammer@netfence.it) Message-ID: <440C961A.1060107@netfence.it> Date: Mon, 06 Mar 2006 21:05:46 +0100 From: Andrea Venturoli User-Agent: Thunderbird 1.5 (X11/20060130) MIME-Version: 1.0 To: freebsd-net@freebsd.org Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Scanned-By: MIMEDefang 2.53 on 212.31.247.179 Subject: vr0: rx packet lost X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 06 Mar 2006 20:06:17 -0000 Hello. I've just installed 6.0/AMD64 on an Asus A8V, which features a vr interface. I'm getting tons of "vr0: rx packet lost" kernel messages as soon as I start transfering some files on my LAN. Needless to say, network performance is VERY poor (ranging from 100KB/s to 1MB/s *). I've looked in the archives and in bug reports and I've seen someone has reported this too, but found no solutions. Someone says that the same board connected to a different switch behaves correctly. I can only add that an identical system running 5.4/i386 does not show this problem. So, getting to the questions: _ is this a known problem? _ was it introduced in 6.0? (And possibly would it be fixed in 6.1?) _ is it an AMD64 only issue? If so, would I be better of starting from scratch with i386? _ any insight on which switches should work, which shouldn't and why? _ any other info is welcome. *) Tests indicate ~1MB/s with a generic kernel and 100-150kB/s with my custom one, which introduces ipfw). I haven't tested thoroughly, however, so don't take this figures seriously. bye & Thanks av.