From owner-freebsd-stable Wed Aug 21 10:32: 8 2002 Delivered-To: freebsd-stable@freebsd.org Received: from mx1.FreeBSD.org (mx1.FreeBSD.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 00A0D37B400; Wed, 21 Aug 2002 10:32:04 -0700 (PDT) Received: from tesla.distributel.net (nat.MTL.distributel.NET [66.38.181.24]) by mx1.FreeBSD.org (Postfix) with ESMTP id 572B743E65; Wed, 21 Aug 2002 10:32:04 -0700 (PDT) (envelope-from bmilekic@unixdaemons.com) Received: (from bmilekic@localhost) by tesla.distributel.net (8.11.6/8.11.6) id g7LHV0r31105; Wed, 21 Aug 2002 13:31:00 -0400 (EDT) (envelope-from bmilekic@unixdaemons.com) Date: Wed, 21 Aug 2002 13:30:59 -0400 From: Bosko Milekic To: Gavin Atkinson Cc: freebsd-stable@FreeBSD.ORG, dillon@FreeBSD.ORG Subject: Re: mbuf usage - how do i track it down? Message-ID: <20020821133059.A31078@unixdaemons.com> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5.1i In-Reply-To: ; from gavin@ury.york.ac.uk on Wed, Aug 21, 2002 at 05:30:48PM +0100 Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG On Wed, Aug 21, 2002 at 05:30:48PM +0100, Gavin Atkinson wrote: > > Hi, > > I'm having problems on 4.6-STABLE, with running out of mbufs. I have in > the past had problems with running out of mbuf clusters, but have since > upped that from 1024 to 4096 with kern.ipc.nmbclusters="4096" in > /boot/loader.conf. I now find I am running out of mbufs. > > endor# netstat -m > 16288/16384/16384 mbufs in use (current/peak/max): > 16227 mbufs allocated to data > 3 mbufs allocated to packet headers > 58 mbufs allocated to socket names and addresses > 291/1742/4096 mbuf clusters in use (current/peak/max) > 7580 Kbytes allocated to network (61% of mb_map in use) > 30504 requests for memory denied > 9746 requests for memory delayed > 12606 calls to protocol drain routines > > I have upped it to 32768 with kern.ipc.nmbufs="32768" in > /boot/loader.conf, but as soon as I did that I found I was frequently > getting "No buffer space available" messages while trying to create some > outgoing TCP connections. However, after commenting the line out and > rebooting, i am still getting these messages so they may not be related. > > This box is acting as a nat gateway, with an adsl usb mopdem attached, and > an fxp card for the internal network. Currently I have a maximum uptiime > of around 8 days before I need to reboot because of the leaking mbufs. Interesting. It really does look like there's a leak. Try ifconfig down-ing your interfaces, one by one, and checking if any get freed. -- Bosko Milekic * bmilekic@unixdaemons.com * bmilekic@FreeBSD.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message