From owner-freebsd-stable Mon Aug 19 19:54:45 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 4438037B400 for ; Mon, 19 Aug 2002 19:54:42 -0700 (PDT) Received: from smtp2.southeast.rr.com (smtp2.southeast.rr.com [24.93.67.83]) by mx1.FreeBSD.org (Postfix) with ESMTP id 469D743E6A for ; Mon, 19 Aug 2002 19:54:41 -0700 (PDT) (envelope-from bts@fake.com) Received: from mail8.nc.rr.com (fe8 [24.93.67.55]) by smtp2.southeast.rr.com (8.12.5/8.12.2) with ESMTP id g7K2tQtu013653; Mon, 19 Aug 2002 22:55:26 -0400 (EDT) Received: from this.is.fake.com ([24.162.238.30]) by mail8.nc.rr.com with Microsoft SMTPSVC(5.5.1877.757.75); Mon, 19 Aug 2002 22:54:37 -0400 Received: by this.is.fake.com (Postfix, from userid 111) id 797E0BA12; Mon, 19 Aug 2002 22:54:32 -0400 (EDT) Content-Type: text/plain; charset="iso-8859-1" From: "Brian T. Schellenberger" To: Matthew Dillon , freebsd-stable@freebsd.org Subject: Re: root (/) not soft-updates by default ? Date: Mon, 19 Aug 2002 22:54:32 -0400 User-Agent: KMail/1.4.2 References: <20020819144928.GA6628@nebula.wanadoo.fr> <200208191604.g7JG4DOU077232@apollo.backplane.com> In-Reply-To: <200208191604.g7JG4DOU077232@apollo.backplane.com> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Message-Id: <200208192254.32172.bts@babbleon.org> 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 I find that that the performance of my IDE drives is perfectly acceptable with write-caching turned off *if* I also enable softupdates on everything. This is for desktop use, not a server, but then again if I were running a server I believe that I'd use SCSI anyway. On Monday 19 August 2002 12:04 pm, Matthew Dillon wrote: | This conversation is moot, since modern IDE drives (and apparently | seeping into SCSI drives as well) do track-at-once writes to the physical | media, there is no ordering guarentee no matter what you do, and IDE | performance with write caching turned off sucks so badly you pretty much | *have* to turn it on if tags are not supported or suffer penalties so | severe that the last time we tried it our user community was up in arms | over the result. | | So short of caching the data in NVRAM you are *screwed* no matter what | you do. Oh, wait, even caching the data in NVRAM doesn't save us, | since track-at-once writes rewrite sectors we never modified. So we might | as well turn write caching on. If you don't like the default, you can turn | it off in /boot/loader.conf. | | In any case, Jaime Bozza's explanation (the smallish root can fill up | with softupdates enabled when installing a new system over the old one) is | the correct one. This issue is fixed in -current's softupdates but not | fixed in -stable's. The business about IDE not ordering writes is also | correct, but the statement that softupdates is somehow *worse* then a | normal mount with an IDE disk is not correct. Both are equally dangerous | due to the track-at-once writes which IDE drives tend to do these days. | | Since very little on the root partition is modified once a system has | come up, not enabling softupdates on root is not a big deal. | | -Matt | Matthew Dillon | | | To Unsubscribe: send mail to majordomo@FreeBSD.org | with "unsubscribe freebsd-stable" in the body of the message -- Brian, the man from Babble-On . . . . bts@babbleon.org (personal) To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message