From owner-freebsd-isp@FreeBSD.ORG Fri Apr 9 11:10:43 2004 Return-Path: Delivered-To: freebsd-isp@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 3459916A4CF for ; Fri, 9 Apr 2004 11:10:43 -0700 (PDT) Received: from annapolislinux.org (alinux.washcoll.edu [192.146.226.81]) by mx1.FreeBSD.org (Postfix) with ESMTP id C9FE543D72 for ; Fri, 9 Apr 2004 11:10:42 -0700 (PDT) (envelope-from tjk@annapolislinux.org) Received: by annapolislinux.org (Postfix, from userid 1000) id 3F08CC00777; Fri, 9 Apr 2004 14:10:44 -0400 (EDT) Date: Fri, 9 Apr 2004 14:10:44 -0400 From: Theodore Knab To: freebsd-isp@freebsd.org Message-ID: <20040409181044.GA18389@annapolislinux.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.3.28i Organization: Annapolis LUG Subject: ETINC bwmgr on FreeBSD 4.4: Possible memory problems ? X-BeenThere: freebsd-isp@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Internet Services Providers List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 09 Apr 2004 18:10:43 -0000 Hi, I have a Dennis device, the ETINC bandwidth shaper. ;-) It runs Free-BSD version 4.4 running ET/BWMGR v3.23d. This ETINC device is a bridged firewall and exhibiting strange behavior when I run CPU intensive tasks. For example, I recently recompiled the kernel with the BWMGR daemon running and 'make depend' kept existing with a SIGNAL 27. I was adding these items to the LOCAL kernel. options BRIDGE options IPFILTER options IPFILTER_LOG device scbus # SCSI bus (required) After putting the device in fail-over [also know as turning off the BWMGR daemon] 'make' was able to compile my new kernel. Additionally, I added the ports tree to my ETINC device. So, that I could use BASH, my favorite shell. After compiling, BASH and installing I accidently typed 'make clean' in the '/usr/ports ' directory. Rather than stopping the process, I figured it would not hurt anything. 'make clean' just cleans up the object files right ? Anyways, the ETINC box crashed. The BWMGR core dumped: Apr 8 12:05:39 maddog /kernel: pid 582 (bwmgrd), uid 0: exited on signal 8 (core dumped) I am also getting this error on the box when looking at text: bash-2.05# tail -f /var/log/messages | grep -i 'IP .*->25)' Profiling timer expired This 'profiling time expired' was interrupting my kernel compilation also and leading on to SIGNAL 27's. So, I think they are related. Before Dennis, the ETINC support, learned that I was making modifications to his appliance, he said that I could have some bad memory. After I told him I made the above to one of his boxes, he chastised me and told me he could not support a hacked up box. Does this sound like a memory problem or a bleeding bwmgrd problem ? -- ------------------------------------------ Ted Knab Chester, Maryland 21619 USA ------------------------------------------ Conquest is easy. Control is not. -- Kirk, "Mirror, Mirror", stardate unknown