Seagate firmware fix bricks Barracudas The updated firmware Seagate issued to solve 1TB Barracuda drive faults is causing 500GB Barracuda drives to fail. The original fault was that Barracuda 7200.11 drives made in Thailand (ref: ST31000340AS with firmware level SD15) were failing at boot time with a firmware error that locked the drive up and rendered it …
Taking a leaf out of the MS book and using your paying customers as beta testers...
Release something that doesn't work properly, it's OK, we'll update it, update causes more problems, release another update that causes more problems, repeat ad nauseum/infinitum.
What happened to fully testing your product and making sure it works before you start selling it?
Seagate released a bios flash to address problems with 1TB drives, and some users applied to to 500GB drives also?
Why oh why would you apply a bios upgrade that didn't provide any benefits? Happy to be corrected and pointed out to be stupid :) My rule of thumb with flashing the bios is "Don't do it unless you've got a problem and make sure it's the absolute last resort"
I have one of the affected drives too, a 500GB 7200.11 one Model ST3500320AS firmware SD15.
Fortunately it's been and still is running fine but for how long I don't know. It's good that the drive really only has a boot partition on most of my data is on other disks but still it would be a bl**dy inconvenience if it failed. Partition backup is called for I think.
What a clusterf*ck from Seagate.
For one thing, as Seagate says, why would you upgrade a drives firmware without backing up your data? You should KNOW if anything goes wrong, you're screwed. Also like Seagate says, why did these users apply a firmware update when they didn't seem to have issues in the first place? Wasn't the original issue with the 1Tb drives? Why did all these 500Gb users go and apply a new firmware? Sometimes it's best to ignore your geek urges and just leave something alone if it's working fine.
It should be pointed out that the KB article at:
http://seagate.custkb.com/seagate/crm/selfservice/search.jsp?DocId=207931
lists the 500GB models with firmware SD15 as recommended for update to SD1A. The people who applied the firmware were not "stupid" they were just following Seagate's instructions in the article.
Yes you are reading this wrong. The firmware 'fix' was issued for a number of 7200.11 drives including the 500GB models. The Seagate website explains that a number of drives are affected and goes on to list a number of drive model numbers - not just the 1TB drives. Admittedly most of the articles about the problems are not very clear on the 'fix' and which drives it applies to.
And, after some bad firmware updates of my own lately, I do agree on the point of not upgrading firmware when nothing has gone worng and/or doesn't provide any additional benefits!
Yes you are. The original register story focused on the 1Tb drives, but seagate's web site listed which drives were affected and included the 500Gb model.
I checked my drives's serial number and was tol dit was affected.
Masses of posts about this on seagate's customer forums and a really interesting post here from a Seagate employee - http://slashdot.org/comments.pl?sid=1098793&cid=26542735
John - waiting for a firmware update to fix my brick!
Why the hell would you update all the drives in a RAID5 at the same time? Do one at a time and make sure everything is fine after each one... Otherwise it kind of defeats the purpose of having redundancy... If you're doing anything to the drives you should always do one at a time, that way if anything goes wrong you still have all your data!
My old Seagate 80 and 160 GB drives just keep on a-hummin'. Until, that is, they fail for some other reason like 20% already have.
ALL things mechanical fail and the solution is: Backup, Backup, Backup. Any scheme should include a backup to a media type other than your primary storage type and off-site storage is usually a pretty good idea.
The instructions for upgrading the firmware say you should only have the drive you want to flash plugged in. Unfortunately some people skipped that bit. A feature of the firmware updater app is that it will quite happily flash any drive that looks like its target. So if you happen to have more than one drive plugged in (the same model as the drive you want to flash) then that will get done too.
Meanwhile I'm just sitting tight and waiting for some new (and tested!) firmware to appear so that I can upgrade my 2 x 500GB which have the (alleged bad) SD15 firmware. But I'll make sue the forums are full of success stories before I attempt anything.
Bought a 500GB Barracuda on Jan 9th, firmware SD15, and the Seagate's support page confirms it is affected. I was just beginning to assemble a new system, so I never actually used or installed the drive. But if I had, I would have upgraded the firmware as soon as possible, seeing that was what the manufacturer suggested to avoid problems. That doesn't sound idiotic to me...
The drive is sitting on a shelf, while I wait for a new (and possibly QA tested!!!) firmware version. In the meanwhile I'll buy something else.
About the "Deathstar": years ago I installed 3 IBM Deathstars in a brand-new (at the time) PowerMac G4 dual 533. Two of those failed within a couple of weeks. Needless to say, I never bought another drive from IBM or (later on) Hitachi.
Now Seagate... Unbelievable...
Is WD any better or should I expect to be robbed of more euros by them as well? To tell the truth, on that PMG4 I later installed a Raptor 10K (the only WD drive I bought so far), and it's still working fine, but then again I'm sure most of you know these things far more than I do.
Seagate lost another customer, for what it's worth...
Not sure I agree with you on that: from past experience, having the same make and model of drives in a RAID ensures good efficientcy on the RAID (a bit like matched memory). I would probably say that you should have the same make and models in a RAID, but if you were going to update a drive in the RAID, you would do so one 1 only not all(!). Just my personal opinion.
And on a different comment: I agree wholeheartedly that if you can not backup your data on a drive, do NOT change the firmware! Only do that if you risk losing your data, in which case there's nothing to lose (apart from you data that will be lost anyway). And if you can't backup your drive, then you will need to beg/borrow/buy something where you can back it up BEFORE updating.
Damn. Looks like I'll be using WD from now on then.
The first "nail" was Seagate dropping their 5 year warranty. That suggests a lack of confidence in the product, not that they were "matching industry norms". This 'k-up is the second and final nail for me and I suspect others, too.
I point blank refuse to use IBM/Hitachi after the deathstar drive experience - 3 dead 40GB drives back in the day was not a pleasant experience.
I've never had a seagate drive fail on me, yet. The best examples are that I've still got 2 working Seagate ST-225 20MB MFM drives hooked to classic Amigas, and they work perfectly along with the ST-251 40MB beast ;) .
Maxtor, Fujitsu, IBM/Hitachi have all given me hassles, but never Seagate. Or WD.
Looks like it's WD. Come on now, we just want the Scorpio Black 500GB 2.5 drive, please!
> My rule of thumb with flashing the bios is "Don't do it unless you've got a problem and make sure it's the absolute last resort"
> Why the hell would you update all the drives in a RAID5 at the same time? Do one at a time and make sure everything is fine after each one.
Yes, indeedy. I'd add that if the drive has been in production for a couple of months (mine have) and if the system has a UPS attached (what big RAID server doesn't?) then it's probably a good idea to let other folks beta- and gamma-test the firmware update. Especially this time!
The post from a Seagate engineer at
> http://slashdot.org/comments.pl?sid=1098793&cid=26542735
referred to above, is a very informative read. The moral, in a follow-up comment there, is surely DON'T FARKING LET MIDDLE MANAGEMENT BYPASS YOUR TRIED AND TRUE TEST/RELEASE PROCEDURE. Amen. There is no problem in the world that management can't make much worse very fast.
> People moaning about dead RAID have just learnt an important lesson, namely never by all the disks from the same company.
But unfortunately if you are setting up a 5-disk 4Tb RAID-5 you'd need five different companies, and are there that many today? Hitachi (formerly IBM), Seagate (took over Maxtor), Western Digital, er.... and if just two out of five drives drop dead that the same time you are still screwed. The lesson I'm taking away is to make sure that if I do disk to disk backup, make sure all the disks in the backup machine are a differnt make to all the disks in the primary.
Finally, something I often tell people: there is no way that any disk drive by any manufacturer can be tested with respect to long-term reliability before it is shipped. Every drive is effectively a prototype, you are relying on what the engineers learned from the previous model and its ancestors for reliability over the next few years. Sometimes the unknowns in the technology will get you. Sometimes a human error will be made. Shit happens, get used to it, unless you can afford to use tried and tested 160Gb drives when state of the art is 1500Gb. Come to think of it, can one still buy the same 160Gb drives as those that were state of the art and proved reliable over several years since? Probably not!
"Backup, Backup, Backup. Any scheme should include a backup to a media type other than your primary storage type and off-site storage is usually a pretty good idea."
So what online backup services do Reg readers recommend - ie. actually use? I'm talking about home use here, and only for important irretrievable stuff - photos, music, docs, maybe videos, etc.
Gigabytes of TV shows lost would be a pain but not the end of the world. Yes this kind of stuff could be backed up to DVD but you know what it's like... ideally a service like this will automatically, and daily, copy off specified filetypes or folders. A bit like a Time Machine in the sky.
Alternatively, name some backup software that'll do the same thing to a 2.5" portable drive.
Well?
Seagate has some serious damage control to do in the coming months/year. I use and recommend Seagate drives all the time, but I'm watching these events carefully to see if their once-great quality and reputation is FUBAR. Which would be a shame. I hate those noisy WD drives.
...to standardize on spare Seagate 7200.11 replacement 500GB drives, as used in the first two. I was dismayed to discover that the third and fourth units are using drives from Samsung and a third brand. Now I'm thinking that variety ain't such a bad thing.
The author of the original thread is off on some major points. Seagate just released further info about the problem to some big OEMs ... details are at http://storagesecrets.org/2009/01/seagate-boot-death-analysis-overhyped-fud/
Bottom line, while it is a firmware bug, the stars all have to line up in a certain way for it to happen. The big one is that the disk had to have been QC'd on one of a small number of test beds that left some debug junk in reserved areas.
8 days ago i send a request to discsupport@seagate.com for a firmware upgrade.
No response.
They could at least send me a notification of receiving my request + assuring to send me the new firmware soon.
The new firmware for my ST3500320NS ES.2 is not afailable from their website.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
