From owner-freebsd-fs@FreeBSD.ORG Wed Sep 22 17:14:08 2010 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 74C7C1065695 for ; Wed, 22 Sep 2010 17:14:08 +0000 (UTC) (envelope-from gull@gull.us) Received: from mail-ey0-f182.google.com (mail-ey0-f182.google.com [209.85.215.182]) by mx1.freebsd.org (Postfix) with ESMTP id 0F6688FC15 for ; Wed, 22 Sep 2010 17:14:07 +0000 (UTC) Received: by eyx24 with SMTP id 24so260344eyx.13 for ; Wed, 22 Sep 2010 10:14:07 -0700 (PDT) MIME-Version: 1.0 Received: by 10.213.36.12 with SMTP id r12mr5705615ebd.96.1285175646829; Wed, 22 Sep 2010 10:14:06 -0700 (PDT) Received: by 10.14.29.76 with HTTP; Wed, 22 Sep 2010 10:14:06 -0700 (PDT) X-Originating-IP: [69.91.158.134] In-Reply-To: <201009221300.o8MD0Cbm030033@higson.cam.lispworks.com> References: <4C9385B0.2080909@shatow.net> <20100917161847.GA58503@icarus.home.lan> <201009221300.o8MD0Cbm030033@higson.cam.lispworks.com> Date: Wed, 22 Sep 2010 10:14:06 -0700 Message-ID: From: David Brodbeck To: freebsd-fs@freebsd.org Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Subject: Re: what happens to pool if ZIL dies on ZFS v14 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: Wed, 22 Sep 2010 17:14:08 -0000 On Wed, Sep 22, 2010 at 6:00 AM, Martin Simmons wrot= e: >>>>>> On Tue, 21 Sep 2010 15:38:22 -0700, David Brodbeck said: >> >> If you don't have a separate log device, synchronous writes are very >> slow with the ZIL enabled. =A0This isn't such a big deal unless you're >> using NFS, where essentially every write is synchronous. > > Is that true for all versions of NFS? =A0In my experience (on 8.0-RELEASE= ), > NFSv2 is indeed synchronous, but NFSv3 does asynchronous flushing (for a > variety of different client OSes). It does allow clients to request asynchronous flushing. My statement that "essentially every write is synchronous" was a bit of an overstatement; the problem comes when the client issues a COMMIT, which happens frequently when doing some operations, such as extracting tar files. These are the operations that can get quite slow when using NFS with the ZIL enabled and no separate log device. By "quite slow," I mean several minutes to extract a tar file that takes less than a minute with the ZIL disabled.