From owner-freebsd-net@FreeBSD.ORG Mon Apr 10 19:43:25 2006 Return-Path: X-Original-To: freebsd-net@freebsd.org Delivered-To: freebsd-net@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id E6DCA16A403 for ; Mon, 10 Apr 2006 19:43:25 +0000 (UTC) (envelope-from zanecb@midwest-connections.com) Received: from mail.midwest-connections.com (mail.midwest-connections.com [69.148.152.10]) by mx1.FreeBSD.org (Postfix) with ESMTP id 6C4BB43D49 for ; Mon, 10 Apr 2006 19:43:25 +0000 (GMT) (envelope-from zanecb@midwest-connections.com) Received: (qmail 15127 invoked by uid 503); 10 Apr 2006 19:46:20 -0000 Received: from unknown (HELO zerda) (zanecb@69.155.32.130) by 0 with ESMTPA; 10 Apr 2006 19:46:20 -0000 Date: Mon, 10 Apr 2006 14:43:35 -0500 From: "Zane C.B." To: freebsd-net@freebsd.org Message-ID: <20060410144335.44ae68a3@zerda> In-Reply-To: References: Organization: Midwest Connections Inc. X-Mailer: Sylpheed-Claws 2.0.0 (GTK+ 2.8.17; i386-portbld-freebsd6.1) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Cc: dima <_pppp@mail.ru> Subject: Re: is NFS production-ready ? X-BeenThere: freebsd-net@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Networking and TCP/IP with FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 10 Apr 2006 19:43:26 -0000 On Mon, 10 Apr 2006 23:26:40 +0400 dima <_pppp@mail.ru> wrote: > First, searching through the archives I'm about to say "No". > > My goal is to provide NFS service to many FreeBSD clients sharing the > exports. The usage pattern appears to be "many reads and not as much > writes". The deployment might look like the following: a SAN and 2 > NFS servers sharing its LUNs. The servers use hot-standby scheme > provided by CARP (or its equivalent). Many FreeBSD clients would > share their exports. I wish servers ran FreeBSD also since it's the > best known OS for the company administrators. > > The majors are: > - no data corruption > - no hangs (this seems to be the largest problem with current > implementation) > - client retry on failure > - a reasonable read speed > > My questions: > 1. NFS/UDP (it's stateless!) is considered to be "evil". Why > (assuming I can grant a balanced network bandwidth)? 2. NFS server > implementation seems to be very buggy. Any success stories? Well, NFS > servers can easily run Linux, Solaris etc. 3. Is at least > implementation of NFS client (either kernel-side or user-space) > stable enough for production use? Client OS replacement is impossible > (hardly suitable, really) in my project. > > PS: The competing options are either SMB or CODA for now. Any other > suggestions? > > PPS: I'd be happy to hear that FreeBSD supports at least one really > clustered FS (proprietary ones are also OK). But I think I wouldn't :( I have been using NFS on FreeBSD for years and never have a had a problem. As long as the network is sound it works nicely.