Any ideas on what I should do?

[fincoder]

If you want to use SMART to diagnose a hard drive, look at RAW values only. If you do that then you could spot it is actually reallocating sectors for example.

That's what Crystal Disk Info does using it's built-in algorithms, and issues a CAUTION which I regard as faulty. With lots of experience and knowledge I suppose manual analysis of raw values might be slightly better.

 

[VISA MC]

I have a box of probably 300+ drives in our office of 1TB, 5400RPM drives that crystal disk info said were fine. We swap to SSD anytime we see 1TB, 5400 RPM and it has solved the issue 100% of the time. Everytime I see another one go by I think about starting a thread here or doing other research to learn exactly why that particular set of specs has such a big failure rate for us but get busy and forget.

As to the OP - If you can’t go on site I would ship the machine back and forth. We’ve done that many times for our customers that are too far to drive to for things like that. Actually just sent a MacBook to Rossmann in New York and back for motherboard repair.

 

[16bwhitt]

Okay guys, thanks for your responses.

It is an Acer Aspire T3-715A, so not a laptop.

Device name DESKTOP-FMUT7AD
Processor Intel(R) Core(TM) i7-6700 CPU @ 3.40GHz 3.40 GHz
Installed RAM 16.0 GB

Hard Drive is a Barracuda (ST2000DM001-1ER164) by Seagate, 2TB. HDD does run at 7200rpm, so it is a bit faster than the usual HDD's we're used to.

Hard Disk Sentinel shows the drive health as 35%. I repaired some bad sectors. Before it was at 27%. CPU is running at full speed (not underclocked due to performance). The computer is actually running normally as far as I can tell, no lag experienced so unsure as to what the client is talking about.

I am not 100% fluent with drive tech, does this imply it's failing? HDS says it is at an acceptable level, though I'd hardly consider 35% acceptable.

 

[frase]

Hard Disk Sentinel shows the drive health as 35% Eggad! Seagate not a fan of - always prone to fail in my opinion, though am not a HDD Recovery Specialist. Some of the other members here in that specific area of knowledge could point out the particulars.

 

[Porthos]

Hard Drive is a Barracuda (ST2000DM001-1ER164) by Seagate, 2TB. HDD does run at 7200rpm

I do not trust the DM series drives whatsoever.

 

[16bwhitt]

Looks like I am leaning towards a drive replacement somehow, will replace with an SSD. Of course, if I don't need to do this, I would be even happier if 35% health is indeed acceptable. The good news is they subscribe to my cloud backup solution, so they're covered.

 

[alexsmith2709]

Like @frase I am not a Seagate fan. I find windows 10 runs like sh!t on a spinner at the best of times, let alone one being reported as 35%. I would replace that drive with the SSD and im pretty sure your issues will be sorted

 

[nlinecomputers]

Hard Disk Sentinel shows the drive health as 35%. I repaired some bad sectors.

In my book ONE BAD SECTOR is enough to require replacement. 6th gen CPU so the system is probably 6 or 7 years old. I don’t think it supports a M.2 slot so you can only use a SATA drive.

 

[britechguy]

In my book ONE BAD SECTOR is enough to require replacement.

On massive drives this metric would result in the vast majority of them "requiring replacement" long before their useful lives are anywhere near over.

It is entirely normal for a few bad sectors to appear, slowly, over long periods of time on platter drives. One of the former system HDDs, now one of my backup HDDs, has had a single bad sector and only a single bad sector for at least 5 years. I've dealt with plenty of drives with a couple of bad sectors with no changes in the number over very, very long periods of time.

Bad Sectors Explained: Why Hard Drives Get Bad Sectors and What You Can Do About It
“A few bad sectors don’t indicate that a hard drive is about to fail — they can just happen. However, if your hard drive is rapidly developing bad sectors, it may be a sign that your hard drive is failing.”

What Is a bad sector and how can I repair it?
Bad sectors are fairly common with normal computer use and the imperfections of the world we live in . . .

 

[nlinecomputers]

And I disagree. On modern hard drives, hard bad sectors almost never occur and when they do it’s a sign of what will soon be rapidly failing hard drive. I’ve been bitten by failing to act on them and hard drives are too cheap.

 

[britechguy]

hard bad sectors almost never occur and when they do it’s a sign of what will soon be rapidly failing hard drive.

And, yet, my experience is directly at odds with this statement. So we will have to agree to disagree.

 

[Sky-Knight]

Hard bad sectors do not appear on modern drives, they haven't been a real thing since the original IDE drives!

They do however happen in bursts... So you'll get a bunch in the first month. After that, they should go quiet. Every time I've seen them reappear, it's in a cascade as the drive dies which is where the intermittent performance loss comes from... the drive is trying to fix these damaged bits instead of reading what the user wants.

But even if you detect this condition, the drive is still often dead before I can warn the user. So I gave up. I just build systems that do not require backup, and happily stopped using magnetic storage for everything that wasn't bulk storage.

P.S. The above links are flat incorrect, contact the manufacturer.

 

[britechguy]

Hard bad sectors do not appear on modern drives, they haven't been a real thing since the original IDE drives!

And we didn't get into hard versus soft bad sectors prior to now.

All I know, and I do know it, is that it's entirely possible for a modern hard drive to report a bad sector or two and stay just that way for extended periods of time, sometimes seemingly permanently. Probably a soft bad sector, true, but it reports as a bad sector.

It is the rate of acquisition that shows something's off. And if that rate of acquisition is glacially slow, nothing's wrong in any meaningful sense.

Those who refuse to do watchful waiting (which I have done, for a long time) don't see this "lack of drive death." The one bad sector and it's bad means you don't bother to try to do watchful waiting repeatedly over time.

Each of us should practice exactly as we see fit. But the idea that a couple of bad sectors means a hard drive is dying, when the rate of acquisition is so slow as to not even able to tell how long it actually was between development of the last bad sector and a new one unless you were *really* watchfully waiting is just wrong.

 

[nlinecomputers]

Hard drives are cheap, my labor is expensive, and data is irreplaceable. Maybe a drive with one bad sector is fine and maybe it will be totally dead next month. Why risk it? Hard drives are like tires whose performance deteriorates over time. If there’s any questions about the potential for drive failure it simply makes more sense to just replace it and be done with it.

 

[nlinecomputers]

Screw my opinion. What do our hard drive experts say. Calling @lcoughey ,@DataMedics.

 

[Blues]

In my book anything less than 50% drive health means it is time for caution and below 40% means recommending a replacement that is assuming there are issues with the system overall but generally I wouldn't be checking the drive health on a system with out a problem or one where any problem has been identified and fixed.

 

[lcoughey]

Hard Drive is a Barracuda (ST2000DM001-1ER164) by Seagate, 2TB. HDD does run at 7200rpm, so it is a bit faster than the usual HDD's we're used to.

Hard Disk Sentinel shows the drive health as 35%. I repaired some bad sectors. Before it was at 27%. CPU is running at full speed (not underclocked due to performance). The computer is actually running normally as far as I can tell, no lag experienced so unsure as to what the client is talking about.

I am not 100% fluent with drive tech, does this imply it's failing? HDS says it is at an acceptable level, though I'd hardly consider 35% acceptable.

Yikes! This model of Seagate is horrible for head failure and crashes. Any sector remapping may result in unnecessary irreversible data loss. But, I'd assume that the very first thing you did was to make sure that the data is 100% backed up, so that shouldn't be a concern. If not, tell them to power it off and get it to someone who can get the data off safely.

Either way, replace the drive.

 

[mikeroq]

If a drive fails short or long DST, smart is tripped, or has pink attributes in gsmartcontrol, it gets replaced.

 

[lcoughey]

If a drive fails short or long DST, smart is tripped, or has pink attributes in gsmartcontrol, it gets replaced.

...or if Seagate, replace.

 

[britechguy]

...or if Seagate, replace.

It really is funny, over time, how virtually every brand out there has a model (or several models) that are just complete dogs, while those that came before/after are fine.

It's also funny how many brands, that no longer exist as manufacturing entities, originate with the same actual manufacturers, which are few. Drives are much akin to retail department stores - they keep various names for the cachet of their histories but are mostly owned (or made, or both) by a couple of conglomerates.