From owner-freebsd-questions@FreeBSD.ORG Fri Jun 2 23:54:52 2006 Return-Path: X-Original-To: freebsd-questions@freebsd.org 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 C69E516A421 for ; Fri, 2 Jun 2006 23:54:52 +0000 (UTC) (envelope-from freebsd@sopwith.solgatos.com) Received: from schitzo.solgatos.com (pool-71-117-252-190.ptldor.fios.verizon.net [71.117.252.190]) by mx1.FreeBSD.org (Postfix) with ESMTP id 36A0043D58 for ; Fri, 2 Jun 2006 23:54:49 +0000 (GMT) (envelope-from freebsd@sopwith.solgatos.com) Received: from schitzo.solgatos.com (localhost.home.localnet [127.0.0.1]) by schitzo.solgatos.com (8.13.6/8.13.4) with ESMTP id k52NsmfP018820; Fri, 2 Jun 2006 16:54:48 -0700 Received: from sopwith.solgatos.com (uucp@localhost) by schitzo.solgatos.com (8.13.6/8.13.4/Submit) with UUCP id k52NsmCK018817; Fri, 2 Jun 2006 16:54:48 -0700 Received: from localhost by sopwith.solgatos.com (8.8.8/6.24) id XAA07715; Fri, 2 Jun 2006 23:45:18 GMT Message-Id: <200606022345.XAA07715@sopwith.solgatos.com> To: freebsd-questions@freebsd.org, arsptr@optusnet.com.au Date: Fri, 02 Jun 2006 16:45:18 +0100 From: Dieter Cc: Subject: Re: Filesystem corruption when drives are mirrored, but not otherwise X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list Reply-To: freebsd@sopwith.solgatos.com List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 02 Jun 2006 23:54:52 -0000 > It seems that this was caused by the (infamous) problem when > softupdates and write caching are both enabled. Disabling write > caching seems to have fixed the problem. Interesting. The disk write cache being on should only cause problems if the system goes down hard/unclean, e.g. power failure. Unless the disk firmware has a bug? Anyway the disk write cache should be off, since even a UPS can't guarantee that the system will not go down hard. > Unfortunately the disk write performance is woeful. There must > be a better way surely? I think that getting NCQ support would help greatly. There are also problems like the one with nforce4 and some brands of disks: http://forums.nvidia.com/lofiversion/index.php?t8171.html It is not obvious what any of this would have to do with mirroring vs. not mirroring.