From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Stefan G. Weichinger" Subject: Re: Pending sectors in valid array - how to proceed? Date: Thu, 29 Jul 2010 10:45:01 +0200 Message-ID: <4C513F8D.5000604@xunil.at> References: <4C506CF5.2080307@xunil.at> <4C5079C4.2050308@seoss.co.uk> <4C5092C4.6050708@gmx.net> <20100729031100.69f583d7@natsu> Reply-To: lists@xunil.at Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20100729031100.69f583d7@natsu> Sender: linux-raid-owner@vger.kernel.org To: Roman Mamedov Cc: st0ff@npl.de, st0ff@gmx.net, Tim Small , linux-raid@vger.kernel.org List-Id: linux-raid.ids Am 28.07.2010 23:11, schrieb Roman Mamedov: > That drive is most likely a Seagate, and if so, there's nothing to worry > about. Literally every Seagate drive will have a high value in > Hardware_ECC_Recovered, it's just a peculiarity of their SMART. Other vendors' > drives recover read errors using ECC too, but don't report that into the SMART > metric. Yep, it's a Seagate. All four are Seagate: sda, sdb: ST3250310NS (should have ERC as far as I found online) sdc, sdd: ST3250621NS (still don't know if they have ERC) I now decided to run that check-action on all three arrays. So far it looks good. All three arrays re-synced OK, without any drive failing. Good :-) Still no reallocated sectors on all four drives. "Current_Pending_Sector" and "Offline_Uncorrectable" on /dev/sdb still at the old value of "13". Do you think I should swap that drive or not? (added difficulty: that server is around 400km from me ... I would have to direct an employee there to swap the hdd ...) Migrating to RAID6, sure, would make sense, but this would need a kernel-upgrade and involves quite some work. Right now I have 2.6.25-gentoo-r7 there :-( thank you all for your replies, Stefan