From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932986AbXIFU65 (ORCPT ); Thu, 6 Sep 2007 16:58:57 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S932304AbXIFU6t (ORCPT ); Thu, 6 Sep 2007 16:58:49 -0400 Received: from hawking.rebel.net.au ([203.20.69.83]:33093 "EHLO hawking.rebel.net.au" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932203AbXIFU6r (ORCPT ); Thu, 6 Sep 2007 16:58:47 -0400 Message-ID: <46E069F9.2040700@davidnewall.com> Date: Fri, 07 Sep 2007 06:28:33 +0930 From: David Newall User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.8.1.2) Gecko/20070221 SeaMonkey/1.1.1 MIME-Version: 1.0 To: Jan Engelhardt CC: Guilherme Vilela , linux-kernel@vger.kernel.org Subject: Re: Problem to recognize that the file system is full References: <89a10cbc0709040951u1786e7e3g76708bfcb7117f44@mail.gmail.com> <89a10cbc0709040953i58f64a9cx957d69dd35519083@mail.gmail.com> In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Jan Engelhardt wrote: > On Sep 4 2007 13:53, Guilherme Vilela wrote: > >> I'm tryng to mount a nfs file system with the option async and run a >> program that writes to the file system. The problem is that the >> program keep writing even when the file system is full. >> > > man 5 exports > > async This option allows the NFS server to violate the NFS protocol > and reply to requests before any changes made by that request > have been committed to stable storage (e.g. disc drive). > (That means it's a feature.) >> This program does'nt occur when mounting with the sync >> option or with async and noac option, but the performance get very >> poor and that is important to my application. The problem doesnt occur >> too with the local file system. I'm running linux centos 5.0. >> > > About ac/localhost, I am not really sure what makes it work. man 5 nfs noac Disable attribute caching, and force synchronous writes. This extracts a server performance penalty but it allows two different NFS clients to get reasonable good results when both clients are actively writing to common filesystem on the server.