Anyone with true HDDs still around here?

[lexluthermiester]

Some WD Greens just had abysmal failure rates.
The ones I remember were the *EARS lineup.
I think *EARX was in there too but these were 1TB and 2TB models, so 500GB drives might have been spared.
The biggest issue with them was Intellipark that would aggressively timeout and repark the head every few minutes, leading to wear.
They were basically the kind of disk you'd want in non-RAID jobs and keep them as nearline/offline backup media.
24/7 duty even when disabling parking didn't save these from going bad.
I kept mine in the rack when it housed the Pentium 4. The sector test looks like a round of Space Invaders.

View attachment 364348

WOW is that a lot of bad sectors!! That's a done drive.

Also, what the hell is with that Windows theme?

 

[64K]

Never had a HDD or SSD die on me. The only odd experience I ever had was a WD Caviar Black HDD that was unusually noisy with the clicking but iirc that was normal.

 

[lexluthermiester]

Never had a HDD or SSD die on me. The only odd experience I ever had was a WD Caviar Black HDD that was unusually noisy with the clicking but iirc that was normal.

You should consider yourself very lucky! For someone with as much experience as you not to have ever had a drive fail on them is very rare.

That said, statistically, it's bound to happen sooner or later so I recommend you employ some form of backup for your important data, which IIRC you do, but it's worth mentioning.

 

[DaemonForce]

WOW is that a lot of bad sectors!! That's a done drive.

The funny thing is I was still able to partition around it for a fair amount of scratch space but yeah, it's a brick.

Also, what the hell is with that Windows theme?

Windows Server 2012 defaults. R2 might have been the same. This must have been back when I was making the jump from 4-8GB ram, which was only 7 years ago.

Never had a HDD or SSD die on me. The only odd experience I ever had was a WD Caviar Black HDD that was unusually noisy with the clicking but iirc that was normal.

SSDs fail in a very sudden and non-spectacular way.
The first failure I had was a 30GB OCZ Vertex II and one day without any warning the controller went poof.
No more SSD detected in the bios, in other machines, just gone.
I think SSDs give a bit more warning these days since we rely on much better controllers but back in 09-10 they were real bad.

 

[lexluthermiester]

The funny thing is I was still able to partition around it for a fair amount of scratch space but yeah, it's a brick.

I've done that myself with drives that had some bad sectors but were otherwise fine. Just calculated the sector spacing and partitioned around the bad sectors. Still have one of them, 120GB 2.5" IDE in a USB enclosure. Still works fine, just minus 1GB worth of space and divided up into two partitions.

 

[TheLostSwede]

4x 4 TB Toshiba N300 in my NAS. No spinning rust in my desktops for years now.

 

[lexluthermiester]

4x 4 TB Toshiba N300 in my NAS. No spinning rust in my desktops for years now.

A lot of people are doing that now that 10gbps networking is a thing. The network latency isn't as much of a bandwidth limit as it once was compared to the SATA 6gbps.

 

[64K]

You should consider yourself very lucky! For someone with as much experience as you not to have ever had a drive fail on them is very rare.

That said, statistically, it's bound to happen sooner or later so I recommend you employ some form of backup for your important data, which IIRC you do, but it's worth mentioning.

I think I have been unusually fortunate with hardware. In almost 45 years of gaming I have never had a problem with storage of any type all the way back to a cassette tape drive on my C64. I am a believer in backups anyway. I even go to the extent of keeping my important files on USB drives in my bank SDB. I did lose one USB drive that got zapped somehow.

SSDs fail in a very sudden and non-spectacular way.
The first failure I had was a 30GB OCZ Vertex II and one day without any warning the controller went poof.
No more SSD detected in the bios, in other machines, just gone.
I think SSDs give a bit more warning these days since we rely on much better controllers but back in 09-10 they were real bad.

The longest I have ever used a SSD is on my gaming rig which has been in use for several hours just about every day for 7.5 years. I have a SSD in my laptop that is 5 years old.

Maybe it's time for me to contact Guinness. I might be eligible for some award.

 

[Ruru]

Some WD Greens just had abysmal failure rates.
The ones I remember were the *EARS lineup.
I think *EARX was in there too but these were 1TB and 2TB models, so 500GB drives might have been spared.
The biggest issue with them was Intellipark that would aggressively timeout and repark the head every few minutes, leading to wear.
They were basically the kind of disk you'd want in non-RAID jobs and keep them as nearline/offline backup media.
24/7 duty even when disabling parking didn't save these from going bad.
I kept mine in the rack when it housed the Pentium 4. The sector test looks like a round of Space Invaders.

View attachment 364348

EARX here. Need to backup my music folder like I promised to @lexluthermiester

 

[Apocalypsee]

A couple of 4TB HDD still in my main rig. SSD isn't that cheap enough for mass storage. I'm storing old TV shows/cartoons that hard to come by. I don't like streaming either

 

[lexluthermiester]

EARX here. Need to backup my music folder like I promised to @lexluthermiester

Don't do it for me, do it for yourself!

A couple of 4TB HDD still in my main rig. SSD isn't that cheap enough for mass storage. I'm storing old TV shows/cartoons that hard to come by. I don't like streaming either

Right there with you!

 

[chrcoluk]

Then why are all my previous HDDs dead after 5-6 years when three 850 Pro's I got in 2015 are going to be 10 yo soon and work perfectly ?
I have nothing against HDDs, they're very useful for backups and data storage, but not as reliable. Unless you mean some server dedicated HDDs, then yes. I guess you need a good SSD to last you a decade or more too, so it's always down to the quality of the drive you're buying, be it HDD or SSD. Crappy HDD will fail faster than a cheapo SSD though.

Feels like you doing something wrong.

I cool all my HDDs and historically I havent let them spin down, so if using windows the first thing you should be doing is disable the hdd spin down setting. Even better never turn the system off. If you do let them spin down set the idle quite high so its in hours rather than minutes to prevent yo yo states, and also decrease head parking, I prefer to make head parking less aggressive rather than completely disable it.

 

[agent_x007]

What does that do

Shows values in 0-9 range, instead of [HEX]

 

[lexluthermiester]

so if using windows the first thing you should be doing is disable the hdd spin down setting.

This, yes.

If you do let them spin down set the idle quite high so its in hours rather than minutes to prevent yo yo states, and also decrease head parking, I prefer to make head parking less aggressive rather than completely disable it.

This is also something worth doing. It what I do. My HDDs power cycle is set for 210minutes.

Even better never turn the system off.

This, no. Windows needs to be fresh booted at minimum once a week, if not every day or two. Windows is more stable than it once was, but it's still not perfect.

 

[DaemonForce]

Windows needs to be fresh booted at minimum once a week

Nah.

Once/Twice a season maybe. 2016 Core and Nanoserver behave well enough that power outages are the concern.
It's probably something to do with the way Windows manages memory as this is before the threshold of what we have for Win10/11.
I'm quite happy with the result but I also haven't shifted this box into full time 24/7 distribution duty.
I can expect CPU spikes in the 60-70% range and all manner of HDD chatter throughout the day without Steam or Epic spun up.

 

[Zazigalka]

I don't use my main i7+4070S rig daily, I bought a 7730u laptop for daily work and entertainment. PC only works for gaming sessions. If a drive cant handle that for 5 years without developing issues, then it's not worth anyone's money.

 

[Harlow]

What I dread, are the "shingled-magnetic-recording" HDDs!

I steer well clear of SMR drives in all of my TrueNAS Core arrays and check even more carefully after the WD Red fiasco in 2020.
https://www.servethehome.com/wd-red-smr-vs-cmr-tested-avoid-red-smr/

In the the case of FreeNAS and TrueNAS, truly horrible things can happen when it comes to re-silvering an array of SMR drives. Instead of taking 3 to 5 hours, it can take 3 to 5 days, with the attendant risk of another drive going down.

I'm sure most of my 3.5in USB3 desktop drives are SMR, but they're only used for off-line archives, so speed isn't important. For backups, I use 800GB LTO4 tapes.

At a guess, I'd say I have at least 70 working hard disks (three 8-drive and one 6-drive NAS) plus various tower systems with up to 10 hard disks each and 11 external USB3 drives. It sound excessive, but you can never have enough backups and I create 600GB+ in RAW+JPG photos on each vacation, let alone 4K video.

 

[chrcoluk]

This, yes.

This is also something worth doing. It what I do. My HDDs power cycle is set for 210minutes.


This, no. Windows needs to be fresh booted at minimum once a week, if not every day or two. Windows is more stable than it once was, but it's still not perfect.

Oh still reboot for updates and stuff of course, but rather I meant HDD's like been powered on all the time, reduce those cycles.

 

[lexluthermiester]

Nah.

View attachment 364352

Once/Twice a season maybe. 2016 Core and Nanoserver behave well enough that power outages are the concern.
It's probably something to do with the way Windows manages memory as this is before the threshold of what we have for Win10/11.
I'm quite happy with the result but I also haven't shifted this box into full time 24/7 distribution duty.
I can expect CPU spikes in the 60-70% range and all manner of HDD chatter throughout the day without Steam or Epic spun up.

Fair enough.

 

[Solaris17]

I never had any problems with the Green line either. Granted, I've only seen a few of them, but only once was one dead.

There was a real problem with them, but it was like a decade ago. I think people just hang on. Though I personally wont run a green because of its performance characteristics they havent dropped like flies in some time.

The biggest issue with them was Intellipark that would aggressively timeout and repark the head every few minutes, leading to wear

yup destoyed them. I remember flashing a few to blues back in the day to prevent premature failure when I came across an affected model

 

[lexluthermiester]

I remember flashing a few to blues back in the day to prevent premature failure when I came across an affected model

Didn't know you could do that! Nice. Did that work?

 

[Solaris17]

Didn't know you could do that! Nice. Did that work?

Yeah it did! Atleast back when this saga was happening the only difference was the firmware regulating when the head was parking and the drive was spinning down.

 

[lexluthermiester]

Yeah it did! Atleast back when this saga was happening the only difference was the firmware regulating when the head was parking and the drive was spinning down.

Cool! Had no idea. I generally stuck/stick with the 7200RPM series and avoided the slower stuff until recent years. 5400rpm drives are surprising speedy in the last 6 or so years.

 

[A Computer Guy]

Some WD Greens just had abysmal failure rates.
The ones I remember were the *EARS lineup.
I think *EARX was in there too but these were 1TB and 2TB models, so 500GB drives might have been spared.
The biggest issue with them was Intellipark that would aggressively timeout and repark the head every few minutes, leading to wear.
They were basically the kind of disk you'd want in non-RAID jobs and keep them as nearline/offline backup media.
24/7 duty even when disabling parking didn't save these from going bad.
I kept mine in the rack when it housed the Pentium 4. The sector test looks like a round of Space Invaders.

View attachment 364348

I remember the EARS ones coming with their entry level NAS boxes WD MyBookLiveDuo. By the time they released MyCloudMirror I think they switched to WD Reds.

 

[RJARRRPCGP]

When shipping them around to wherever they're marked FRAGILE for a reason.

"Fraaahgeeelayyy"!

WOW is that a lot of bad sectors!! That's a done drive.

Also, what the hell is with that Windows theme?

It at least almost looks like a dropped HDD! Yikes!

I think SSDs give a bit more warning these days

Actually, faulty SSDs, have the propensity to suddenly corrupt system files out-of-the-blue, possibly before you can even see the signs from the SMART.