[LUNI] Re: how to recover from "offline uncorrectable sectors"
David Ehle
ehle at agni.phys.iit.edu
Fri Aug 20 19:45:17 CDT 2004
Plastic Drive bays are BAD news. They turn into little ovens to bake your
drives into a RMA. 4 drive with the little plastic bays as a mail server.
Really Bad Idea. I went through about a drive every 6 months till I
finally pulled out all the elcheapo plastic bays and got some nice all
metal IcyDocks with front fans. On examinig the removed bays I found
brown places and areas where the plastic had warped and begun to melt
from the heat. I agree with the other posters, you'd be better off
mounting them without a bay then putting them in the little plastic
coffins :(
Next step is to get a 3ware card so I can have hardware raid (less finicky
and prone to false alarms) and IDE hot swapping :).
Good luck!
--
David Ehle
Computing Systems Manager
CAPP CSRRI
rm 077
LS Bld. IIT Main Campus
Chicago IL 60616
ehle at iit.edu
312-567-3751
On Fri, 20 Aug 2004, luni-request at luni.org wrote:
> Send luni mailing list submissions to
> luni at luni.org
> I'm lookin for some advice about how to deal with an IDE drive for
> which smartd has just started reporting "1 Offline uncorrectable
> sectors". Every time smartd does a disk check (every 30 minutes) I
> get another pair of those error messages. What should I do to
> stabilize the system, assuming that I want to keep using this hard
> drive if possible?
>
> Longer story: Around 00:58 today /dev/hdc, a Maxtor 6Y080P0 IDE drive,
> reported some I/O errors of this general form:
>
> Aug 20 00:58:54 example kernel: hdc: dma_intr: status=0x51 { DriveReady SeekComplete Error }
> Aug 20 00:58:54 example kernel: hdc: dma_intr: error=0x40 { UncorrectableError }, LBAsect=12100877, sector=382768
> Aug 20 00:58:54 example kernel: end_request: I/O error, dev 16:03(hdc), sector 382768
>
> Those appeared for a dozen or so sectors. From that point on, the
> routine smartd disk check reports errors every time. They started out
> like this:
>
> Aug 20 01:20:19 example smartd[7125]: Device: /dev/hdc, 2 Currently unreadable (pending) sectors
> Aug 20 01:20:19 example smartd[7125]: Device: /dev/hdc, 2 Currently unreadable (pending) sectors
> Aug 20 01:20:20 example smartd[7125]: Device: /dev/hdc, ATA error count increased from 0 to 19
>
> A smartd short self-test ran at about 02:00 and failed for the first
> time ever (I believe), reporting 19 errors. Only 5 have details, all
> of the following general form and triggered by a READ DMA command:
>
> Error: UNC 3 sectors at LBA = 0x00b8a508 = 12100872
>
> At about 5:20 the error reports from the routine disk check changed to
> this form:
>
> Aug 20 05:20:19 example smartd[7125]: Device: /dev/hdc, 1 Offline uncorrectable sectors
> Aug 20 05:20:19 example smartd[7125]: Device: /dev/hdc, 1 Offline uncorrectable sectors
>
> This disk is quite new; I put it in service on June 25 2004. It
> replaced a WD disk that failed with similar I/O errors. The disk is
> running pretty hot, reporting 50 degC via smartctl. It's been running
> that hot from day one. The disk sits in a plastic drawer that can be
> pulled from the front of the server cabinet and I'm wondering if it's
> not getting enough airflow. The drawer unit has its own fan but,
> still, that 50 degC temp worries me. But not as much as these recent
> I/O errors and bad sector reports worry me.
>
> --
> Fred Yankowski fred at ontosys.com tel: +1.630.879.1312
> OntoSys, Inc PGP keyID: 7B449345 fax: +1.630.879.1370
> www.ontosys.com 38W242 Deerpath Rd, Batavia, IL 60510-9461, USA
>
> ------------------------------
>
> Message: 7
> Date: Fri, 20 Aug 2004 14:31:54 -0500
> From: Martin Maney <maney at pobox.com>
> Subject: Re: [LUNI] how to recover from "offline uncorrectable
> sectors"
> To: Linux Users Of Northern Illinois - Technical Discussion
> <luni at luni.org>
> Message-ID: <20040820193154.GA10266 at furrr.two14.net>
> Content-Type: text/plain; charset=us-ascii
>
> On Fri, Aug 20, 2004 at 12:24:54PM -0500, Fred Yankowski wrote:
> > I'm lookin for some advice about how to deal with an IDE drive for
> > which smartd has just started reporting "1 Offline uncorrectable
> > sectors". Every time smartd does a disk check (every 30 minutes) I
> ...
> > This disk is quite new; I put it in service on June 25 2004. It
> > replaced a WD disk that failed with similar I/O errors. The disk is
> > running pretty hot, reporting 50 degC via smartctl. It's been running
> > that hot from day one. The disk sits in a plastic drawer that can be
>
> Plastic drive trays are a bad idea at 7200 RPM and above. Given the
> evidence of a previous failure in the same setting, I wouldn't hesitate
> a moment: replace the tray (or mount the drive without a tray, maybe).
> Unless you want to assume it's already sustained lasting damage, and
> it's worth the shot at a warranty replacement to run it until it out
> and out dies; the latter seems unlikely. As to the former, who knows?
> 50C is within spec, if not by very much. How long did the WD drive
> survive? That may be suggestive.
>
> --
> Here's my message to the record industry and its allies:
> I'm not a thief. I'm a customer. When you treat me like a
> thief, I won't be your customer. -- Dan Gillmor
>
>
> ------------------------------
>
> Message: 8
> Date: Fri, 20 Aug 2004 14:54:14 -0500
> From: Stef <stefmit at gmail.com>
> Subject: Re: [LUNI] Shameless plug
> To: Linux Users Of Northern Illinois - Technical Discussion
> <luni at luni.org>
> Message-ID: <cad4cac7040820125448484b44 at mail.gmail.com>
> Content-Type: text/plain; charset=US-ASCII
>
> Friday rambling: when you've sent this message, the price of a 17" HP
> monitor (refurbished) was advertised as $149.99. Now I went to visit
> that site again, and the price "jumped" to $299.99. Pretty
> disapppointing, to say the least. I am trying to find in the original
> web site the fine print about changing the price at will, but cannot
> find it ... I hate to think of what may happen to people driving there
> based on your first email ...
>
> Stef
>
> On Wed, 18 Aug 2004 00:50:58 -0500, Tim Wielgos <wiggles at xnet.com> wrote:
> > Hey all--
> >
> > A friend of mine asked me to post this to the LUG...
> >
> > http://www.valueonedirect.com
> >
> > They're having a sale on refurbed electronics and computers on 8/21 and
> > 8/22 up in DesPlaines.
> >
> > I'm assured of some good deals on some interesting stuff. For those
> > interested, this might be fun.
> >
> > Enjoy!
> >
> > Tim
> >
> > --
> > Linux Users Of Northern Illinois - Technical Discussion
> > http://luni.org/mailman/listinfo/luni
> >
>
> ------------------------------
>
> Message: 9
> Date: Fri, 20 Aug 2004 15:26:18 -0500
> From: Fred Yankowski <fred at ontosys.com>
> Subject: [LUNI] Re: how to recover from "offline uncorrectable
> sectors"
> To: Linux Users Of Northern Illinois - Technical Discussion
> <luni at luni.org>
> Message-ID: <20040820202618.GA12775 at ontosoft.com>
> Content-Type: text/plain; charset=us-ascii
>
> On Fri, Aug 20, 2004 at 02:31:54PM -0500, Martin Maney wrote:
> > Plastic drive trays are a bad idea at 7200 RPM and above. Given the
> > evidence of a previous failure in the same setting, I wouldn't hesitate
> > a moment: replace the tray (or mount the drive without a tray, maybe).
>
> I'm going to get that drive out of the tray and mount it directly, as
> you suggest.
>
> > 50C is within spec, if not by very much. How long did the WD drive
> > survive? That may be suggestive.
>
> I installed the WD drive in early October, 2003, and replaced it with
> the Maxtor in late June, 2004. So the WD drive lasted less than 9
> months. The WD drive did not report temperature via SMART, but it sat
> in the same tray in the case and I recall that it felt very hot to the
> touch when I opened the tray to work on it. That WD drive, which was
> nearly inoperable in the server under discussion, works fine when
> installed in a server here in my office.
>
> --
> Fred Yankowski fred at ontosys.com tel: +1.630.879.1312
> OntoSys, Inc PGP keyID: 7B449345 fax: +1.630.879.1370
> www.ontosys.com 38W242 Deerpath Rd, Batavia, IL 60510-9461, USA
>
> ------------------------------
>
> --
> Linux Users Of Northern Illinois - Technical Discussion
> http://luni.org/mailman/listinfo/luni
>
>
> End of luni Digest, Vol 18, Issue 23
> ************************************
>
More information about the luni
mailing list