Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752628AbZK0R6S (ORCPT ); Fri, 27 Nov 2009 12:58:18 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1752070AbZK0R6S (ORCPT ); Fri, 27 Nov 2009 12:58:18 -0500 Received: from netrider.rowland.org ([192.131.102.5]:55816 "HELO netrider.rowland.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with SMTP id S1751947AbZK0R6R (ORCPT ); Fri, 27 Nov 2009 12:58:17 -0500 Date: Fri, 27 Nov 2009 12:58:23 -0500 (EST) From: Alan Stern X-X-Sender: stern@netrider.rowland.org To: tmhikaru@gmail.com cc: Jan Kara , Boaz Harrosh , Kernel development list , USB list , Jens Axboe , SCSI development list , Subject: Re: Weird I/O errors with USB hard drive not remounting filesystem readonly In-Reply-To: <20091127094339.GA9047@roll> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1370 Lines: 30 On Fri, 27 Nov 2009 tmhikaru@gmail.com wrote: > It is important to me to know exactly how the failure path operates. Please > explain to me what I will see happen. - Not knowing is driving me nuts. It goes like this. The computer sends a lot of READ commands to the drive. (For all we know the same thing might happen with WRITEs, but you didn't do any writing in the test data you sent.) Every now and then the drive fails to carry out a READ, for no apparent reason. Normally the computer then retries, and the READ succeeds the second time. But of course, success isn't guaranteed. When the retry does succeed, no error messages are printed in the log and everything continues normally. Occasionally the computer does not retry the READ -- in circumstances where it doesn't really need the data (optimistic readahead). When this happens, the failed READ does cause an error message to appear. The same thing would happen if the attempted retries were to fail. Whether or not this would result in lost or corrupted data, or remounted readonly filesystems, depends on the kind of data being read. Alan Stern -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/