From owner-freebsd-fs@FreeBSD.ORG Fri Sep 25 23:51:39 2009 Return-Path: Delivered-To: freebsd-fs@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 3DCE31065679 for ; Fri, 25 Sep 2009 23:51:39 +0000 (UTC) (envelope-from andrew@modulus.org) Received: from email.octopus.com.au (email.octopus.com.au [122.100.2.232]) by mx1.freebsd.org (Postfix) with ESMTP id 00CF98FC18 for ; Fri, 25 Sep 2009 23:51:38 +0000 (UTC) Received: by email.octopus.com.au (Postfix, from userid 1002) id E8CD017DA9; Sat, 26 Sep 2009 09:53:13 +1000 (EST) X-Spam-Checker-Version: SpamAssassin 3.2.3 (2007-08-08) on email.octopus.com.au X-Spam-Level: X-Spam-Status: No, score=-1.4 required=10.0 tests=ALL_TRUSTED autolearn=failed version=3.2.3 Received: from [10.20.30.102] (60.218.233.220.static.exetel.com.au [220.233.218.60]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) (Authenticated sender: admin@email.octopus.com.au) by email.octopus.com.au (Postfix) with ESMTP id 020461723B; Sat, 26 Sep 2009 09:53:09 +1000 (EST) Message-ID: <4ABD56D6.50301@modulus.org> Date: Sat, 26 Sep 2009 09:48:38 +1000 From: Andrew Snow User-Agent: Thunderbird 2.0.0.6 (X11/20070926) MIME-Version: 1.0 To: Scott Ullrich References: In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: freebsd-fs@freebsd.org Subject: Re: Slow disk write IO with ZFS / NFS 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: Fri, 25 Sep 2009 23:51:39 -0000 Scott Ullrich wrote: > Does anyone have any pointers on how to speed this up without putting > the data in jeopardy during a power failure, etc (ie: leaving ZIL on). NFS writes are syncronous so ZFS is constantly syncing small datablocks to disk. By default the transaction log is stored on the same disks as the rest of the data. You might like to try an async NFS mount, but this isn't much different from just turning off the ZIL. Failing that, Add a log device to the pool, SSD or ramdisk is ideal, but separate disk HDD spindles to your data zpool also works pretty well. - Andrew