From owner-freebsd-fs@FreeBSD.ORG Mon Mar 22 16:34:28 2004 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 500C316A4CE for ; Mon, 22 Mar 2004 16:34:28 -0800 (PST) Received: from zardoc.esmtp.org (adsl-63-195-85-27.dsl.snfc21.pacbell.net [63.195.85.27]) by mx1.FreeBSD.org (Postfix) with ESMTP id E7A8543D2D for ; Mon, 22 Mar 2004 16:34:27 -0800 (PST) (envelope-from ca+envelope@esmtp.org) Received: from zardoc.esmtp.org ([127.0.0.1]) by zardoc.esmtp.org (sendmail 9.0.0.PreAlpha12) with ESMTP id S00000000404775DB01; Mon, 22 Mar 2004 16:35:57 -0800 Received: (from ca@localhost)i2N0Zun5026306 for freebsd-fs@freebsd.org; Mon, 22 Mar 2004 16:35:56 -0800 (PST) Date: Mon, 22 Mar 2004 16:35:56 -0800 From: Claus Assmann To: freebsd-fs@freebsd.org Message-ID: <20040323003556.GA22741@zardoc.esmtp.org> References: <20040317060617.GA23526@zardoc.esmtp.org> <20040319221643.GA90277@dan.emsphone.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20040319221643.GA90277@dan.emsphone.com> User-Agent: Mutt/1.5.6i Subject: Re: softupdates and two different MTAs X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 23 Mar 2004 00:34:28 -0000 On Fri, Mar 19, 2004, Dan Nelson wrote: > In the last episode (Mar 16), Claus Assmann said: > > | program | FS | writes | reads | > > |---------------+------------------+---------+-------| > > | sm8.12.11 | UFS, softupdates | 236 | 0 | > > | sm9.0.0.12 | UFS, softupdates | 3500 | 4 | > So something is either not fsyncing, or there is clustering going on > behind the scenes. The sm8 softupdates count is disturbingly low, even > assuming good clustering. You are right, it is too low. After following your advice about adding the disk I/O stats to sendmail 8 itself, I finally found that the cf file had SuperSafe=m, which causes it to not issue most of the fsync(2) calls. With SuperSafe=true sm8 uses about twice as many disk writes as sm9 and hence the latter is about two times faster than the former. Thanks for your reply!