From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jean-Baptiste Thomas Subject: Re: Paranoid mode for RAID-1 ? Date: Mon, 27 Apr 2015 14:36:51 +0200 (CEST) Message-ID: <1441590395.23298447.1430138211592.JavaMail.zimbra@laposte.net> References: <972437107.22507366.1430116679975.JavaMail.zimbra@laposte.net> <4ADD42F9-407A-4266-A1FC-B92F3D46C36F@insync.za.net> <743316986.23005258.1430129889889.JavaMail.zimbra@laposte.net> <553E1559.2010305@hesbynett.no> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <553E1559.2010305@hesbynett.no> Sender: linux-raid-owner@vger.kernel.org To: David Brown Cc: linux-raid@vger.kernel.org List-Id: linux-raid.ids On 2015-04-27 12:54 +0200, David Brown wrote: > The problem with all of these is that they /might/ be right - > but they /might/ be wrong and make matters worse. Even if you > have 3 copies of the sector, and get two matches and one > different, there is no way to determine that the odd one is > wrong. Perhaps a common bus or connector fault caused the > other two to be wrong. Picking the "majority vote" may > decrease your chances of losing data (but may not - it depends > on the cause of the fault), but it certainly does not avoid > the worst case scenario. So Neil's objection is that it's too paranoid and yours is that it's not paranoid enough ? :-) > Perhaps the best choice during normal usage (as distinct from > recovery or rebuild, when the drive is not mounted) is to > simply report a failure to the layers higher up - that way you > won't make matters worse by giving returning data. You may be right. The main points I think are that a) the inconsistency be caught and reported and b) writes be disabled before the propagation of errors buggers up the whole file system.