Seagate customers swamped by Barracuda drive failures An update to this story can be found here Seagate 1TB Barracuda drives are failing at an alarming rate, with users complaining of a plague of such failures spreading across the globe. Barracuda 7200.11 drives made in Thailand (ref: ST31000340AS with firmware level SD15) are failing at boot time with a firmware error that is …
My earlier post concerning their reduced warranty period:
"They were losing their butts by warrantying drives good for 3 years (or less) for 5 years. Simple solution, don't improve the drives, cut the warranty period. Typical corporate response."
Looks to me they are getting what they deserve. Unfortunately buyers are not getting what they deserve. I stopped buying Seagate months ago, and I'm so glad. Well, glad until my current maker of choice does the same thing Seagate has done.
Yesterday I got a return number for a failed, slightly earlier, Baracuda (500GB ST3500641AS 7000.9) which will be winging it's way through the post for a replacement later today (once it's finished secure erasing.)
According to DataCent's web site there is a problem to do with a coating on the surface peeling off and sticking to the drive heads. They also have some sounds on their web site to give you a clue as to what the drive sounds like prior to failure. I was convinved it was a firmware issue though, Seagate said not...
Well there we are.. It was coming, from the time the "Freeagent" drives power saving wakeup played up with Linux, but was fine with Windows.
The last big bastion of quality drives is falling. Shame.. My ST506's and ST-225s are still running, some 25 years on, but my freeagent has had a fall from grace at only a few months old! Sign of the times etc.
Back to Hitachi and WD for me.
Although it was due to bad blocks. However it was part of a mirrored pair (Linux mdadm), so we just replaced it.
This sort of story makes me really glad that when setting up a mirror I always use disks from different manufacturers -- just in case. This also avoids getting 2 disks from the same batch that are then, maybe, prone to fail in the same week.
Within the UK, the Sale of Goods Act makes it clear that buyers can claim (against the seller, not the manufacturer) for any design or manufacturing defect. Firmware errors come clearly into those categories. Such lilability lasts for "the reasonable life of the product, which can be longer than the manufacturer's guarantee". Reasonable life of a hard drive is probably 5 years, so anybody with a drive that failed after 2 years should be resonably claiming a 60% refund.
Note: quote paraphrased from Trading Standards explanation, not necessarily taken verbatim from SOGA.
This post has been deleted by its author
I just sent back a 1TB and a 500GB. Two somewhat different problems, but both bricks.
I have 4 WD drives, and 6 Barracudas. I was told the WDs are inferior and unreliable, but they are still going, nearly a year later (constant on), and I have had two of six Seagates croak within two months.
Ironic, eh?
"Just bought 4 of these for my Home NAS..... quite worrying."
I was **JUST** about to buy 2 for a QNAP NAS... I'll opt for the Samsung Spinpoint drives then...
RAID1 - will take Gulfie's advice and test recovery prior to transferring all my data from the various PCs in the house onto them.
There's still a single point of failire is someone comes in and steals the NAS though - shame none of them (including the QNAP) can't automatically backup their data onto Amazon S3. Perhaps one day...
Why do people think seagate are so good? Failed seagate drives caused huge problems for Amstrad. Amstrad lost millions, took seagate to court and won. Buy a variety of brands and reduce the risk.
Raid 5 isn't really suitable for large drives or large arrays. Raid 6 is fast becoming popular especially so in larger arrays for enterprise etc. With two drives instead of one for parity you don't have to worry about the time consuming and intensive rebuild process killing off another drive. You really can't afford to have another drive fail whilst going through rebuild process because you have no redundancy. (Raid 6 requires a minimum of four drives)
In an a raid 6 system with six drives you could have two from WD, two from Seagate and two from hitachi, you could then have both drives fail from one of the manufacturers and still read your data. This gives you brand redundancy and will protect you from a design faults. (provided you limit each manufacturer to two drives or less in a single array)
Maybe it's just us, but we've had perhaps 6 out of 100 (estimated) Seagate Momentus 80GB SATA laptop drives fail within 6 months of being in service.
Regarding all brands: It seems like all brands made some stinkers and some 'golden' drives. Our failure rate in desktops, laptops, and servers seem about even for each brand.
Oddly, perhaps the best reliability I've seen is at my old job, where they have ancient POS equipment running 2.5GB Maxtor drives. Most of them are 11+ years old and still in service with no errors. (yes, the company's too cheap to upgrade)
"As always users should backup their data, but there is an obvious additional problem here. What happens if you backup to disk and the backup disk is also a Barracuda 7200.11?"
Actually, the 7200.11 is kind of meaningless as a drive identifier - it's the speed/cache size designation.
A 7200.10 is a 7200RPM drive with 16MB cache. A 7200.11 is a 7200RPM drive with 32MB cache.
What you really need to ID the specific drive type is the model number, in this case the ST31000340AS.
Getting the picture now on why they reduced their warranty period a couple months back?
I own 3 of their 500GB 7200.11 drives - and two of them failed. The two that failed? Made in China. The one that didn't? Made in Singapore. So one would think that maybe - just maybe - there is a QC problem in the China manufacturing facility.
Fortunately, the replacements I got were refurbed in Singapore, and running well.
My next drive purchase - have years of loyalty to Seagate - will not be one of their drives. Theeir quality has taken a plunge, and instead of fixing the issue - they change the warranty. Been nice knowing ya, Seagate.
Errrmm ... I wouldn't. I've had no end of problems with Samsung F1 drives, especially if they go anywhere near RAID setups.
I'm a WD green power fan at the moment - had no problems at all with them, and the WDTLER tool lets you switch the cheapo consumer drives into behaving nicely in RAID arrays ...
It just doesn't pay to cut that many corners on QC. You will lose your whole customer
base as word gets around your product is gone in the dumper and you are unresponsive
to complaints. Longer warranty is nice; but I would much rather to NEVER have to worry
about making a warranty claim.
Switched to Seagate back when IBM made all those famous "Deathstar" HDs and I had to replace about half of the IBMs I'd rolled out to clients over the past year. Anyone else remember those? Remember how IBM spun off it's drive unit to Hitatchi 15 minutes later? YOu'd send in defective drives and they'd send back Refurbished defective drives. Cost me a fucking fortune cause you can't replace a brand new HD that died right after you sold it to a customer, with the same broken drive refurbished.
At the time Seagate was impossible to beat, they cost more, but the reliability was better than anything else, especially on SCSI drives. I've been less impressed with their SATA offerings.
Although this particular issue is specific to 1Tb drives, I'd say the average quality across the board is way down. The lowered warranty is the final straw. Unfortunately I"m not sure who's better now.
On a side note anyone considering or using 1Tb drives or larger in a RAID 5 setup might want to read and consider;
http://www.communities.hp.com/online/blogs/information-faster/archive/2008/10/27/the-real-mtbf-the-importance-of-better-raid-levels-in-protecting-your-data-medical-and-otherwise.aspx
I've seen several other discussions of the same issue. Basically If you have a RAID 5 array with 1Tb or larger drives and one fails, it takes so long and works the drives so hard to rebuild the array, that it is statistically likely that you will have a second drive failure before the rebuild is completed.
Joe
I had gotten used to Western Digital's free advance replacement and was shocked by Seagate wanting to charge me $26+ for the same service. I do have to pay for return shipping with WD, but they'll sell me a discounted shipping label during the RMA process.
For a normal return Seagate wanted me to purchase a 3rd party foam RMA kit rather then ship it the way it was received (enclosed in 1" of bubble wrap on all sides) in a box of packing peanuts.
For years I usually selected Seagate drives and recommended them to others but now I'm not so sure about them. Last year at work we had a couple of their 320GB drives fail after a couple of months of use - the motors decided they didn't want to spin the platters any more and instead just sat there going bzzzzzzzzzzzzzzzzzzttt.
The 40GB 7200.7 in my semi-retired home PC is still going strong after 7 years or so. I guess they don't make 'em like they used to.
Bought an Iomega external 1Tb from Tesco yesterday. No one has mentioned Iomega (good) but I have no idea who actually manufactures their drives. It isn't for important data anyway; my HD satellite television tuner has a USB port for recording to a USB drive or memory key. With a terrabyte recording space I won't have to watch it anymore.
Nice one on the Velociraptor I have a similar story myself about a 150GB Raptor I was the one who broke the damn drive. Went to put it in system after re-running all the wires in it and snapped the damn sata connector off. I called them to explain what happened with all intention of replacing the controller board on the drive to not lose data on it. They told me not to worry about paying for repairs they just sent me a new drive no questions asked. Score for WD in my book. As it turned out I had moved the data to another drive anyway :)
WD from now on. . .well until they change anyway
pretty poor reporting, you fail to mention any sorts of estimates for drive failures.
I've just bought two of these drives and am considering returning them now but am unsure as you haven't said wether the rate has increased from 0.05% of drives to 0.1% (100% increase => alarming, but still negligible) or up to 10% failure rate from 9.99% (not so alarming apart from the fact it's high but you get my drift).
Pretty goddamn important thing to state !!
Hi,
All hard drives are dodgy:
Fujitsu made some terrorable IDE drives which failed for no reason - some rumour mill was that it was either a chip design problem or rough designer who left and made this a leaving present (I believe it was a chip design problem though). I do like their SCSI drives though...
Maxtor had some 40Gb and 20Gb drives which suffered from head crashs rather to often.
Now Seagate with this (though I wonder if these are Maxtor drives or Seagate drives??? - a model number would be nice!).
Personally I think if you wish to protect your datam you should:
-RAID all drives which have important data (anything over like 80gb in my opinion unless it's only a boot drive with no important data on it),
-Daily backup on tape (autoloader!).
And before anyone thinks SSDs will be better, they seem to all work completely differently from one another (ie proprietary different methods of drive management with different wear leveling standards which I predict will make recovery differcult).
Mike
As another article on El Reg shows Seagate has major managerial problems at the top. Seagate has really screwed up in the last year or so and now even lowly WD is eating their lunch. They seem to be in panic cost cutting quality cutting mode so I would avoid their products until they get the ship turned around.
Just had one of these drives fail in an array on our main backup server. We figured that the server might be a bit unreliable, so we bought 4 more of these drives to configure in another fail over backup server. I just gave my boss a heads up about the 8 ticking time bombs now residing in our backup array.
If only I had seen this article a week ago.
I used to have a pair of hard drives with external flywheels, back when 32 MEGAbytes was a big drive. One of them always refused to start spinning in the morning, so I had a specially carved stick to poke through a hole in the case and prod the flywheel. The drive continued to work, with prodding, until 32 meg became small.
transcript::
Hello. How may I help you?
AC: hiya, I've read your KB article (http://seagate.custkb.com/seagate/crm/selfservice/search.jsp?DocId=207931) about your drives having dodgy firmware, I'm wondering what I need to do to update the firmware now to fix the problem
Matt C.: We won't have a firmware update for those drives until Tuesday of next week
AC: ok no problem, will you update that KB article with a link to the update or will I have to contact you guys again?
Matt C.: depending on the capacity of the drive, you may have to call in for a firmware update
Matt C.: chat can only update firmware on the 1.5TB drives
AC: hmm, ok mine are the 1.0TB drives
AC: what do you recommend I do?
Matt C.: I would wait until next tuesday, and call 1 800 SEAGATE
Matt C.: they should be able to help you with firmware
AC: I'm in the UK so that contact number's no good for me :/
Matt C.: I can get you the uk number
AC: if you wouldn't mind that'd be great
Matt C.: not a problem
AC: How will the update need to be applied. Will it be done when booting with a floppy or can it be done when in Windows?
Matt C.: 00.800.4732.4283
Matt C.: you will download the firmware, and there will be a text file that explains everything.
AC: cool, sounds simple enough, thanks a lot Matt
Matt C.: no problem, a firmwae update is pretty easy to do. You should be fine
AC: out of interest, what happens if the update fails and the drive is bricked? Can we return it to you for a new one free of charge?
Matt C.: it's fairly straightforward, the text document you get with the firmware download is pretty good.
Matt C.: if the drive is in warranty, you can return the drive at:
Matt C.: https://store.seagate.com/webapp/wcs/stores/servlet/SgCheckWarrantyView?langId=-1&krypto=lZV0ifY90uy6PUbeNYxmlJ%2FeOamQTU8vX%2FyVcskL7Ra8YMmV34QjfaXbpO6NlvwR%2BTEC3tVdnhJ0%0AtmaQWKyGUNscSUkjrgNNSrX6sZWZFxPTN7qiNOvGlw%3D%3D&ddkey=SgSSORedirect
Matt C.: Our drives come with a 5 year warranty
AC: sounds good, so are you sure of your timings on the update or is there a potential for it drag out until later on in the week do you think?
Matt C.: I was told that they will have it available in Tuesday
Matt C.: on Tuesday
Matt C.: they're testing it now, so it should be out then.
AC: ok Matt, thanks a lot you've been most helpful
Matt C.: glad I could help. Is there anything else I can assist you with today?
AC: not unless you can give me some winning lottery numbers (;
Matt C.: I wish I could
AC: I guess I shall leave you to it, you've answered all I can think of asking you now so thanks
Matt C.: you're welcome
"No one has mentioned Iomega (good) but I have no idea who actually manufactures their drives."
If it's anything like mine, it's a Seacrate (ST31003.40AS in my case)
Their quality control has gone right down the Gary Glitter ever since they acquired Maxtor, and I had some real stinkers from them, including a OneTouch drive that failed after 3 months - the replacement took a couple of hours(!)
Every drive manufacturer seems to be following the price curve downwards, and I suspect they all have similar skeletons in the cupboard.
I have two WD 320GB laptop drives that mysteriously disconnect while in use. They are fine after a reboot, but hey. Having yer linux system freeze is just no fun. I am casting round for an alternative but find horror stories all over the place.
The free market is supposed to provide choice and opportunity. If the only choice is between idendical dodgy products then what is the point?
Where are the twice-the-price but more-than-twice-the-quality alternatives?
Seagate has isolated a potential firmware issue in certain products, including some Barracuda 7200.11 hard drives and related drive families based on this product platform, manufactured through December 2008. In some circumstances, the data on the hard drives may become inaccessible to the user when the host system is powered on*.
As part of our commitment to customer satisfaction, we are offering a free firmware upgrade to those with affected products. To determine whether your product is affected, please visit the Seagate Support web site at http://seagate.custkb.com/seagate/crm/selfservice/search.jsp?DocId=207931.
Support is also available through Seagate's call center: 1-800-SEAGATE (1 800 732-4283)
Customers can expedite assistance by sending an email to Seagate (discsupport@seagate.com). Please include the following disk drive information: model number, serial number and current firmware revision. We will respond, promptly, to your email request with appropriate instructions. There is no data loss associated with this issue, and the data still resides on the drive. But if you are unable to access your data due to this issue, Seagate will provide free data recovery services. Seagate will work with you to expedite a remedy to minimize any disruption to you or your business.
For a list of international telephone numbers to Seagate Support and alternative methods of contact, please access http://www.seagate.com/www/en-us/about/contact_us/
*There is no safety issue with these products.
I have an HP MSA with, yup..you guessed it...seagate drives...lost 2 at once..ouch...bye, bye, RAID 5. Seagate wouldn't help with anything...blamed HP...although it was a firmware issue with the seagate drives.
I had to go to OnTrack for recovery....$15k later..OnTrack didn't get any data back that was usable...oiy!
HP RMA'd the whole MSA...but they sent 12 more Seagates!!!
Holy mother of cantelope...how can I trust my backup to disk setup now??
I have been buying WD almost exclusively for going on 10 years now. The one time I bought a Seagate it was for my TiVo, because Seagates were supposedly quieter. But the one I received made a quiet but very noticeable high pitched whine and I had to return it. My TiVo and my parents' TiVo both have WD drives now that are always on and seek a lot and they've been working for 5+ years each. Just got a new WD drive for my laptop and it breathed new life into the machine.
I Swore by IBM in the late nineties, and got hit by the Deathstar issue...100% failure rate, however their warranty replacements are still going strong. Put me off IBM/Hitachi drives for a while though.
Im currently on WD drives, but ive lost 3 out of 4 5000AAKS 500Gb drives, 3 out of 8 drives bought for the increasing amount of video editing I do. Still bought WD for other things, but now im trialling some Samsungs as well.
Fact is that every drive manufacturer has failures, occasionally some more than others, its horses for courses. Some people will get stung by a drive going bad and never buy that drive again, while others have years of pain-free use out of the same drives. I recommended WD drives to a friend in the early noughties (before I had my big fail), and his failed in 3 months. Theres very little we can do other than research a drive to see if the failrate is above normal before buying...
...and by backing up, natch.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
