News 71-TiB NAS with twenty-four 4TB drives hasn't had a single drive failure for ten years — owner outlines key approaches to assure HDD longevity

[Admin]

Tech enthusiast Louwrentius says that turning off the NAS when unused likely helped the 24 4TB hard drives last ten years without any failures.

71-TiB NAS with twenty-four 4TB drives hasn't had a single drive failure for ten years — owner outlines key approaches to assure HDD longevity : Read more

 

[bit_user]

The NAS owner said the 4TB HGST drives have only accumulated 6,000 hours since their deployment, translating to about 600 hours or about 25 days annually. Louwrentius turns off the NAS when unused, which the user claims is the likely secret to its longevity.

First, HGST drives of this vintage were really good.

Second, I had a fileserver with 5x 1 TB WD Black HDDs that also lasted over 10 years with zero unrecoverable sectors on any of the drives. As in his case, I turned it off when not in use. I used it mostly for backups and it probably had a similar number of hours as his.

Finally, I've seen machines at work with HDDs in continuous service for more than 10 years. I think one of them was even running for more than 15 years! So, it's possible to have old hard drives last that long, even when in continuous use!

 

[das_stig]

The Hitachi drives in my servers were all manufactured over 10 years ago and since I purchased them off ebay as ex-DC drives have been online 24/7 for nearly 2 years without any issues. THe call centre I used to work in doing support, had old HP desktops, that had drives 15 years+ and still running until we refreshed to W10. As bit_user said, drives back in the day just better quality

 

[ex_bubblehead]

He should consider himself lucky as most failures actually occur during power up where current draw is maximum and this is when things like spindle motors fail.

 

[konky]

Why mix units here? TiB to TB just to have to explain the size difference is foolish. Are you just trying to show of that you know the difference?

 

[thestryker]

So less than a year of powered on time which is not exactly how the vast majority of people would use a NAS. I'd be more curious about the number of times the drives were powered up than the length of time they've been installed for.

With my previous server box (24/7 operation) I had one drive actually fail with its original installation. The second set of drives I had a failing drive that I replaced before it failed. Neither one of which do I consider a very big problem as it was running 8 drives in RAID 6 so it was protected from drive failure. This would have been over a period of about 11 years.

 

[USAFRet]

Optimizing drive life vs usability of the NAS.

My NAS is on 24/7.
The 12 drives in or attached are between 4 and 15 yrs old. Various sizes and makes.

Only one has failed, an 8TB Toshiba Enterprise. 7 months old at the time of death. Its 4 year old replacement is just fine.

It is ON 24/7, because it is the movie and music repository, and receives nightly backups from the other house systems, and 24/7 video from the house security cams.

 

[bit_user]

Only one has failed, an 8TB Toshiba Enterprise. 7 months old at the time of death. Its 4 year old replacement is just fine.

When I upgraded my file server to 4 TB drives, one of the drives hit an unrecoverable sector immediately after RAID initialization. I think it was during the first consistency-check I ran. That was a WD Gold drive, apparently designed by the WD team prior to the merger with HGST completing.

I replaced it with another 4 TB WD Gold drive that was designed by HGST and not only was it faster and cooler, but also quieter! I assume the speed difference was due to it containing fewer, higher-density platters, since all drives had the same capacity and were 7.2 kRPM. It could complete a self-check in 9 hours, whereas the older, non-HGST drives took 11 hours.

 

[USAFRet]

When I upgraded my file server to 4 TB drives, one of the drives hit an unrecoverable sector immediately after RAID initialization. I think it was during the first consistency-check I ran. That was a WD Gold drive, apparently designed by the WD team prior to the merger with HGST completing.

I replaced it with another 4 TB WD Gold drive that was designed by HGST and not only was it faster and cooler, but also quieter! I assume the speed difference was due to it containing fewer, higher-density platters, since all drives had the same capacity and were 7.2 kRPM. It could complete a self-check in 9 hours, whereas the older, non-HGST drives took 11 hours.

That 8TB Tosh went from 0 to 14k+ bad sectors in less than a week.
RMA.

 

[bit_user]

That 8TB Tosh went from 0 to 14k+ bad sectors in less than a week.
RMA.

I never got around to RMA'ing mine. I had waited too long to even do the upgrade, to avail myself of Newegg's 30 day return period. So, that meant returning it to WD and receiving a refurb drive in exchange. Even though I was using RAID-6, I didn't want refub drives in my array, so I just bought a new replacement with my own money.

I still wanted to RMA it, just to punish them for shipping a bad drive. However, laziness overtook my indignation - a tale as old as time, I'm sure.

 

[NinoPino]

He should consider himself lucky as most failures actually occur during power up where current draw is maximum and this is when things like spindle motors fail.

This case prove at least that for HGST 4TB drives this is not true.

 

[ex_bubblehead]

This case prove at least that for HGST 4TB drives this is not true.

Just because it hasn't happened in no way indicates that it can't and won't happen in the future. It could happen the very next time or be years down the road. When the odds are non zero it means that it WILL at some time happen.

 

[USAFRet]

This case prove at least that for HGST 4TB drives this is not true.

Survivor bias.

This one case does not reveal all the other drives that DID fail after that many ON/OFF cycles.

 

[Rookie_MIB]

I've got a NAS with eight HGST 3 and 4TB drives (ZFS, 2 VDEVs, 4x3tb, 4x4tb, 20tb array), currently been powered on 204 days since last shutdown. No errors on the drives currently, I've had one fail several years ago, nothing since.

I actually have a few failed sectors on my Crucial boot drive which surprised me. Still, HGST drives are among the best IMHO.

 

[Amdlova]

IF the hdd hit 8000 hrs mark it's time to think of change it. Some disk don't tell right on smart and don't report.
I have one machine with five disk with raid 5... loose my sleep thinking if the motherboard dies. How i will build the raid again.

 

[USAFRet]

IF the hdd hit 8000 hrs mark it's time to think of change it. Some disk don't tell right on smart and don't report.
I have one machine with five disk with raid 5... loose my sleep thinking if the motherboard dies. How i will build the raid again.

That is why an actual backup is also needed.

A RAID 5 is but a single copy of your data.

 

[bit_user]

Still, HGST drives are among the best IMHO.

Of that vintage, they were. However, the group from HGST is also who pushed SMR into the mainstream at WD, I believe.

 

[NinoPino]

Just because it hasn't happened in no way indicates that it can't and won't happen in the future. It could happen the very next time or be years down the road. When the odds are non zero it means that it WILL at some time happen.

Yes, but also your affirmation that most failure happens during power up in no way indicate that is was lucky.
In my experience with servers, I never got a single HD broken during power up.
In the case of the article, he used 24 HD and for lot of years he power down and up the NAS countless times, this is indicative at least of the fact that power cycles are not so dangerous as sometimes claimed.

 

[ex_bubblehead]

A sample of 1 is not proof.

 

[bit_user]

In my experience with servers, I never got a single HD broken during power up.

Wow! You've never gotten the "click of death" upon switching on a computer? Or power up a server and see a fault light on a HDD that wasn't there before? I certainly have.

I can't say whether failures at power-on are more likely, but I think it probably depends a lot on how often you reboot/power-cycle. It's certainly true that the longer you leave your machine running, the greater the chance you'll have a failure before the next reboot.

 

[Amdlova]

The Samsung HDD long time ago. IF you change the disk from a machine to another they die instantly.
some of them die if just plug and power up the machines.
Thank God they sell the division to the seagate years later... Seagate Eat all the good brands of the market...
quantum>maxtor>samsung> and others I don't remember... Seagate = explosive highend tech
After the Flood on Thailand in 2011-2012 The HDD industy only goes down... chinese drivers are the worst.
The CMR and SMR scandall it's another hit. Today market is flooded with "FACTORY REFURBISHED".
I only get big drivers before see what is inside ou read tons of it.

 

[Christopher_115]

Tech enthusiast Louwrentius says that turning off the NAS when unused likely helped the 24 4TB hard drives last ten years without any failures.

71-TiB NAS with twenty-four 4TB drives hasn't had a single drive failure for ten years — owner outlines key approaches to assure HDD longevity : Read more

"TiB"

Stop. It's over. You've lost.

 

[bit_user]

"TiB"

Stop. It's over. You've lost.

Yeah, storage devices, link speeds, and clock speeds are pretty much always powers of 10.

RAM capacity and (usually) file sizes are the common places the power-of-2 "binary prefixes" are used. It's generally appropriate for anything that naturally scales as a power of 2.

https://en.wikipedia.org/wiki/Binary_prefix
​

 

[NinoPino]

Wow! You've never gotten the "click of death" upon switching on a computer?
Or power up a server and see a fault light on a HDD that wasn't there before? I certainly have.

Yes, but at power on only with client machines, never with servers or NAS.

Some servers needed power cycles 1/2 times per year, others like small NAS more often, but the "broken disk on power up" never happen. Instead I changed a lot of disks for degradation during use (always excluding PCs).

Survivor bias.

This one case does not reveal all the other drives that DID fail after that many ON/OFF cycles.

This is my experience, not statistically relevant, ofcourse, but equally statistically irrelevant is the comment that say it was "only luck" that 24 HGST 4TB are all in full service after many years.

I can't say whether failures at power-on are more likely, but I think it probably depends a lot on how often you reboot/power-cycle. It's certainly true that the longer you leave your machine running, the greater the chance you'll have a failure before the next rereboot.

I knows it's said that power cycle is highly stressing on HDD and I also disabled auto power off on frquently used NAS to prevent frequent parking of disks.
But I never see a real statistic that compares power cycles reliability between drive brands and segments.
So, who knows, maybe some disks electrically and mechanically manages the power cycles better than others.

 

[jeremyj_83]

When you buy enterprise drives, you expect them to last. At work we have a 16 bay NAS filled with 14TB WD Gold HDDs. In the last 5 years of being powered on 24/7 and writes to it all day at random intervals (log backups) we have had 2 drive failures. Therefore this NAS lasting 10 years isn't surprising.