|
Yeah, that's why I said I'm trying real hard not to get my prejudice to get into the way. But like I said, I cannot ignore the yearly Backblaze reports - where they put thousands of drives of all brands to real-world use.
If you've had nothing but good luck with Seagate - obviously I can't dismiss that and I can't tell you you're wrong.
From my position, I've had good luck with all the WDs I've been purchasing, so I'd have to have some really bad luck to get incentivized again to try another brand...
|
|
|
|
|
FYI:
The HGST drives were bought late October 2023 and early to mid November. The first failure was the first week of December, the second in January and the third about 2 weeks ago.
The Seagate drives were bought in March 2018 and the first failure was October 2023.
|
|
|
|
|
StarNamer@work wrote: The HGST drives were bought late October 2023 and early to mid November. The first failure was the first week of December, the second in January and the third about 2 weeks ago.
That's seriously bad. Do other systems also report them as dead? Have you tried other SATA ports?
Quite a few years ago I got a Sandybridge motherboard replaced (under warranty) because after a few months, SATA ports started disappearing. I believe there was an actual recall. I'm not suggesting the same thing might apply here, but it might be worth simply trying different ports. Who knows.
|
|
|
|
|
dandy72 wrote: Do other systems also report them as dead?
None of the drives were actually completely dead. In fact, in each case, the parameter which caused them to be reported as "FAILED!" was the Spin_Retry_Count, with one of the drives having a raw value of 1441811 when I replaced it.
I suspect that, once spinning, the drives would probably continue to work fine for months (years?) but since they were being reported as having the potential to fail within 24 hours and were all within their return window, I felt it was better to replace them rather than risk a more serious failure.
|
|
|
|
|
The listing didn't actually say they were refurbished drives. That was just assumed from the price. Since they came with a card which indicated 5 year warranty (which turned out not to be valid outside the US) I concluded the were actually new and just old stock.
I think I've only ever bought something like 20+ drives over the years, but the only 2 I'd had trouble with before the HGSTs was one Seagate (out of about a dozen over the years) and one unbranded drive which came preinstalled in a microserver I bought. It was only 250Gb so I moved it out and put it in an external enclosure which probably shortened its life!
|
|
|
|
|
There is at least one site that uses massive numbers of hard drives (thousands at least) and have been doing so for years.
They collect failures stats for all of those at an individual basis and make the results public. They have been doing that for quite some time.
Lots of detail.
Fun random read. And probably more relevant it you really need to get some real world stats on failure rates.
|
|
|
|
|
dandy72 wrote: Also - I'd never buy a refurbished drive. Ask yourself: What are the reasons anyone would ever send a hard drive back?
I would never buy even a new a drive from Amazon. I thought to save some money and bought 2 8TB drives from them (in 2022, IIRC). Both were dead on arrival. Buying the drives locally in Israel is a bit more expensive, but none of them have ever arrived dead.
(Sample of >30 drives over the last 20 years)
Freedom is the freedom to say that two plus two make four. If that is granted, all else follows.
-- 6079 Smith W.
|
|
|
|
|
StarNamer@work wrote: I'm wondering what his experience has been with them so far and how it matches with mine So, I guess not knowing is driving you crazy. Yeah I know, I'll get my coat.
StarNamer@work wrote: I'm beginning to think this may have been false economy and I'd have done better sticking with newer Seagate drives despite the price! Keep in mind, I haven't built a RAID in years. So, my experience is old and crusty, but back in the day Seagates were always known to fail before drives like from WD. I'm sure someone online will get upset and emotional about that, but whatever. Anyway, what time span are you talking here? Assume the warranty was still valid then did all these drive failures happen within a year? If so, that's crazy.
Also assuming prior to this HGST fiasco you didn't have drives failing like popcorn popping in the microwave, which would indicate a problem with your housing... maybe it overheats (which is a big problem), or you're setting your enclosure on top of a large speaker magnet for funzies, etc... Then you might be onto something.
If HGST was acquired by WD in 2012 then it's safe to assume they it was acquired with inventory. Given the fact that we had serious economic trouble in 2007-2008 and inventory can be manufactured a few years before it's actually sold to the customer (depends on the size of the company), or even if manufactured in 2012 maybe they started in 2008 being cheap and continued with it. So, in theory it's possible there was some cheap batches made you were unlucky to get. It's just conjecture though. Either way, might be time to try a different brand.
Side note, Google used to keep a list of which drive brands suck. They go through millions of them and they know. They refused to release that list though as it would effectively put that company out of business... even though in a real free market that can and should happen. Which is to say, the consumer is the last person companies care about these days.
Jeremy Falcon
modified 19hrs ago.
|
|
|
|
|
Jeremy Falcon wrote: Side note, Google used to keep a list of which drive brands suck.
Backblaze isn't shying away from that, and my conclusions seem to match their yearly reports.
RE: HGST...weren't they the ones that had a major flood at their manufacturing plant a decade+ ago, and subsequently had a huge batch of unreliable drives?
|
|
|
|
|
dandy72 wrote: Backblaze isn't shying away from that, and my conclusions seem to match their yearly reports. Respect.
dandy72 wrote: RE: HGST...weren't they the ones that had a major flood at their manufacturing plant a decade+ ago, and subsequently had a huge batch of unreliable drives? Dunno. I've been doing cloud everything lately, so me old and crusty with that. Never even heard of HGST until this post. It would explain a lot though.
Jeremy Falcon
|
|
|
|
|
I think HGST used to be a Hitachi brand, and I've purchased external drives from WD, and the drives inside had an HGST label. Even recent ones, so even though they might no longer promote the HGST brand (at least on the box), WD is clearly still using the name internally...
And I don't think I've had any sort of bad failure rate with the drives I have that I know are HGST.
|
|
|
|
|
They were and Hitachi bought that division from IBM earlier. The flooding happened in Thailand if I recall correctly.
"They have a consciousness, they have a life, they have a soul! Damn you! Let the rabbits wear glasses! Save our brothers! Can I get an amen?"
|
|
|
|
|
Weren't the IBM drives originally branded as "deskstar" which, of course, became corrupted to death star?
I've had reasonable experiences with Hitachi and Western Digital drives - no unexpected failures before they were replaced because they were either getting old anyway, or were just replaced for more capacity.
I did have to laugh when many years ago, having just bought and installed a Fujitsu Robin drive a "very knowledgeable" friend (a technical writer journo) posted a list of drives not to be touched with a dirty stick online, and the Fujitsu was top of the stack. That drive eventually got retired some 6 years later without a hiccup when the machine it was in was upgraded.
|
|
|
|
|
If you deep doin whatcha been doin, you'll keep gettin whatcha been gettin.
Definition of a burocrate; Delegate, Take Credit, shift blame.
PartsBin an Electronics Part Organizer - Release Version 1.3.1 JaxCoder.com
Latest Article: EventAggregator
|
|
|
|
|
In a closed society where everybody's guilty, the only crime is getting caught. In a world of thieves, the only final sin is stupidity. - Hunter S Thompson - RIP
|
|
|
|
|
Or a crackpot idea.
MSN[^]
|
|
|
|
|
|
It's a crackpot idea, don't get sucked in.
Definition of a burocrate; Delegate, Take Credit, shift blame.
PartsBin an Electronics Part Organizer - Release Version 1.3.1 JaxCoder.com
Latest Article: EventAggregator
|
|
|
|
|
|
Stop it with link only posts. (grumpy max)
At least copy the title or the summary if there is one.
CI/CD = Continuous Impediment/Continuous Despair
|
|
|
|
|
MSN[^] 18 Reasons Why Men Get Grumpier As They Age
|
|
|
|
|
|
touché!
CI/CD = Continuous Impediment/Continuous Despair
|
|
|
|
|
2 things:
(a) I see MSN still hasn't lost their habit of spreading their "[N] reasons why XYZ" type of articles over [N+M] pages (where M are just extra ad pages). So make that 19 reasons.
(b) Was there anything in there that was actually unique to men, and could not equally apply to women? Besides, aren't we're all supposed to be gender-fluid now?
|
|
|
|
|
Black holes generating enormous power has been around in science fiction for a while (Arthur C. Clark - Imperial Earth (1976)). More recent in the Star Trek Universe with the Romulan D'deridex-class warbird uses an artificial quantum singularity as a power source for its warp propulsion drive.
from the Wikipedia article ( Black hole starship - Wikipedia[^])
" A black hole weighing 606,000 metric tons (6.06 × 108 kg) would have a Schwarzschild radius of 0.9 attometers (0.9 × 10–18 m, or 9 × 10–19 m), a power output of 160 petawatts (160 × 1015 W, or 1.6 × 1017 W), and a 3.5-year lifespan"
That's a lot of power from a tiny space.
|
|
|
|