From owner-freebsd-stable@FreeBSD.ORG Tue Aug 31 15:28:48 2010 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id D3C1510656AA for ; Tue, 31 Aug 2010 15:28:48 +0000 (UTC) (envelope-from rmacklem@uoguelph.ca) Received: from esa-jnhn.mail.uoguelph.ca (esa-jnhn.mail.uoguelph.ca [131.104.91.44]) by mx1.freebsd.org (Postfix) with ESMTP id 9018C8FC21 for ; Tue, 31 Aug 2010 15:28:48 +0000 (UTC) X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: ApwEAB69fEyDaFvO/2dsb2JhbACDGJ4vrRKSJIEigyJzBIoR X-IronPort-AV: E=Sophos;i="4.56,298,1280721600"; d="scan'208";a="92351997" Received: from erie.cs.uoguelph.ca (HELO zcs3.mail.uoguelph.ca) ([131.104.91.206]) by esa-jnhn-pri.mail.uoguelph.ca with ESMTP; 31 Aug 2010 11:28:46 -0400 Received: from zcs3.mail.uoguelph.ca (localhost.localdomain [127.0.0.1]) by zcs3.mail.uoguelph.ca (Postfix) with ESMTP id CA703B3E95; Tue, 31 Aug 2010 11:28:46 -0400 (EDT) Date: Tue, 31 Aug 2010 11:28:46 -0400 (EDT) From: Rick Macklem To: Giulio Ferro Message-ID: <332176575.321479.1283268526770.JavaMail.root@erie.cs.uoguelph.ca> In-Reply-To: <4C7D01F7.4010003@zirakzigil.org> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [24.65.230.102] X-Mailer: Zimbra 6.0.7_GA_2476.RHEL4 (ZimbraWebClient - SAF3 (Mac)/6.0.7_GA_2473.RHEL4_64) Cc: freebsd-stable@freebsd.org Subject: Re: About zfs + nfs stability X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 31 Aug 2010 15:28:48 -0000 > > The freezes are gone, thankfully, but I often get huge slow-downs: > looking in the logs of the nfs clients I get plenty of: > ... kernel: nfs server ...:/path/to/dir: lockd not responding > ... kernel: nfs server ...:/path/to/dir: lockd is alive again > If you don't need file locking to work across multiple clients concurrently (ie. multiple clients aren't locking the same file at the same time), then you can avoid the NLM by using the "nolockd" mount option on the clients. (Linux has a similar mount option under a different name.) > I don't know if this has anything to do with zfs. I don't believe it has anything to do with zfs. The NLM is a separate protocol from NFS. > 3) Is it a good idea to switch to nfsv4? Performance? Stability? > NFSv4 will provide better file locking (if you need that) imho, but is still considered experimental, so it is hard to say how well it will work for you. Some seem to use it without difficulties, whereas others have problems. There is a recent unresolved thread where a guy has perf. problems on some of his clients, but not all. rick