A statistical anomaly to say the least. If anything, two mirrored drives failing simultaneously would more likely indicate an environmental factor (humidity, temperature, power delivery, etc). Even with unreliable drives, the odds of failing at exactly the same time would be extremely unlikely. The most likely scenario of unreliable drives in RAID is one fails, you replace and rebuild the array, and the other fails shortly after.
1 Like
Wenn it comes to mechanical drivers, there isnt realy much to say about it.
In my opinnion its mainly a matter of luck.
In the enterprise world, there might be certain brands thats have been noted for having a higher failure rate then others.
But i still dont think that there is realy one is more reliable then others sorta thing.
2 Likes
For SSD's: mushkin samsung kingston adata
for hard drives: seagate hgst wd
If your seagate randomly died like that I would blame the power supply.
1 Like
I doubt it would be the PSU. I have a pretty good one: Rosewill (SuperFlower) Tachyon 650W
Honestly for years I have been afraid to play with a raid array I would much rather just have 2 separate drives
and manually copy/clone data
back in the day I would use as a Windows briefcase that keeps two separate folders synchronized
The only drives I've ever had die were a Seagate 1tb barracuda that clicked in 2 weeks, and a Samsung 500gb laptop drive that lasted for 8 years. I buy WD whenever I buy hard drives, as they have always treated me well.
Out of the 3 Seagate hard drives myself and family have owned, every single one has failed. Meanwhile, I have a plenty of old HGST/Hitachi and Western Digital drives happily chugging along, some being well over 10 years old. The Seagate ones all died within of purchase.
Maybe it was a fluke, but all I can say is that by my experience, I will never buy another Seagate drive again, nor can I recommend them to anyone.
1 Like
I think @MisteryAngel is right on this - it's more luck than anything. There hasn't really been any studies into the long-term reliability of consumer drives, and everyone's experience is different. I've used Seagate drives almost exclusively since I had my first "gaming" computer, Some Dell with a Pentium D 820, and a slapped on Radeon 4670. The few WD drives I've had were the problem ones. And reading up on reviews on WD drives, especially the Green, which seemed universally panned, didn't inspire confidence.
perhaps the RAID controller was dodgy it got replaced on a subsequent job but the point is the first thing that happened was both drives failing, not sure of the time frame between first and second drive failure, they may not have been monitoring the server very often and one failed and nobody noticed until the second failed is a likely possibility. but yeah no idea if it was hours or days or weeks or months between the first and second failing.
Alright everyone, I appreciate the healthy discussion on the topic!
I decided to take the dive on the WD 4TB Blue and hoping for the best. Kinda wish I forked over the ~$25 or so for the HSGT, but since both are WD, oh well.
I've got like 4 of the same model desktop consumer grade seagate drives on my failed pile it's not the first time either, they get flaky after a year or two of use and no way to really tell when they're going to go, 3 out of the 4 customers lost data because it was dead completely, internal failure before i got to it, one showed signs of impending failure so we copied the data to WD black and sent her on her way. I don't have any other brands of drive where I've got 3 or 4 of the same model drive that all failed after approx the same amount of usage except for maybe WD greens.
but having said that apparently the 4TB and 6TB seagates are showing better reliability than the 2TB and 3TB supposedly so yeah and environmental factors, movement/vibration, temperature, humidity etc. play a role too.
Go back 5-6 years seagate had a firmware glitch that they didn't pick up on until a lot of affected drives were sold, if the drives wrote past a certain point in their internal log while entering or leaving a low power state the drive would brick itself or something like that. It was random luck if it would ever happen to you, if it happened to a raid array they all brick at once...yayyy you can use a firmware fixed PCB of another same drive to get the data but it was a shit problem and it took a while before seagate was like ok we have a problem and issued firmware fixes.
Seagate/Samsung - highest chance of manufacturing defects or unreliable design.
Western Digital - not as bad but yeah, don't trust any hard drives, duplicate backups everywhere.
Hitachi - Almost 0 chance of manufacturing defects. (owned by western digital these days) They mostly live as long as they're supposed to if not longer, actually about 10 macbook customers in a row all had the 5 year old macs with hitachi drives only failing because they've endured 5 years of punishment, new HDD and clean the internals and maybe fresh thermal paste and it's as good as new. it's only the mac's made since jobs passed have starting using seagates a lot, because they're making an extra $20-$50 profit on the grossly overprices machines using cheaper and less reliable drives.
Toshiba - I think they're still made by hitachi but a cheaper spec drive than the hitachi branded ones, i could be wrong.
That deffect would explain my dead array...
yep if you've still got them the data is there, you just need like i dunno how ever many working PCB's you need to run them all at once again, not sure if seagate is still helping people that suffered that defect any more, all you can is ask, send them 8 dead PCB's and they should be able to send them back working if they want to.
Wish me luck! I suppose I need it...
Well my personal experiance, i have multiple drives seen dying from multiple manufacturers over the years.
This includes WDĀ“s and SeagateĀ“s, HitatchiĀ“s, Maxtors, Samsungs.
So to me personaly, i still think thats its just a matter of luck mainaly.
Aside from a few exceptions to that rule with drives that had known issues.
If iĀ“m not mistaken some certain BaracudaĀ“s and WD greens had some known issues.
If i have to pick one particular brand that i have seen die more frequently then the others, then i have to say Maxtors.
But Maxtors now days arent made anymore, since Maxtor was boaght by Seagate somewhere in 2006.
Still i do agree that the envoirement and usage circumstances had allot of influence on dying drives.
In large enterprise buisneses there will ofc be more often a drive failure then by a normal home user.
So wenn it comes to the enterprise envoirements, there might a recond of certain drives failing more often then others.
But this far to me, it looks mainly like a matter of luck.
If i have to nominate one particular drive to be one of the best ever made in history.
I probably going to say a Quantum Bigfoot.
But those arent made anymore lol.
they all fail sooner or later, 3-5 years is the expected lifespan of most drives made these days, but any of the dirt cheap ones from any manufacturer tend to have a greater chance of failing before they make it to 2-3 years, I'm sure a lot of failures I see are from someone treating an external drive or laptop like shit and it gets bumped and knocked too much or overheating from a system desperately needing a clean out but it's left there running screaming hot until something fails.
5400 RPM drives tend to last longer too don't they? if reliability is more of a concern than performance.
Also with those backblaze failure rates, keep in mind the drives they use cop a fair bit of workload, not just turned on 24/7 but reading and writing. So think of it as the drives that handle a flogging the best.
Yes definitely, i wont argue with that. āŗ
I was surprised at how difficult it was to find a decently priced 2TB HDD recently.
All of the cheap ones have a 2 year warranty or are <7200 RPM or possibly used workstation pulls
I didn't want to take another chance at a Seagate Barracuda drive as my 1TB one died just after the warranty period (3 years) - seems to be common according to NewEgg reviews
I decided to take a chance on an HGST Deskstar 7K4000 that was listed on Amazon. Manufactured in 2013 but never used. I'm praying that it last since drive failures are damn annoying
I'm using a 256GB Toshiba Q Series Pro for my boot and programs drive on my desktop. WD Blue for storage. Eventually I'd like to get a high capacity SanDisk Ultra II for installing games. I had a 640GB WD Black as my boot drive originally and it went bad on me, but it was also 6+ years old (I believe; it was used and most of the other hardware that the computer came with was about that old) so that's not surprising.
I'm running a 480GB SanDisk Ultra II in my laptop.
I had three of the Seagate 1TB SSHD's go bad on me in succession within a few years (they each lasted a about a year before either bugging out or completely hitting the bed). After that I swore never to go back to Seagate.
From my experience
Good:
- SanDisk / WD
- HGST / Hitachi
Bad:
- Seagate