From owner-freebsd-questions@FreeBSD.ORG Mon Nov 10 14:26:00 2003 Return-Path: Delivered-To: freebsd-questions@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id CC88E16A4CE for ; Mon, 10 Nov 2003 14:26:00 -0800 (PST) Received: from lilzmailfe01.liwest.at (lilzmailfe01.liwest.at [212.33.55.13]) by mx1.FreeBSD.org (Postfix) with ESMTP id 140D643FBF for ; Mon, 10 Nov 2003 14:26:00 -0800 (PST) (envelope-from dgw@liwest.at) Received: from [212.33.58.27] (helo=cm58-27.liwest.at) by lilzmailfe01.liwest.at with esmtp (Exim 4.14) id 1AJKTv-0004gg-86; Mon, 10 Nov 2003 23:25:59 +0100 From: Daniela To: "Jack L. Stone" , freebsd-questions@freebsd.org Date: Mon, 10 Nov 2003 23:21:44 +0000 User-Agent: KMail/1.5.3 References: <3.0.5.32.20031110151819.01431468@10.0.0.15> <3.0.5.32.20031110160157.0142b020@10.0.0.15> In-Reply-To: <3.0.5.32.20031110160157.0142b020@10.0.0.15> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200311102321.44756.dgw@liwest.at> Subject: Re: Help! Runaway NATD X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 10 Nov 2003 22:26:00 -0000 On Monday 10 November 2003 22:01, Jack L. Stone wrote: > At 10:42 PM 11.10.2003 +0000, Daniela wrote: > >On Monday 10 November 2003 21:18, Jack L. Stone wrote: > >> FreeBSD 4.8-RELEASE-p13 - GW/router/NAT/FW/Caching DNS - Bind-8.3x + > >> IPFW(8) > >> > >> Have just setup the above to route mainly for a LAN (in place of > >> hardware router), but natd runs non-stop as per TOP(1) and keeps eating > >> up the CPU until it crashes. Takes about 1/2 hour to eat up the > >> resources and requires a reboot -- then the problem just starts over > >> again. > > > >Have you tried to attach to the process with gdb(1) or truss(1)? > >Also please show us what `ps ax` says about natd. > >Can you get a core dump? > > > >Good luck! > >Daniela > > Thanks for the quick reply. No, haven't had chance to redo the kernel for > gdb, etc. But, here's some more info from the "ps -ax" and log. The tcp > start out very small and keep growing in big leaps as below -- this is just > before running out of CPU.... I'd definitely need at least the output of truss(1) to tell what's wrong. The commands would be: # killall natd # truss /sbin/natd -f /etc/natd.conf -n rl0