From owner-freebsd-fs@FreeBSD.ORG Tue Sep 18 13:14:48 2012 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [69.147.83.52]) by hub.freebsd.org (Postfix) with ESMTP id 6B1C0106566B for ; Tue, 18 Sep 2012 13:14:48 +0000 (UTC) (envelope-from gpalmer@freebsd.org) Received: from noop.in-addr.com (mail.in-addr.com [IPv6:2001:470:8:162::1]) by mx1.freebsd.org (Postfix) with ESMTP id 24FDE8FC14 for ; Tue, 18 Sep 2012 13:14:48 +0000 (UTC) Received: from gjp by noop.in-addr.com with local (Exim 4.80 (FreeBSD)) (envelope-from ) id 1TDxdR-000PhC-GT; Tue, 18 Sep 2012 09:14:41 -0400 Date: Tue, 18 Sep 2012 09:14:41 -0400 From: Gary Palmer To: Volodymyr Kostyrko Message-ID: <20120918131441.GC77784@in-addr.com> References: <000001cd9239$ed734c80$c859e580$@goelli.de> <5052EC5D.4060403@gmail.com> <000a01cd9274$0aa0bba0$1fe232e0$@goelli.de> <505322C9.70200@gmail.com> <000001cd9377$e9e9b010$bdbd1030$@goelli.de> <50559CD8.1070700@gmail.com> <000001cd94f1$a4157030$ec405090$@goelli.de> <50581033.4040102@gmail.com> <20120918112355.GB77784@in-addr.com> <50586B99.40108@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <50586B99.40108@gmail.com> X-SA-Exim-Connect-IP: X-SA-Exim-Mail-From: gpalmer@freebsd.org X-SA-Exim-Scanned: No (on noop.in-addr.com); SAEximRunCond expanded to false Cc: freebsd-fs@freebsd.org Subject: Re: AW: AW: AW: AW: AW: ZFS: Corrupted pool metadata after adding vdev to a pool - no opportunity to rescue data from healthy vdevs? Remove a vdev? Rewrite metadata? X-BeenThere: freebsd-fs@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Filesystems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 18 Sep 2012 13:14:48 -0000 On Tue, Sep 18, 2012 at 03:39:53PM +0300, Volodymyr Kostyrko wrote: > 18.09.2012 14:23, Gary Palmer wrote: > >> From my point of view all hype about moving to 4k sectors is highly > >> irrelevant to ZFS and current products on the market. > >> > >> 1. ZFS tends to use big recordsize for storing any data. This means most > >> files on your drives are already stored in 128k sectors. Storing small > >> tails in 512b or 4k sectors shouldn't give big difference. > > > > Performance testing has shown that running "advanced format" (aka 4kilobyte > > sector disks) with 512 byte alignment with ZFS seriously degrades performance > > compared to running with 4 kilobyte alignment. > > Please understand me correctly, this is only my point of view on the > problem as I never saw any tests that show difference between correct > alignment of _partitions_ and alignment on _records_ on ZFS. This area > is not thoroughly covered with test data. I seem to recall that people made 4 kilobyte aligned partitions on advanced format drives without doing the gnop trick and still suffered worse performance than when they did the gnop trick to make ashift=12. Check the list archives. If you believe there is insufficient testing here and are saying that conventional wisdom regarding this is wrong, it is resonable to request that you prove your position. Gary