From owner-freebsd-scsi Sat Nov 30 5:35:28 2002 Delivered-To: freebsd-scsi@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id F388037B401 for ; Sat, 30 Nov 2002 05:35:26 -0800 (PST) Received: from salmon.maths.tcd.ie (salmon.maths.tcd.ie [134.226.81.11]) by mx1.FreeBSD.org (Postfix) with SMTP id AB03543EC5 for ; Sat, 30 Nov 2002 05:35:25 -0800 (PST) (envelope-from dwmalone@maths.tcd.ie) Received: from walton.maths.tcd.ie by salmon.maths.tcd.ie with SMTP id ; 30 Nov 2002 13:35:23 +0000 (GMT) Date: Sat, 30 Nov 2002 13:35:20 +0000 From: David Malone To: Daniel Geske Cc: 'Daniel Geske' , freebsd-scsi@FreeBSD.ORG Subject: Re: fsck: PLEASE RERUN FSCK - does not fix problem :-( Message-ID: <20021130133520.GA83659@walton.maths.tcd.ie> References: <20021130105251.GA82084@walton.maths.tcd.ie> <001901c29863$bb15d670$a52efea9@Bowman> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <001901c29863$bb15d670$a52efea9@Bowman> User-Agent: Mutt/1.3.25i Sender: owner-freebsd-scsi@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org On Sat, Nov 30, 2002 at 12:29:20PM +0100, Daniel Geske wrote: > Thanks for your reply. Is there anything I can do like make the disk > skip the bad parts and keep on using the parts that are still good? As I unserstand it, the "MEDIUM ERROR" is the disk saying that it tried to read the requested block, but couldn't. SCSI Drives should be clever enough to be able to map these blocks to spare blocks elsewhere on the disk, but this remapping can only be done on a write. (This is usually enabeled by default, but you may need to enable it with camcontrol.) So, if the information on the disk isn't too important you can try rewriting the sectors on the disk to get the disk to remap them. Something like "dd if=/dev/zero of=/dev/da1" should work, but remember it will wipe the information on the disk. While this often works, if the disk is going bad you find that it will quickly reach a state where you are loosing blocks often enough that the disk is useless and you're better off buying a new one. (On our busier disks we probably see one or two blocks go bad a month, as shown by "camcontrol defects daX". Disks that are going bad any faster than that should be backed up before they die...) David. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-scsi" in the body of the message