From owner-freebsd-questions@FreeBSD.ORG Mon Nov 16 00:06:57 2009 Return-Path: Delivered-To: freebsd-questions@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 2A4A51065670 for ; Mon, 16 Nov 2009 00:06:57 +0000 (UTC) (envelope-from rfg@tristatelogic.com) Received: from outgoing.tristatelogic.com (segfault.tristatelogic.com [69.62.255.118]) by mx1.freebsd.org (Postfix) with ESMTP id 081C48FC08 for ; Mon, 16 Nov 2009 00:06:55 +0000 (UTC) Received: from segfault-nmh-helo.tristatelogic.com (localhost [127.0.0.1]) by segfault.tristatelogic.com (Postfix) with ESMTP id B857FBDC47 for ; Sun, 15 Nov 2009 16:06:55 -0800 (PST) To: freebsd-questions@freebsd.org Date: Sun, 15 Nov 2009 16:06:55 -0800 Message-ID: <42307.1258330015@tristatelogic.com> From: "Ronald F. Guilmette" Subject: Bad Blocks... Should I RMA? X-BeenThere: freebsd-questions@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: User questions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 16 Nov 2009 00:06:57 -0000 In one of my systems, I've got a Seagate SATA 500GB drive (ST3500320AS) which is actually not very old... purchased 12/11/2008. It's never given me any problems, but just a few minutes ago, while compiling a small C program, I got a set of three irrecoverable errors in quick succession... apparently all read errors from the same single block. Here's the relevant lines from /var/log/messages: Nov 15 15:24:17 coredump kernel: ad4: FAILURE - READ_DMA status=51 error=40 LBA=256230591 Nov 15 15:24:43 coredump kernel: ad4: FAILURE - READ_DMA status=51 error=40 LBA=256230591 Nov 15 15:24:46 coredump kernel: ad4: FAILURE - READ_DMA status=51 error=40 LBA=256230591 (Don't be confused... The name of the host system here is "coredump"... my lame attempt at humor.) So anyway, this is one of those Seagate drives with 5-year warranty. (I only buy the 5-year ones these days... don't trust anything less.) This situation happened at a (relatively) opportune moment. I have zip, nada, nothing on the drive that needs to be either backed up or relocated to another drive. This drive is essentially blank at the moment. So, the question is, should I: 1) RMA the drive back to Seagate? 2) Somehow try to lock-out the bad sector(s)? (If so, how?) 3) Other? If it was failing all over the place (and on multiple blocks), then yea, sure, I'd RMA it back to Seagate in a heartbeat. But heck! It's only one sector. And what's one sector between friends? Before posting this, I googled around a bit for the crrent Accepted Wisdom regarding such sitiations. Most seems to say that bad blocks (even one?) are an early warning of impending doom (for the drive), and suggest trashing or RMA'ing the drive. I just sorta wanted to know if folks here would agree or disagree with that. One thing concerns me about the thought of RMA'ing the drive back... The last time I RMA'd a drive (years ago & a different brand) I got back as a replacement a ``refurb'' drive. Hummm. If I RMA this drive, it is possible that Seagate would replace it with a refurb whose remaining life may perhaps prove to be even less than the drive I am RMA'ing? Do Seagate RMA drive replacements come with fresh platters? Regards, rfg P.S. If I _do_ end up RMA'ing the thing back, do I need to worry about scrubing the drive squeaky clean first... you know... using one of these multiple write-over progs (like `wipe') if I am paranoid... as I am... about the possibility of old credit card numbers lying around in unallocated sectors on the drive? (The drive is empty _now_, but earlier it was in serious/heavy use.) I guess what I'm asking is: Do Segate and the other manufacturers care enough about their customer's privacy to securely wipe old drives/platters that come in to them for RMA? Or do I need to worry 'bout that for my own self?