Seagate Hit with Class Action Lawsuit for High Failure Rates

[Notional]

8 hours ago, Comic_Sans_MS said:

I am pretty sure that one of the of my HDDs I use all the time is a 7200.11, it hasn't failed, it was from a computer that someone smashed up and it is still fine. 

 

Could be. I can only assume it was firmware updated. If not, it could lock up at any time.

 

7 hours ago, kirashi said:

Can confirm the 7200.11 series were absolute crap. It wasn't even Seagates' inability to fix the firmware caching bugs.
For those who don't know, basically if the write-cache filled up completely, the drive would freeze as the cache would not actually write-out data to the platters.
This happened if you wrote a specific size of smaller-ish files, or the drive managed to fill it's own cache during a large continuous write.
Or (for me) literally anytime I slept or woke up my computer from sleep or hybernation state.
I had a 1.5TB 7200.11, the worst offender in the series. The smaller the drive, the less it was affected.

No, the best part was Seagate denying anything was ever wrong with the drives, until the community on Seagates' since-deleted/migrated forums performed some very specific tests with small groups of files during writes to the drive. Seagate finally acknowledged the problems, and released 2 firmware updates, neither of which actually resolved the caching issue.
One of the firmware update was actually designed to slow down writes to the drive in hopes that users wouldn't fill up the cache as often.
All of the recent information and press releases from Seagate point to a BuSY state during a power on as the root cause, which happens due to a full cache.

Seagate then claimed that it didn't matter anymore as all these drives were now at least 3 years old and thus completely out of warranty.
(This wasn't exactly public information unless you emailed them attempting to open a warranty claim due to the firmware bug.)
(Also note that while some users were given the run around, others were actually offered free data recovery services.)
It wasn't until I recently discovered The Solution for Seagate 7200.11 HDDs over at msfn.org that I knew of a solution.

TL;DR: Seagate manufactured an entire series of defective drives, denied it, made the drives worse with firmware updates, then said "bite me" to their customers.

 

This is exactly the problem, and what my 1TB drive suffered from too. I assume it happened during a windows shutdown, as the drive was not seen by the BIOS at the next startup. When I got into Windows the drive was simply gone.

When the drive corrupts its cache, it goes into a fail safe, that shuts down the drive, so it can no longer be seen by the motherboard.

 

The solution you post is the one I used, but I used one of these:

 

At the end I re activated the drive, and firmware updated it. It worked fine, but after less than half a year it perma died to clicking sound of death, where the read/write head just smashes back and fourth because it can no longer see anything. Think Elle Driver in kill bill after she has her second eye removed.

 

A nice example here:

My 1TB drive had a total lifetime of less than 1 year. And yes the clicking sound of death is caused by shitty firmware too, but could also be shitty read/write heads.

 

Either way, as you mention, the way Seagate handled this catastrophe, was completely unacceptable. If the consumers who have joined this class action law suit, has experienced similar response from Seagate, then this lawsuit is completely founded in my opinion.

[Dabombinable]

17 minutes ago, Notional said:

 

Could be. I can only assume it was firmware updated. If not, it could lock up at any time.

 

 

This is exactly the problem, and what my 1TB drive suffered from too. I assume it happened during a windows shutdown, as the drive was not seen by the BIOS at the next startup. When I got into Windows the drive was simply gone.

When the drive corrupts its cache, it goes into a fail safe, that shuts down the drive, so it can no longer be seen by the motherboard.

 

The solution you post is the one I used, but I used one of these:

 

 

 

 

At the end I re activated the drive, and firmware updated it. It worked fine, but after less than half a year it perma died to clicking sound of death, where the read/write head just smashes back and fourth because it can no longer see anything. Think Elle Driver in kill bill after she has her second eye removed.

 

A nice example here:

My 1TB drive had a total lifetime of less than 1 year. And yes the clicking sound of death is caused by shitty firmware too, but could also be shitty read/write heads.

 

Either way, as you mention, the way Seagate handled this catastrophe, was completely unacceptable. If the consumers who have joined this class action law suit, has experienced similar response from Seagate, then this lawsuit is completely founded in my opinion.

What happened in 2003 is why I only trust Samsung HDD at the moment (2.5" and 3.5", had excellent experiences with both, and not their SSD):

http://redhill.net.au/d/119.php

They kept their 3 year warranty while Seagate, WD and Maxtor at the same time reduced theirs to 1 year. I've still got a 1992 WD AC280 running like new though,and my WD Blue has seen 2 years now with no change to its SMART status, so ATM its Samsung>WD for me.

 

Edit: For those that don't follow links,

Of the five main manufacturers, only Samsung continued to stand behind its products with a three year guarantee. The other manufacturers stridently denied that the warranty cuts had anything to do with reliability problems but that was demonstrably untrue. It was no coincidence that the one compay still offering three year warranty was also the one company with significantly lower in-service failure rates.

[HalGameGuru]

Many offered longer warranties with higher end devices. WD Black, Samsung in general, Hitachi and Toshiba continued to offer longer warranties. WD Blacks still get 5 years last I saw.

[Dabombinable]

16 minutes ago, HalGameGuru said:

Many offered longer warranties with higher end devices. WD Black, Samsung in general, Hitachi and Toshiba continued to offer longer warranties. WD Blacks still get 5 years last I saw.

Point is, all of their HDD used to have at least a 3 year warranty.

[HalGameGuru]

Just now, Dabombinable said:

Point is, all of their HDD used to have at least a 3 year warranty.

To expect all of them to, forever, on every product is to be disingenuous. Warranties are a reflection of product tier, placement, price, support, and longevity. People wanted 50 dollar 1TB drives more than they wanted 5 year warranties. No company is obligated to provide any more or less warranty than what they opt to provide that people who wish to buy their product are willing to accept in order to give that company their money. I have no problem with a 1 year warranty, I have a problem with those specific drives because they were garbage no matter how long the warranty is. And getting a free replacement after 4 years rather than 1 doesn't get me my data back.

[LAwLz]

17 hours ago, Notional said:

Backbllazes numbers are 100% legitimate. All drives are tested similarly, but Seagates drives just dies faster. Sure they are put in extreme situations, but that's how you test things. Remember that 1Petabyte SSD test? Yeah that's not normal usage either, but it still paints a picture of the reliability of the drives.If the test was Seagate consumer drives versus WD enterprise drives, then fair enough, but that is not the case. It's an apples to apples comparison, whether people like it or not.

They are 100% legit in the same way "4 out of every 5 people involved in gang-rape enjoys it".

You do NOT test things by putting them in "extreme situations" for which they were not even designed to operate in. You don't test a sports car by driving it in a muddy field for example, but you might test an SUV by doing that. You do different tests for different products. The one petabyte test was testing the endurance of NAND. The operations and testing environment were very controlled and doing the same kind of thing a regular consumer would (write data, over and over). The fact that you compare BackBlaze to the petabyte test shows how little you understand about this entire deal. Backblaze are not even an apples to apples comparison. All drives are just thrown into different racks with different amounts of vibrations and heat output. That's not a controlled test at all.

 

18 hours ago, Notional said:

Now the numbers are primarily due to 2 issues:

1. The 7200.11 series was abysmal with a failure rates of up to 50% within a year. They had defective firmwares that would brick the vast majority of the drives within a month or two. Mine died (I even reset the firmware using hyperterminal with a USB to serial port and removing the PCB). Backblazes drives dropped like flies, completely in conjunction with the consumer experience.
2. 3 TB drives seems to have fundamental issues. Not just on Seagate, but also WD. Idk if it's because of platter density going into a new generation or whatever, but it does hit Seagate pretty damn hard.

Seagate crapped all over itself with the 7200.11 series and has never really been good since. Blaming WD flooding makes no sense. Only Seagate is responsible for quality, and they have failed.

Your numbers doesn't show the picture of the life time of these new drives. Especially within guarantee or MTBF. So they really don't say all that much.

The 7200.11 was badly affected.

Yes, the 7200.11 was bad. So what? All companies can mess up with one product. Like I said in the previous thread, there have been AMD cards (specific models) which have had very high failure rates as well. Does that mean we should completely stop recommending AMD cards altogether? Or does that mean we should maybe not recommend that specific model? I personally vote for the latter, but with your logic we should do the former (and stop recommending AMD). Do you see the flawed logic? You can't base your entire image of a company on a single product while ignoring all their other ones.

If you look at some actual facts which are applicable to consumers (such as the ones I posted) then you would see that you are incredibly wrong regarding the reliability of Seagate drives. They are actually MORE reliable than WD drives (within the first year). Usually when a HDD survives the first year they keep going strong for a long time.

 

 

TL;DR: You are a moron and are trying to apply statistics (which might be 100% true) to the wrong situation. You use consumer workloads to evaluate consumer hardware, and server workloads to evaluate server hardware. The two should not mix when evaluating reliability.

The statistics I posted are consumers using their drives for consumer things, and Seagate are just fine there. Your statistics (BackBlaze) are consumer drives for server workloads, and all drive manufacturers fail there, including WD which also got abysmal failure rates. WD would go bankrupt if the RMA rates were anywhere near those numbers for consumers.

[Mikensan]

Not sure why saying drives tested in a data center is a poor test... You're talking about a room where humidity and temperature are controlled as well as limited access.

 

Why is anyone crying about reliability? If it lasts longer than the warranty, you got your money's worth. If it dies within warranty, replace it. If it dies and you lose all your data... that's your fault. Any drive you own has a chance to die and you could be the lucky winner - protect your data. If you expect a drive to be reliable after 5 years you're kidding yourself.  No company is going to lose money by charging consumer prices for enterprise equipment. 

[Notional]

22 minutes ago, LAwLz said:

They are 100% legit in the same way "4 out of every 5 people involved in gang-rape enjoys it".

 

TL;DR: You are a moron and are trying to apply statistics (which might be 100% true) to the wrong situation.

 

Seriously you need to chill right now. Are you for real? Do you have financial investment in the company or something?

 

Fact of the matter is that Seagate consumer drives and other consumer drives have been tested in the same extreme use case, and the Seagate drives are more easily broken. If you take 1000 Ford sedans and 1000 Hyundai sedans and drive them off road, and more fords than Hyundais break, then the Hyundais are probably of better build quality.

 

So why do the Seagate crap out earlier than their direct competing consumer products?

Having a graphics card crap out is not a big issue. Having something with your data on it crash and burn is much more severe. Not all people have the possibility of backup up. In the class action law suit, the problem are 3TB drives, usually external, which IS the backup for most people.

 

The 7200.11 model is very relevant as those are the drives that makes the numbers go through the roof. All 3 TB seems to have issues too (and they are not 7200.11 models), but more across the board om most vendors.

 

Either way, plenty of people have given plenty of arguments of all the BS Seagate has pulled in the last decade or so, from entire lines of products being defective, to omission of responsibility, to poor customer service, to poor firmware solutions.

 

Seagate bought Maxtor, one of the worst HDD vendors out there in 2006. Two years later, the catastrophic 7200.11 series launched. Sorry but Seagate crapped all over itself after that.

My 80 GB 7200.7 and 250 GB 7200.10 still works and runs on my system still.

[iamdarkyoshi]

Sorry, but HAHAHAHAHAHAHAHAHA

Why do you think I moved to WD?

[Colonel_Gerdauf]

44 minutes ago, Notional said:

-snip-

 

I implore you to read over what @LAwLz has said once over. You may be correct with the method of data analysis, but the application of such data is outside the context of this topic. He has called you out on your erroneous generalization, which I have to agree with. Unless you have worked in a retail warehouse, your anecdotal evidence does not say a whole lot. You have essentially purchased a bad egg of the bunch, and you have negatively labelled the entire basket as a result.

 

Both Western Digital and Seagate have a few fantastic units and a few horrible units in their history catalogs. You can easily make a top ten list for both companies in both angles. However, on a general basis, when the sole deciding factor is failure rate, their drives are interchangeable. I personally prefer WD, mainly due to how the sub-brands are organized, but I do not necessarily make it a top priority. If I get a Segate drive in a system that I will use, then so be it.

[kelvindeschutter]

i have one of those seagate 4tb external drive after three years its still holding strong

[MrDynamicMan]

On 01/02/2016 at 1:07 AM, Aytex said:

step 1, stop being boneheaded about SSDs

step 2 make working products

Step 3,  delay your fade into irrelvancy by 6 months.

Ftfy.

[LAwLz]

35 minutes ago, Notional said:

Seriously you need to chill right now. Are you for real? Do you have financial investment in the company or something?

Nope, I don't have any financial investment but I really hate fanboys and people who are willfully ignorant.

 

36 minutes ago, Notional said:

Fact of the matter is that Seagate consumer drives and other consumer drives have been tested in the same extreme use case, and the Seagate drives are more easily broken. If you take 1000 Ford sedans and 1000 Hyundai sedans and drive them off road, and more fords than Hyundais break, then the Hyundais are probably of better build quality.

No, that's not how testing works. Let's look at it like this. If you were to buy a car and found out that BMW cars break more often than Mercedes cars when driven off road, but on road the BMWs actually hold up better, which card would you say is the most reliable when it comes to driving to work? Would you say the Mercedes is the most reliable because "it holds up better in scenarios it wasn't designed for" or would you say that the BMW is more reliable because it just flat out breaks less often when used for the thing most people will use it for, including you? According to your logic, the Mercedes is the most reliable despite it breaking more when driven on the road. According to actual logic, the BMW is the most reliable in the scenario most people will use it in.

 

That's the problem I got with you. You are ignoring statistics which disagrees with you, the statistics which proves you wrong and actually shows what normal people, people who actually bought the drives and use them in their own computers. You ignore all of those statistics, but you agree with the statistics of a company who use the wrong tool for the wrong job. What's next, you are going to say Linus' screw driver is bad because if you were to use it as a hammer it would break, while the screw driver you got would actually last longer as a hammer?

This famous quite applies very well to this situation. "if you judge a fish by its ability to climb a tree, it will live its whole life believing that it is stupid". That's what you are doing right now. You are judging a fish by its ability to climb a tree, when what people are actually looking for is who will win a swimming competition.

 

 

43 minutes ago, Notional said:

So why do the Seagate crap out earlier than their direct competing consumer products?

I don't have any definite answer, but what I do know is that there is a huge disparity between the statistics put out by BackBlaze and ACTUAL CONSUMERS. One explanation might be that Seagate drives are more sensitive heavy vibrations. That is not something they will be exposed to during normal consumer usage, but they will be when you put them in a SAN environment. If Seagate drives were more sensitive to that then that would explain why their consumers drives do very well (actually better than their competitors) for regular consumers but bad in BackBlaze's test.

Another explanation would be their very tiny sample size of some drives. Even BackBlaze says this about Toshiba and WD 4TB drives in their own article if you read it.

 

On top of that, we are only talking about 3TB drives here. If we look at for example 4TB drives then the numbers then the stats are very very different.

 

Then there is also the problem of BackBlaze's definition of "failed drive". A lot of the drives they say failed would actually be perfectly fine to use in for example your personal computer. They deem a drive "failed" when it does one of three things.

1) Don't start at all (this is what most people would say is a failed drive)

2) Doesn't sync in a RAID array (the drive might be perfectly fine, but your RAID controller might not like a particular type of drive and rejects it more often than other drives, despite it not being anything wrong with the drive itself. This is still considered failed in their test).

3) The S.M.A.R.T data goes below the thresholds BackBlaze has defined for themselves. This means that drives which accurately predicts when they are about to fail will get a worse rating than drives which just dies all of a sudden without any warning.

 

 

2 hours ago, Notional said:

Seagate bought Maxtor, one of the worst HDD vendors out there in 2006. Two years later, the catastrophic 7200.11 series launched. Sorry but Seagate crapped all over itself after that.

And even more recently they bought Samsung, one of the best (if not THE best) HDD manufacturers. Your point? You need to stop looking at the past and cherry picking products, and instead focus on their current product lineup.

Only a fool defines their entire view/opinion of a company on a small fraction of their product lineup.