From owner-freebsd-questions@FreeBSD.ORG Tue Nov 23 16:40:35 2004 Return-Path: Delivered-To: freebsd-questions@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 401B716A4CE for ; Tue, 23 Nov 2004 16:40:35 +0000 (GMT) Received: from dan.emsphone.com (dan.emsphone.com [199.67.51.101]) by mx1.FreeBSD.org (Postfix) with ESMTP id CE9C243D1D for ; Tue, 23 Nov 2004 16:40:34 +0000 (GMT) (envelope-from dan@dan.emsphone.com) Received: (from dan@localhost) by dan.emsphone.com (8.13.1/8.13.1) id iANGeSt4044615; Tue, 23 Nov 2004 10:40:28 -0600 (CST) (envelope-from dan) Date: Tue, 23 Nov 2004 10:40:28 -0600 From: Dan Nelson To: Nikolas Britton Message-ID: <20041123164028.GH48882@dan.emsphone.com> References: <41A2A7B2.4000601@confabulator.net> <20041123030351.GA31803@dan.emsphone.com> <41A2CCFA.4000408@confabulator.net> <20041123034553.GB48882@dan.emsphone.com> <6.2.0.14.2.20041123062022.00acec58@localhost> <41A35DBF.1010208@confabulator.net> <41A3537D.8090403@nbritton.org> <20041123155546.GF48882@dan.emsphone.com> <41A36111.2050100@nbritton.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <41A36111.2050100@nbritton.org> X-OS: FreeBSD 5.3-STABLE X-message-flag: Outlook Error User-Agent: Mutt/1.5.6i cc: "Ryan J. Cavicchioni" cc: freebsd-questions@freebsd.org Subject: Re: SCSI Hard drive "MEDIUM ERROR" X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 23 Nov 2004 16:40:35 -0000 In the last episode (Nov 23), Nikolas Britton said: > Is this the norm SCSI drives?... With IDE drives bad sector remapping > is transparent, the drive does all that by itself, and if you start > getting bad sectors it's because the drive ran out of spares Remapping on an uncorrectable read error is usually not a good idea, because that means your first read to that block will return an error, and the 2nd time the kernel tries, it will succeed (because the block has been remapped) but return garbage. That's a very bad thing to do :) So remapping is only done on a write or on a read where the data was recovered using ECC. I would expect ATA drives to do the same. One nice tool to install is ports/sysutils/smartmontools, which will monitor the error counters kept by the drive and let you know if it's going bad. Works on both SCSI and ATA disks. -- Dan Nelson dnelson@allantgroup.com