From mboxrd@z Thu Jan 1 00:00:00 1970 From: Goswin von Brederlow Subject: Re: feature suggestion to handle read errors during re-sync of raid5 Date: Sun, 31 Jan 2010 19:04:15 +0100 Message-ID: <87bpgai21s.fsf@frosties.localdomain> References: <4B6471A1.2070407@texsoft.it> <4B6482BD.6090102@anonymous.org.uk> <4B65AD05.5050000@anonymous.org.uk> <4B65B10F.6060501@shiftmail.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: In-Reply-To: <4B65B10F.6060501@shiftmail.org> (asdo@shiftmail.org's message of "Sun, 31 Jan 2010 17:34:23 +0100") Sender: linux-raid-owner@vger.kernel.org To: Asdo Cc: John Robinson , Linux RAID List-Id: linux-raid.ids Asdo writes: > John Robinson wrote: >> On 30/01/2010 21:33, Mikael Abrahamsson wrote: >> [...] >>> I think the 4k sector size on WD20EARS (for instance) is supposed >>> to add more ECC information but I'm not sure how this will affect >>> the 10^14 error rate. >> >> iirc part of the point of moving to 4K sectors is to improve the >> error correction to something like 1 in 10^20 or 22 without losing >> storage density, partly by using what was lost before in >> inter-sector gaps and partly because you can do better with more >> bits of ECC over more data. > I remember the other way around: the purpose of 4k was to keep the > same error rate while saving storage space I thought the purpose of 4k was to reduce the number of blocks so the wear leveling has to handle less data. MfG Goswin