ST3000DM001
en.wikipedia.org
en.wikipedia.org
"In the case of the Fireball SE, the mounting of the [power] socket was so flimsy that it was all too easy to break the PCB and thus ruin the drive. (With most other drives, you would have to be a gorilla and a drunken one at that.)"
To the other comments. We where a good customer that experienced an amazing stupidly high failure rate in a part that others also reported the same issue. The cost of deploying these systems and having them fail randomly broke planning and wasted the teams time and impacted our IT cost metrics. The cost of the support ticket, time and replacing and re-imaging a system was not free. We of course tried very hard to be preemptive in sorting this once we ID’ the issue. The vendor did not care, so another path was taken in order to help express our unhappiness to the vendor which resolve the issue. We paid additional money by paying for a brand name vs for white box with the idea that we would prevent this type of failure.
We've had to fry our failed HDDs because otherwise the vendor would replace them with those we've returned previously. Then they would not endure RAID rebuild and the cycle would repeat.
But you're supposed to hide the murder weapon.
Although the USA can fight back with a slow HFCS poisoning :-)
Unless it's a Creative labs product. You'll never RMA one of those. Ever.
It would be better to slowly raise the voltage until it blows, leading to minimal pyrotechnics on the components (smoke/burning)
https://en.wikipedia.org/wiki/Quantum_Bigfoot
A massive drive. I worked in computer retail at the time, and I lost count of how many customers brought their Compaq Presarios in, DOA, and one of those drives inside.
We'd sell them a new hard drive off the shelf, install it and send them on their way (they were out of warranty, of course).
After running a pair of probes hooked up to 110vac along the fingers of the VESA bus connection on the video card (and a few microprocessor chips exploded), the scorch marks were cleaned up and it was sent back one more time. We finally did get a new card that worked perfectly along with a long, vaguely threatening letter.
Sometimes the clearest path is through the mud.
"It's not flaky now."
The thing is, I knew a lot more than he did about how hard drives work, but he knew a lot more about how vendors work. A flaky drive might be hard to replace. A dead one wouldn't. Lesson learned.
Ended up having to kill the drives on a dedicated rig and send them back to keep our customers from having issues. They knew they were bad but refused to accept them back prior to failure. For a while they were returning a failing drive on the RMA too.
We bought some drives to rotate while we waited for RMAs, but the ones we got back failed too, so we just ended up keeping a stack of drives ready to swap.
They sat next to the spare dongles for the PCMCIA Ethernet cards, that the sales folks would inevitably break by tilting the laptop on it to get at a piece of paper on their desk.
That was a fun time being a small IT shop playing with WinNT and early Linux. We did everything from writing internal apps to running Ethernet.
3 guys in a glorified closet using post-its as our ticketing/user story system. Very IT Crowd.
The best deal we could find was buying these Seagate hard disks as external drives at retail from Costco. We picked up a truck load. Then had somebody with a screw driver remove them all from the enclosures and install them in new servers.
The error rate was high, the drives didn't last long in production.
Back then we referred to it as shucking the modems, so it's definitely not a new term in IT/computing (never mind the actual origin of the word).
Hence my final comment about the origin of the word pre-dating IT/technology.
(Wikipedia page on the facility in question: https://en.wikipedia.org/wiki/Utah_Data_Center)
Nobody is alleging that but you. Personally I don't think either "surveillance system needs a way to store what is collected" or "high-volume customers have economic leverage to demand confidentiality" are particularly intricate concepts.
To claim that the NSA took advantage of an existing event to help their already-planned ordering fly under the radar more easily is perfectly plausible. And, honestly, pretty boring.
Which one are you alleging? Choosing your words carefully might be wise.
https://archive.sltrib.com/article.php?id=2118801&itype=CMSI...
https://archive.thinkprogress.org/nsa-tries-to-keep-its-wate...
How many people work in Thailand's hard drive manufacturing plants? Hundreds? Thousands? How many people in the logistics channels? How many people are vending food to these employees? How many people in the US Government? Oh, and pulling this off would also require co-ordination with Thai Government. That's hundreds of people right there.
Not if you can discredit those who dare to speak out by labeling them "conspiracy theorists," thereby discouraging others from coming out as well.
I remember an episode from an old American sitcom where people make fun of the dad (an airline pilot) when he claims to have seen a UFO. It is the same thing. Any deviation from the standard narrative isn't tolerated by most people.
Only in a technical sense.
When people talk about UFOs, they are not referring to known things like weather balloons, planes or fighter jets, but, instead, to something out of the ordinary. However, every "standard" explanation offered is one of these things.
The standard narrative here is: UFOs don't exist.
That seems like a reasonable narrative considering that the definition of UFO you used basically means "aliens".
Frankly, we don't know, and one of the reasons is, I believe, people are hesitant to take a serious look at it (or even talk about it) because they don't want to be classified as kooks.
I think the rationally correct belief is that we simply have no evidence of aliens at this time, and since we virtually know absolutely nothing about what pilots saw (all we know is that there are these observations), we just fall back to our priors (base assumptions). Sure, it might be aliens. And hence the ancient aliens meme. It might all be secret military stuff. (Or the mix of the two, hence the storm Area 51 meme.)
And in this case I don't see what's wrong with the standard narrative.
If there are truly aliens flying regularly in our atmosphere we should strap a few 4K recorders on regular airline planes, and we just have to wait.
Not to mention the increasing number of space launches and spacecraft up there. And increasing number of downward looking observatories.
While I don't know anything about the theory being discussed here, the pervasive belief in Santa Clause among children disproves your point entirely.
So unless we all are living in a very well orchestrated NSA Truman show, it's unlikely that your argument applies.
"All right," said Susan. "I'm not stupid. You're saying humans need... fantasies to make life bearable."
REALLY? AS IF IT WAS SOME KIND OF PINK PILL? NO. HUMANS NEED FANTASY TO BE HUMAN. TO BE THE PLACE WHERE THE FALLING ANGEL MEETS THE RISING APE.
"Tooth fairies? Hogfathers? Little—"
YES. AS PRACTICE. YOU HAVE TO START OUT LEARNING TO BELIEVE THE LITTLE LIES.
"So we can believe the big ones?"
YES. JUSTICE. MERCY. DUTY. THAT SORT OF THING.
"They're not the same at all!"
YOU THINK SO? THEN TAKE THE UNIVERSE AND GRIND IT DOWN TO THE FINEST POWDER AND SIEVE IT THROUGH THE FINEST SIEVE AND THEN SHOW ME ONE ATOM OF JUSTICE, ONE MOLECULE OF MERCY. AND YET—Death waved a hand. AND YET YOU ACT AS IF THERE IS SOME IDEAL ORDER IN THE WORLD, AS IF THERE IS SOME...SOME RIGHTNESS IN THE UNIVERSE BY WHICH IT MAY BE JUDGED.
"Yes, but people have got to believe that, or what's the point—"
MY POINT EXACTLY -Hogfather
For a TL:DR of Making Money, there's
https://www.ft.com/content/1aebc9f8-70bb-11dc-98fc-0000779fd...
I'd still recommend the book though. There are so many things that are little more than flaky veneers taken for granted in adult human life that the fact anything works at all generally comes down to an intentioned group of people 'conspiring' to make it so, and no one ever bothers to ask about it, as it 'just werkz' and creates value. It's only a problem when too many externalities get ignored, in which case, everyone starts to notice.
Once you understand this, the beauty of the 'conspiracy theorist' meme coming out of an intelligence agency to get people to stop taking those who ask inconvenient questions starts to take on a stroke of genius. You get people who rock the boat discredited out the gate, and reinforce the status quo.
...Until that document gets declassified and blows up in your face anyway.
That would be similar to the way in which the fake Swiss cryptographic machine manufacturer was run. Their only leak risks were from employees who realized that their cryptosystem was easily broken. An HDD manufacturer would not have that problem because all of its work would be perfectly legitimate.
If everything in the world were an illusion, I doubt that any group would be able to orchestrate that for all of us. But large groups can coordinate secret happenings, and they do, all the time.
No, it proves it.
Default hypothesis: Santa is real.
For Santa to be a conspiracy: my parents and all my friends’ parents must be lying to their children.
Therefore: Santa is believed.
Just because a customer orders a few gazillion of product x and wants to keep it quiet doesn’t make some kind of nutty conspiracy theory. It happens all the time. The Thailand floods drew down production even more, but without the giant order, there wouldn’t have been the same impact.
Fun fact, any time two or more people want to keep something on the DL , it’s a conspiracy. Nearly everything that an intelligence agency does, is de facto, a conspiracy.
That seems... totally expected? The problem comes when you're trying to use it as proof of something. Some conspiracy theories do turn out to be true, but that doesn't mean every conspiracy is true.
Companies are uniquely positioned to minimise risk in their ability to execute.
I remember this period well and aware that a number of large players basically bought up whatever supply was available to cover their needs in the short to medium term. The result was that retail pricing more or less doubled (if you could even get drives).
So, just for the fun of it: Forbes estimates a capacity (at the time) of up to 12 EB [1]. Assuming they used 3TB drives, they'd need 4 million drives for this [2]. This article [3] shows the market for consumer drives at about 150 million in 2013 (figure 7). So the NSA order would've been in the ballpark of 2% of consumer drives sales - insanely large at the face of it, yes, but probably not enough to cause such a shortage.
[1] https://www.forbes.com/sites/kashmirhill/2013/07/24/blueprin...
[2] https://www.google.com/search?hl=en&q=12%20exabytes%20%2F%20...
[3] https://www.forbes.com/sites/tomcoughlin/2020/05/29/hdd-mark...
I know of people who were delivering software to certain customers that were required/intended to work on EB scale datasets over a decade ago, and this was well before the Utah data center was a thing.
There's all kinds of things to think about like how 2% may be less than 100% but in an elastic supply of a globally distributed good, 2% of (total yearly) supply called upon all at once can be a _huge_ departure from the norm. Maybe the shortage isn't just a hard disk shortage but a shortage of shipping space to take those hard drives anywhere else but to the NSA. Maybe it's that the factory(s) don't make 100% of the drives in an instant and then just distribute them for the rest of the year, but slowly make a constant stream of drives, all of which were swallowed up for a period of time by the massive order, causing the shortage.
There's a million other factors besides just "this number smaller that this number".
It looks like you've picked only the most favorable statistics and ignored everything else in formulating your opinion/response.
I really have no feelings for this theory either way, I simply don't think it's realistic.
> There's all kinds of things to think about like how 2% may be less than 100% but in an elastic supply of a globally distributed good, 2% of (total yearly) supply called upon all at once can be a _huge_ departure from the norm.
I generally agree with that. However, for one, we're talking about consumer goods within a growing marked - a demand jump of 2% should be within expected deviation and not generate the problems we've seen there.
> Maybe the shortage isn't just a hard disk shortage but a shortage of shipping space to take those hard drives anywhere else but to the NSA.
3,5" hard drives have a volume of ~ 390.000 mm^3. Let's say that's 3.000.000 mm^3 shipped. That would allow ~10.854 drives in a standard container, for a total of 370 (assuming 4 million drives). A large container ship can pack 850,000 containers. So yes, it's a large volume, but for global shipping it's a drop in the bucket.
> Maybe it's that the factory(s) don't make 100% of the drives in an instant and then just distribute them for the rest of the year, but slowly make a constant stream of drives, all of which were swallowed up for a period of time by the massive order, causing the shortage.
Sure they do. But, again: If we're assuming 2% of total drives, it would be the output for one week. Not ideal, but not a major shortage. And that's the worst-case scenario: Orders in the billions of dollars don't usually work in the time frame of one week; it's quite likely that the manufacturers knew of this well in advance and were able to upscale/order accordingly.
> Out of 150 million drives, there might not have been 4 million 3TB drives out there, this was 2014, maybe there was but they didn't want all the exact same drive, if you calculate that some of the drives are 2TB or 1TB the estimated numbers only get larger.
Sure. But 4TB drives and probably 6TB were also available at the time and it's reasonable that for extremely large storage, they'd shoot for larger drives - if only for space and heat efficiency. There surely were a few smaller drives as cache etc. here and there, but I'd presume the lions share to be large drives.
> I think it's totally reasonable that a single order of 12,000,000TB to a single location in Utah could grind the widespread consumer market to a halt for a bit,
I won't disagree with that it surely did impact the market a bit, but I highly doubt it was a major cause for the shortage. Especially when there's a perfectly fine explanation of a tsunami hitting where most drives are manufactured.
> especially if they're almost all made in one place, which it seems most of this discussion is taking for granted.
Most likely they'd order from multiple manufacturers, alone for capacity and failure probability reasons. In fact, if they actually ordered in one place, you could easily get evidence for this theory by showing the problems originating from one specific manufacturer.
> It looks like you've picked only the most favorable statistics and ignored everything else in formulating your opinion/response.
I highly disagree with that. I picked the largest estimate of the data center size and the capacity commonly available to consumers at the time; additionally, I only used a (low!) estimate for the size of the consumer market and completely disregarded the enterprise one. If anything, I probably overestimated the impact.
A said above, the massive buying surely did not help the shortage and I dislike the NSA as much as the next guy. But unless my estimation is off by an order of magnitude the numbers simply don't add up - even assuming the ordered with a delivery date of yesterday! - and given that there's a very reasonable alternative explanation for the shortage, I don't think this is a sound theory.
So is there any 'open secret in the industry' about the current chip shortage, that I'm unaware of?
If something is not indexed by Google, it's effectively non-existent. FAANG censorship is fast and effective. E.g. there were many important US bureaucrats involved in the Theranos scandal (even current US president). When I try to find anything related to that matter now, Google returns pages with many links removed by censors. It's almost impossible to find people responsible for withdrawing money from US budget via Theranos.
Also, assuming that 3 letter agencies were involved in the operation, any whistleblower would understand that destiny of Epstein will await her/him in near future.
Also, there's no "independent" journalism, all media belong to few influential entities, there's only approved articles appear in newspapers, magazines, on TV and YouTube.
If someone posted a plausible/verifiable account of what happened in a HN thread, then it would've made it halfway round the world before the NSA could take it down
So why didn't they bother censoring this comment?
Where did this idea come from?
https://journals.plos.org/plosone/article?id=10.1371/journal...
One thing I learned recently, which made the effect of the 2011 flood on the HD market worst: Aramco purchased a big chunk of several months of the world supply of hard drive to recover from Shamoon in 2012, by replacing every HD in the company [1]
Are you saying that there either was no tsunami or that the NSA created the tsunami as a cover to buy harddrives?
Why not just start a rumour claiming the price is increasing due to huge demand?
I was really careful - I had two NASes, each with raid 5 volumes holding a full copy of the footage, as it was irreplaceable.
I got up one morning to find a disc had failed in one of them. Swapped it. During the restripe, another failed. Oh well, I have the other NAS, I thought, and after replacing the busted disks and reformatting the failed one, started copying the data over again - and then a disk failed in the backup nas. And then another. And then another in the previously failed one. And that was that. Five disks, all dead, within a day of each other, all the same model. The WD drives in them were all absolutely fine.
I haven’t bought a seagate drive since, and never will again.
In many cases it's actually cheaper than buying the bare ICs at their normal price.
zpool history tank | grep "zpool create\|replace"
2013-03-26.23:27:40 zpool create -fo cachefile=/var/tmp/zpool.cache -o altroot=/mnt tank raidz2 gpt/disk0.nop gpt/disk1.nop gpt/disk2.nop gpt/disk3.nop
2014-01-18.19:00:41 zpool replace tank /dev/gpt/disk3
2014-01-19.09:59:45 zpool replace tank /dev/gpt/disk1
2014-11-30.14:44:59 zpool replace tank /dev/gpt/disk2
2015-09-13.21:16:11 zpool replace tank /dev/gpt/disk1
2016-01-03.15:37:40 zpool replace tank gpt/disk0TIL this exists. Neat!
The latter, if memory serves, was the bug where the drive had an internal event log that was a ring of 256 entries, and if it was at the last entry on boot, the hard drive's code on bootup for handling rolling back to 0 was (usually) broken, causing the hard drive to not start up any more. (I say usually because for this particular bug, regardless of whether I'm associating it with this model correctly, there were anecdotes of repeatedly power cycling the drives causing them to eventually boot normally, whereupon you could eventually apply the firmware update.)
If you look up that model, you can find people commenting on stories about the ST3000DM001 asking "so when's the class action for the ST31500341AS?"
edit: I see someone else in this discussion also remembers that family even better than I, because I forgot (or didn't update fast enough to encounter) the bug that was present in the first firmware update to fix this.
Seagate bought Connor in the middle of this and refused to honor the warranties. From what I recall a lawsuit followed and Seagate eventually capitulated.
Other HDDs on Wikipedia include: IBM Deskstar (equally infamous), and other historical milestones like the DEC RK05 (classic HDD for PDP-8 and PDP-11).
I just hope the upcoming Heat-Assisted (Seagate) and Microwave-Assisted (Western Digital) Magnetic Recording drives won't repeat history.
[0] https://en.wikipedia.org/wiki/Category:Hard_disk_drives
[1] Several Seagate product families have their Wikipedia article, but ST3000DM001 and ST506 are the only two specific models.
Decided one day to consolidate my backups from CDR to DVDR. Copied CDRs, deduped the contents and could fit about 15 CDRs worth onto one DVDR. Satisfied, I cut the CDRs and rebooted my system to prepare for a DVD burn.
The drive died on reboot.
Ended up taking a sledgehammer to the drive to get the anger out of my system.
https://i.imgur.com/Di9rQqu.jpg
https://i.imgur.com/zxrKyzE.jpg
Remarkably sturdy, that drive base was.
1) At one point we were experiencing disk failures often enough that we wrote a cronjob to detect drive failures via smartctl and automatically send an email to our hosting company requesting them replace the drive. This saved engineering time, and more importantly reduced time to drive replacement, because
2) On at least one occasion, we had the third drive in a RAID6 array fail before we had rebuilt from the initial failure, leading to loss of the array. We think the increased load of the rebuild increased the chance of subsequent failures. Needless to say recovering from this destroyed all plans (and sleep) for the weekend it happened.
Thats because servers know.... oh they know. Tuesday morning when you get into the office? Naw... that's too easy. Let's shit the bed on Christmas morning at 3:49am. That will let those stupid humans know who is boss around here.
I’ve never seen this posted publicly by them, but I’m fairly sure a revised firmware was offered. We deemed it too risky to upgrade the drives one at a time, so built a new cluster on Western Digital drives.
> On June 15, 2018, Judge Joseph Spero ruled that the class action plaintiffs must separate into multiple classes, as there was too much variability in failure rates to combine all claims a single class. In 2019, the plaintiffs were denied class certification a second time.
So basically Seagate got away with it.
Mmmm Fujitsu, the sweet sweet smell of conifer forest. The flux they used on those drives smelled amazing!
I worked at European Fujitsu distributor in late nineties. Absolutely the best non IBM(1) hard drives you could buy at the time. Cheap, fast, dead silent, super reliable. Then this happened:
https://www.theregister.com/2002/11/05/fujitsu_admits_4_9_mi...
https://www.dataclinic.co.uk/fujitsu-hard-disk-recovery/
“blame was laid on the supplier of epoxy mould compound used in the manufacture of Cirrus’ Himalaya 2.0 and Numbur chips”
Symptoms were drive not being detected, reporting as garbage corrupted strings, not spinning up or even clicking. The irony is they are 100% perfect mechanically. Btw I read somewhere swapping pcbs without making sure they have same firmware rev on board might result in service area corruption requiring actual specialist knowledge to recover (or running premade script in PC3000).
(1) and we all know how IBM Deskstars turned out ;-( Amazing and fast drives, until the sad conclusion.
Here's a list of terms for reference: http://data-recovery-tips.co.uk/wp-content/uploads/2013/09/h...
At some point, the bearing that serves as the pivot that the actuator arm and actuator attach to, is attached to the chassis of the hard drive with a bolt on the bottom of the drive. At that time, most drives also had a bolt that went through the cover. The Deathstar did not. It was only attached to the chassis. Which was fine, and that's the method commonly used since then. This particular drive had no thread lock on the bolt, and it had come loose, leading to a catastrophic head crash.
I don't know if other Deathstar's suffered the same fate for the same reason. The Wikipedia article on the subject says it was due to the magnetic coating of the platters coming off.
https://r-1.ch/hdd.mp3 (warning, loud)
https://soundcloud.com/ryan-flowers-916961339/sets/hard-driv...
The associated blog post: https://miscdotgeek.com/adventures-in-hard-drives/
We'd open the case of a crashed cartridge, extract the platter, and hang it on the wall; the crash would produce a rather spectacular sunburst pattern in the magnetic coating.
I always thought hard disk platters looked especially cool compared to other storage media like floppies or CDs or flash stuff.
So I submitted the article: https://news.ycombinator.com/item?id=27423640
Has nobody dissected it to figure out exactly which component no longer functioned correctly?
It's a bad sign when tech complexity gets out of the expertise of a hobbyist - I believe that's the start of a stagnation of that technology as new young engineers never get excited about disassembling and modding hard drives in their free time, and in turn it becomes paid engineers only who, I suspect, are less innovative.
When was the last time you heard someone who had modded their hard drive to, for example, seek faster?
People also make their own toy CPU's for fun with quite some regularity - using FPGA's and other tools, the same as the pros.
And people decap and reverse engineer new CPU's like Apples M1.
* install Linux to your hard drive firmware
https://spritesmods.com/?art=hddhack
The difference is, these hacks we've mentioned are software and solid-state electronics hack. On the other hand, doing a hard drive failure analysis in the mechanical realm is totally a different challenge, a hard drive is hard.
Firmware mods would be more doable, assuming it isn't locked down.
http://www.brendangregg.com/blog/2021-05-09/poor-disk-perfor...
For example, with the Challenger explosion, it's O-rings vs. the culture and organizational deficits that led to: the O-rings, failure to revise the seal structure despite evidence of problems on prior launches, and the final decision to launch Challenger.
All of that is arguably the most interesting part.
https://www.tomshardware.com/news/seagate-hdd-failure-lawsui...
> The Seagate 3 TB models failed at a higher rate than other drives during the Backblaze deployment, but in fairness, the Seagate drives were the only models that did not feature RV (Rotational Vibration) sensors that counteract excessive vibration in heavy usage models -- specifically because Seagate did not design the drives for that use case.
https://www.dtc.umn.edu/publications/reports/2005_08.pdf
> In the presence of an external vibration caused by adjacent disk drive seek operations the bandwidth performance of a consumer-grade disk drive “feeling” these vibrations will decrease about 10%-15% when reading data and about 25%-40% when writing data
Don't really know if this is enough to actually cause damage if the disk does not have the hardware to detect this vibration.
I don't know if the latency spike is caused by sensors detecting the extra vibrations and delaying a read or seek operation or if its just vibrations causing minor read-head errors which either require a 2nd pass (that isn't reported through to the OS via SMART) or some sort of extra processing to pull the correct signal out of the raised noisefloor.
Edit: person shouting at the disks is Brendan Gregg, not Bryan Cantrill.
https://www.recoveryforce.ca/forums/viewtopic.php?f=12&t=157
Most of them seem to be head problems, and there are some outright head crashes too. Seems like bad media (leading to head failure) is mainly to blame:
https://www.cheadledatarecovery.co.uk/2016/02/seagate-dm001-...
It sounds vaguely reminiscent of the IBM DeathStars, which failed in a similar fashion.
I would question this. Backblaze knows what they're doing. And at least they provide a relative reliability index as they treat all drives the same.
While they were still under warranty we replaced 50% of them and after warranty 100% of them were replaced using another brand.
I do make an effort to source them from as many different batches as possible, though - based on a DeskStar (“DeathStar”) experience a couple of decades ago. And there have been occasional bad batches for many models.
(The iirc 500MB DeathStar and this 3TB seagate are special in having mostly bad batches)
If you can, also stagger the initial power ons. There's a history of disk firmware bugs that are triggered by runtime; if it makes sense, you want to have enough of a difference in power on time to do a lossless replacement on your first disk before your second disk hits the magic number of time on.
Synology, which I’ve been using for the last 10 years or so when the project cannot afford NetApp/EMC class storage, does this out of the box. (I’m sure netapp and EMC do too, but it’s not my problem when the disks inside them fail)
After a couple years, I had multiple of them failing. Some of them were replaced by the shop I bought them at, some of them were replaced by Seagate directly. Every time, the data was lost.
At some point, when the 5th drive failed, I realized that the failure was due to a firmware issue, that could be fixed by upgrading the firmware using a TTL connection and a serial console. I did exactly that, my data was accessible again, and the drive never failed again (it's still running now).
I was very angry at Seagate and other vendors for sending replacement drives with the old firmware when it was known that the firmware would cause the problem again, while the new firmware was out and working correctly.
After that we switched to HGST and never looked back. For the thousands of Helium HGST/WD drives I've set up since 2014, I had to replace only a handful of them overall. In fact my support account generally needs a password reset because it's been sitting unused for 1 year straight... Contrast that with the awful time when I was spending half my days replacing Seagate drives, and the other half asking for RMAs...
I have never put any stock in this Ford vs. Chevy argument with hard drive manufacturers. Any of them are One model away from having a deathstar.
We proof the drives fairly extensively and burn them in for long periods before deploying them. That has Worked out well.
Toshiba took on a huge portion of Hitachi's former business, and seems to have a solid business. Backblaze doesn't buy Toshiba because the sticker is in the wrong location (and Toshiba doesn't want to move the sticker), but otherwise... their Toshiba tests look pretty good.
https://www.backblaze.com/b2/storage-pod.html
A sticker with the Serial Number on the correct location will help you find which of those 60-hard drives are broken. Toshiba didn't have their stickers in a location that worked for Backblaze.
On the other hand, you could just buy WD hard drives at virtually the same price and not have to worry about those labels at all since the sticker is in the correct spot.
Its not that its an unsolvable problem. Its that the solution is not worth the price difference between the drives.
A Seagate may (or may not! see Backblaze stats) die sooner than a WD drive, but at least it's fit for purpose until then.
I also personally bought at least one Quantum 105S, and then later "nobody every got fired for buying IBM" personally bought an IBM Deathstar drive. https://en.wikipedia.org/wiki/Deskstar#IBM_Deskstar_75GXP_fa...
Given my luck managing to find the worst lemon drives, for my subsequent home server, you can be sure that I intentionally used RAID mirroring of two different brands of drive. (Yes, it turned out that one make&model ran hot, while the other ran cool, and the hot one died.)
I'll never buy seagate again.
I retired them both earlier this year, replacing them with a single 2TB WD Blue drive to augment my SSD. They are both still fully functional, not so much as a SMART error.
I chalk it up to two things: The first is an extremely low number of power cycles. The second is that I never moved the computer thy were in until after they'd finished spinning down.
Once the bad drive was out of the pool, I ran a full self-test on it, which it failed. I did at least get five years of service out of it, but in any case, it's time for me to think about some newer hardware.
For my next NAS, I'm kind of leaning towards using 2x8TB mirrored, with a third drive to be rotated into the mirror, splitting off the rotated-out drive as an offline backup.
I was none too pleased when Seagate bought out Maxtor and turned it into their lowball brand.
It held up well for me, but I don't really use it much since it threw that error. SSDs seem better for reliability these days - the TBW numbers are big.
Every single one I have owned with any decent level of usage since the mid-90s has died on me after a couple years. I actually bought a 4 Disk Seagate NAS and all 4 Seagate drives have since been replaced with HGST drives one by one.
Exact same story here.
With the exception of one of the Seagates that did die really early (< 1 year) the rest of them had the same lifetime as my 2TB WD Reds.
I certainly wouldn't recommend them given their track record but it does seem this issue is a tad overblown.
Some anecdata: I have exactly two of these drives in my home NAS. The other two were WDC_WD30EZRX, one of them already failed, while the ST's are still going. All purchased in 2016.
As I recall that pretty much killed the IBM brand when it came to hard drives.
I'm much better about backing data up regularly now.
I guess I'll want to replace that soon.
Nobody ever seems to describe that other than saying vaguely that they use cheaper parts.
Oh boy.
- https://msfn.org/board/topic/128807-the-solution-for-seagate...
- https://www.anandtech.com/show/3482
- https://en.wikipedia.org/wiki/Seagate_Barracuda#Firmware_bug...
seriously, can't something be done about titles for URLs?
When people browse the list of titles on the front mpage, they should know what the story is about. So, in this case, something like:
"ST3000DM001 - A disk model number that would live in infamy"
or:
"Seagate's infamous ST3000DM001"
etc.
Who knows what it could be?
It sounds like one of those sci fi wall-e tier robots.
> Please don't do things to make titles stand out, like using uppercase or exclamation points, or saying how great an article is. [...] please use the original title, unless it is misleading or linkbait; don't editorialize.
Most of the time, this rule does more good than harm to prevent clickbaits, but it's not perfect and occasionally causes problem if the original title is not descriptive. In fact, there are multiple incidents that edited titles were reverted back by the mods, much to the complaints of the readers, I guess it's a small price to pay. Sometimes you can be sneaky so it won't be noticed, saying "Seagate's infamous ST3000DM001" is unacceptable, but the mods don't care if you say "Seagate ST3000DM001".
No, it was a cover for massive flooding that killed more than 800 people and led to most of Thailand being designated as disaster areas.