Saving a partially rotten IBM DTLA-307030 Harddisk
Andreas Ntaflos
I've got the following problem which I hope someone could help me with: One
of my boxes running FreeBSD 4.7-STABLE has an IBM DTLA-307030 (30GB) which
worked very well for more than 2 years now, but I think it starts rotting
away according the following:
* The security output shows something like this every day since Dec 11:
-------8<------
Checking setuid files and devices:
find: /usr/home/user1/path/to/file1: Input/Output error
find: /usr/home/user1/path/to/file2: Input/Output error
.
.
[snip, lots of similar blahs]
.
.
find: /usr/home/user1/path/to/file17: Input/Output error
------->8------
So I know that it is just a specific area on the disk that seems to be
done for, which corresponds to the following console messages:
* ad4: READ command timeout tag=0 serv=0 - resetting
ata2: resetting devices .. done
ad4: READ command timeout tag=0 serv=0 - resetting
ata2: resetting devices .. done
ad4: READ command timeout tag=0 serv=0 - resetting
ata2: resetting devices .. done
swap_pager: indefinite wait buffer: device: #ad/0x20021, blkno: 1656,
size: 4096
ad4: READ command timeout tag=0 serv=0 - resetting
ad4: trying fallback to PIO mode
ata2: resetting devices .. done
ad4s1h: hard error reading fsbn 62679087 of 29163648-29163679 (ad4s1 bn
62679087; cn 3901 tn 151 sn 9) status=59 error=40
ad4s1h: hard error reading fsbn 62679087 of 29163648-29163679 (ad4s1 bn
62679087; cn 3901 tn 151 sn 9) status=59 error=40
* trying to `rm -rf` the affected directories and files results in the
same messages:
rm: /usr/home/user1/path/to/file1: Input/Output error
rm: /usr/home/user1/path/to/file2: Input/Output error
ad4s1h: hard error reading fsbn 62679087 of 29163648-29163679 (ad4s1 bn
62679087; cn 3901 tn 151 sn 9) status=59 error=40
ad4s1h: hard error reading fsbn 62679087 of 29163648-29163679 (ad4s1 bn
62679087; cn 3901 tn 151 sn 9) status=59 error=40
As can be seen, the data in that affected area is virtually lost, but
luckily not very important. So my question is, what can I do to get
the disk fully operational again? Some time ago I read that writing binary
zeroes there could be used as some kind of low-level-format to get rid of
the bad sectors. If this would be of any use, how would I accomplish to
dd /dev/zero to just that specific part of the disk, leaving the rest
intact? Is that possible at all? Or are there any other ways to solve that
problem, apart from buying a new disk or low-level-formating the whole
thing?
I would really appreciate any help or hints on that.
Kind regards
--
Andreas "ant" Ntaflos | "A cynic is a man who knows the price of
a...@overclockers.at | everything, and the value of nothing."
Vienna, AUSTRIA | Oscar Wilde
To Unsubscribe: send mail to majo...@FreeBSD.org
with "unsubscribe freebsd-stable" in the body of the message
Dave Uhring
> Hello list (sorry for crossposting, hope I am doing the right thing),
>
> I've got the following problem which I hope someone could help me
> with: One of my boxes running FreeBSD 4.7-STABLE has an IBM
> DTLA-307030 (30GB) which worked very well for more than 2 years now,
> but I think it starts rotting away according the following:
Download the dft utility from IBM.
http://www.storage.ibm.com/hdd/support/download.htm
When you get the failure code e-mail IBM and get the drive replaced if
it is less than 3 years old.
Andreas Ntaflos
> On Thursday 02 January 2003 02:50 pm, Andreas Ntaflos wrote:
> > Hello list (sorry for crossposting, hope I am doing the right thing),
> >
> > I've got the following problem which I hope someone could help me
> > with: One of my boxes running FreeBSD 4.7-STABLE has an IBM
> > DTLA-307030 (30GB) which worked very well for more than 2 years now,
> > but I think it starts rotting away according the following:
>
> Download the dft utility from IBM.
>
> http://www.storage.ibm.com/hdd/support/download.htm
>
> When you get the failure code e-mail IBM and get the drive replaced if
> it is less than 3 years old.
Wow, that worked like a charm, the Disk Fitness Test was able to repair the
bad sectors without any major problems. Really good.
Thanks a lot for that!
--
Andreas "ant" Ntaflos | "A cynic is a man who knows the price of
a...@overclockers.at | everything, and the value of nothing."
Vienna, AUSTRIA | Oscar Wilde
To Unsubscribe: send mail to majo...@FreeBSD.org
Dave Uhring
> Wow, that worked like a charm, the Disk Fitness Test was able to
> repair the bad sectors without any major problems. Really good.
>
> Thanks a lot for that!
The drive will probably fail shortly. Keep your data backed up.
Matthias Andree
> On Thu, Jan 02, 2003 at 03:11:56PM -0600, Dave Uhring wrote:
>> On Thursday 02 January 2003 02:50 pm, Andreas Ntaflos wrote:
>> > Hello list (sorry for crossposting, hope I am doing the right thing),
>> >
>> > I've got the following problem which I hope someone could help me
>> > with: One of my boxes running FreeBSD 4.7-STABLE has an IBM
>> > DTLA-307030 (30GB) which worked very well for more than 2 years now,
>> > but I think it starts rotting away according the following:
>>
>> Download the dft utility from IBM.
>>
>> http://www.storage.ibm.com/hdd/support/download.htm
>>
>> When you get the failure code e-mail IBM and get the drive replaced if
>> it is less than 3 years old.
>
> Wow, that worked like a charm, the Disk Fitness Test was able to repair the
> bad sectors without any major problems. Really good.
Nope. It hid the problems at the expense of spare sectors. Backup your
data and have the drive replaced. See the pertinent DTLA FAQs on the
web. These drives need special treatment.
--
Matthias Andree
Scott Mitchell
> On Thu, Jan 02, 2003 at 03:11:56PM -0600, Dave Uhring wrote:
> > On Thursday 02 January 2003 02:50 pm, Andreas Ntaflos wrote:
> > > Hello list (sorry for crossposting, hope I am doing the right thing),
> > >
> > > I've got the following problem which I hope someone could help me
> > > with: One of my boxes running FreeBSD 4.7-STABLE has an IBM
> > > DTLA-307030 (30GB) which worked very well for more than 2 years now,
> > > but I think it starts rotting away according the following:
> >
> > Download the dft utility from IBM.
> >
> > http://www.storage.ibm.com/hdd/support/download.htm
> >
> > When you get the failure code e-mail IBM and get the drive replaced if
> > it is less than 3 years old.
>
> Wow, that worked like a charm, the Disk Fitness Test was able to repair the
> bad sectors without any major problems. Really good.
>
> Thanks a lot for that!
I wouldn't trust that disk too far anymore, whatever DFT says. I had an
IBM drive (might even have been the same model) fail in much the same
way... DFT brought it back to life, but it just died again a few weeks
later, and that time there was no saving it. Now they go back to IBM as
soon as they start complaining. Thank goodness for three year warrenties!
If nothing else, make sure your backups are working :-)
Scott
--
===========================================================================
Scott Mitchell | PGP Key ID | "Eagles may soar, but weasels
Cambridge, England | 0x54B171B9 | don't get sucked into jet engines"
scott.m...@mail.com | 0xAA775B8B | -- Anon
Andreas Ntaflos
> Andreas Ntaflos <a...@overclockers.at> writes:
>
> > On Thu, Jan 02, 2003 at 03:11:56PM -0600, Dave Uhring wrote:
> >> On Thursday 02 January 2003 02:50 pm, Andreas Ntaflos wrote:
> >> > Hello list (sorry for crossposting, hope I am doing the right thing),
> >> >
> >> > I've got the following problem which I hope someone could help me
> >> > with: One of my boxes running FreeBSD 4.7-STABLE has an IBM
> >> > DTLA-307030 (30GB) which worked very well for more than 2 years now,
> >> > but I think it starts rotting away according the following:
> >>
> >> Download the dft utility from IBM.
> >>
> >> http://www.storage.ibm.com/hdd/support/download.htm
> >>
> >> When you get the failure code e-mail IBM and get the drive replaced if
> >> it is less than 3 years old.
> >
> > Wow, that worked like a charm, the Disk Fitness Test was able to repair the
> > bad sectors without any major problems. Really good.
>
> Nope. It hid the problems at the expense of spare sectors. Backup your
> data and have the drive replaced. See the pertinent DTLA FAQs on the
> web. These drives need special treatment.
I know, I think about half of all the DTLA 30xxxx disks that were shipped 2
or 3 years ago where faulty and died soon after being put to work. A friend
of mine had one and it failed 3 months after he had bought it. Mine survived
up until now and I was hoping that it would last another 2 or 3 years. But it
seems that I was hoping to much. Disks are quite expensive though :)
Anyway, thanks for the help everybody!
--
Andreas "ant" Ntaflos | "A cynic is a man who knows the price of
a...@overclockers.at | everything, and the value of nothing."
Vienna, AUSTRIA | Oscar Wilde
To Unsubscribe: send mail to majo...@FreeBSD.org
Dave Uhring
> I know, I think about half of all the DTLA 30xxxx disks that were
> shipped 2 or 3 years ago where faulty and died soon after being put
> to work. A friend of mine had one and it failed 3 months after he had
> bought it. Mine survived up until now and I was hoping that it would
> last another 2 or 3 years. But it seems that I was hoping to much.
> Disks are quite expensive though :)
>
> Anyway, thanks for the help everybody!
You get your drive replaced under warranty if it is less than 3 years
old.
http://www.storage.ibm.com/warranty/
I shipped mine to California and waited over a month, but IBM did
replace the drive.
Barry Pederson
> On Thursday 02 January 2003 06:27 pm, Andreas Ntaflos wrote:
>
>
>>I know, I think about half of all the DTLA 30xxxx disks that were
>>shipped 2 or 3 years ago where faulty and died soon after being put
>>to work. A friend of mine had one and it failed 3 months after he had
>>bought it. Mine survived up until now and I was hoping that it would
>>last another 2 or 3 years. But it seems that I was hoping to much.
>>Disks are quite expensive though :)
>>
>>Anyway, thanks for the help everybody!
>
>
> You get your drive replaced under warranty if it is less than 3 years
> old.
>
> http://www.storage.ibm.com/warranty/
>
> I shipped mine to California and waited over a month, but IBM did
> replace the drive.
The thing I wasn't sure about was... does it get replaced with a new drive?
or just some other returned drive of the same flaky model that they've
reformatted and mapped out the bad sectors?
IIRC the website where you got an online RMA only guaranteed a refurbished
drive. Can't tell right now though since the above-mentioned URL only says
warranty support is moving to Hitachi in a few days.
Barry
Yeasah Pell
replacement had a "used part" sticker on it. It failed within a month.
I foolishly returned the drive again, this time including a plea to
replace the drive with another model that doesn't fail, but I just got
the same thing back again, 2 weeks later. Now it rots in a drawer.
I'd say I know lots of other people who have had these drives
fail in the same manner, but that would be pretty redundant at
this point.
I wonder if the Hitachi thing will change anything about the
(pretty bad, IMO) way these problems are being handled?
----- Original Message -----
From: "Barry Pederson" <b...@barryp.org>
To: <freebsd...@FreeBSD.ORG>
Sent: Friday, January 03, 2003 1:24 AM
Subject: Re: Saving a partially rotten IBM DTLA-307030 Harddisk
> The thing I wasn't sure about was... does it get replaced with a new
drive?
> or just some other returned drive of the same flaky model that they've
> reformatted and mapped out the bad sectors?
>
> IIRC the website where you got an online RMA only guaranteed a refurbished
> drive. Can't tell right now though since the above-mentioned URL only
says
> warranty support is moving to Hitachi in a few days.
To Unsubscribe: send mail to majo...@FreeBSD.org
Matthias Andree
> The thing I wasn't sure about was... does it get replaced with a new
> drive? or just some other returned drive of the same flaky model that
> they've reformatted and mapped out the bad sectors?
Well... you never know, but they ship refurbished used
equipment. Whatever that means. Our returned DTLA-307045 failed again
after some months. I wonder if it's possible to call the IBM support and
ask for a different model. I'd even take a set of used DPTA drives of
same capacity.
> IIRC the website where you got an online RMA only guaranteed a
> refurbished drive. Can't tell right now though since the
> above-mentioned URL only says warranty support is moving to Hitachi in a
> few days.
Huh. If that's any good... I guess I'd better buy a spare drive.
--
Matthias Andree
Sean Kelly
> I returned a drive, it took a couple weeks to get it back, and the
> replacement had a "used part" sticker on it. It failed within a month.
> I foolishly returned the drive again, this time including a plea to
> replace the drive with another model that doesn't fail, but I just got
> the same thing back again, 2 weeks later. Now it rots in a drawer.
I went through four of them like this. I'm still using the fourth one,
until it explodes. All of them are "servicable used parts", so it means
some other sucker returned them and they're just redistributing them after
"fixing" them.
My first drive lasted about a year. My second drive lasted about 6 months.
My third drive lasted lasted a few weeks. Along with the return of the
third drive, I included
http://www.zombie.org/ibm.txt
with the drive. I also managed to get an 'Ask Slashdot' story posted which
asked people if they were having problems with this series of drive. To
this day, I still get e-mail from people complaining about these IBM
drives.
(And yes, I apologize for contributing to something that is pretty far
offtopic by now.)
--
Sean Kelly | PGP KeyID: D2E5E296
smk...@zombie.org | http://www.zombie.org