From owner-freebsd-current@FreeBSD.ORG Sat May 2 21:12:41 2009 Return-Path: Delivered-To: current@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 73B891065675 for ; Sat, 2 May 2009 21:12:41 +0000 (UTC) (envelope-from tmclaugh@sdf.lonestar.org) Received: from freefall.freebsd.org (freefall.freebsd.org [IPv6:2001:4f8:fff6::28]) by mx1.freebsd.org (Postfix) with ESMTP id D599E8FC14; Sat, 2 May 2009 21:12:40 +0000 (UTC) (envelope-from tmclaugh@sdf.lonestar.org) Received: from straycat.dhs.org (root@localhost [127.0.0.1]) by freefall.freebsd.org (8.14.3/8.14.3) with ESMTP id n42LCdPn093500; Sat, 2 May 2009 21:12:40 GMT (envelope-from tmclaugh@sdf.lonestar.org) Received: from tomcat.straycat.dhs.org (tomcat.straycat.dhs.org [192.168.3.130]) by straycat.dhs.org (8.14.1/8.14.1) with ESMTP id n42LCdYh000402; Sat, 2 May 2009 17:12:39 -0400 (EDT) Message-ID: <49FCB742.2080502@sdf.lonestar.org> Date: Sat, 02 May 2009 17:12:34 -0400 From: Tom McLaughlin User-Agent: Thunderbird 2.0.0.21 (X11/20090320) MIME-Version: 1.0 To: Ryan Stone References: <49D851FC.4090103@sdf.lonestar.org> <49EBC778.7080305@sdf.lonestar.org> In-Reply-To: X-Enigmail-Version: 0.95.6 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: current@freebsd.org Subject: Re: NFS lockd/statd lock up network connection X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sat, 02 May 2009 21:12:43 -0000 Ryan Stone wrote, On 04/20/2009 10:14 PM: > Well, that does confirm that your system is running out of clusters. > Because you lose all network connectivity I'd suspect a leak, probably > in code exercised by lockd. I'm afraid that I know absolutely nothing > about it so I can't offer any kind of solution. Hopefully somebody who > does know something is paying attention. > > One thing you could try is setting the tunable kern.ipc.nmbclusters > higher than 25600 -- maybe 40000? If it's a leak that won't help > anything but if your system just doesn't have enough clusters that will > fix it. Yup, upped the value to 40000 and it still runs out of mbuf clusters. Appears to be something lockd is triggering. I'll check with dfr. tom -- | tmclaugh at sdf.lonestar.org tmclaugh at FreeBSD.org | | FreeBSD http://www.FreeBSD.org |