From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from brick.kernel.dk ([93.163.65.50]:53011 "EHLO kernel.dk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751753AbZDRRsN (ORCPT ); Sat, 18 Apr 2009 13:48:13 -0400 Date: Sat, 18 Apr 2009 19:48:12 +0200 From: Jens Axboe Subject: Re: Recent changes Message-ID: <20090418174812.GD4593@kernel.dk> References: <20090409040002.0D18237A271@kernel.dk> <512785.43007.qm@web63808.mail.re1.yahoo.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <512785.43007.qm@web63808.mail.re1.yahoo.com> Sender: fio-owner@vger.kernel.org List-Id: fio@vger.kernel.org To: Girish Satihal Cc: fio@vger.kernel.org On Fri, Apr 17 2009, Girish Satihal wrote: > Hi Jens, > I am seeing a serious problem with fio testing. I tried running fio > test on ext3 partition with Sequential Reads and Writes. I keep > getting the I/O write error for Sequential writes. I tried with > multiple servers with different controllers and I still get the same > error. With Sequential Reads and Random Reads/Writes there is no issue > seen. Please find the attached fio results file for Sequential > Reads/Writes which has the error details pertaining to Sequential > writes on ext3 file system. Thanks, Girish That's pretty strange. Your test job(s) seem pretty massive, you don't happen to have a cut down version that exhibits the problem? There also seems to be parts of the error missing from io_u_log_error(), like the file associated with the error and the io_u details (offset= and buflen=, for instance). What version of fio are you using? -- Jens Axboe