Hard drive on way out?

[jcdenton]

Hi,

I have a Samsung F3? 1TB drive for media and some games and it has developed an issue where I can hear it spin up and whatever the PC (Windows 10) is doing seems to be delayed for a few seconds.

I have run some basic drive checking utilities which did not reveal any errors.

I have checked /tweaked some settings as well related to power saving to no avail.

Has anyone encountered this problem before?

TIA

[chuckskull]

Sounds like you might be right but something to check first, especially with SMART reading OK(I have seen a couple die like this but it's pretty rare in my experience); the cables/socket going bad(it happens) can cause the same symptoms. Swap them out. See what happens.

[satrow]

Check the stats with Hard Disk Sentinel, it might pick something up.

[DanceswithUnix]

I had a couple of Samsung drives fail on me, I wonder if they might have made an iffy batch as they used to be very dependable.

In my case, checking SMART stats showed some "pending sectors". If I see some of those, I would replace the drive. Heck, 1TB drives are quite cheap again, might be worth migrating the data to a spare drive anyway if you care about it.

[jcdenton]

Thanks for the suggestions, hard disk sentinel reports the following:

The hard disk status is PERFECT. Problematic or weak sectors were not found and there are no spin up or data transfer errors.

1,Raw Read Error Rate,51,100,100,OK,000000000460,0,Enabled
2,Throughput Performance,0,252,252,OK (Always passing),000000000000,0,Enabled
3,Spin Up Time,25,71,68,OK,00000000234C,0,Enabled
4,Start/Stop Count,0,96,96,OK (Always passing),00000000129C,0,Enabled
5,Reallocated Sectors Count,10,252,252,OK,000000000000,0,Enabled
7,Seek Error Rate,51,252,252,OK,000000000000,0,Enabled
8,Seek Time Performance,15,252,252,OK,000000000000,0,Enabled
9,Power On Time Count,0,100,100,OK (Always passing),0000000011D4,0,Enabled
10,Spin Retry Count,51,252,252,OK,000000000000,0,Enabled
11,Drive Calibration Retry Count,0,96,96,OK (Always passing),00000000101B,0,Enabled
12,Drive Power Cycle Count,0,97,97,OK (Always passing),000000000E82,0,Enabled
191,G-Sense Error Rate,0,100,100,OK (Always passing),00000000002B,0,Enabled
192,Power off Retract Cycle Count,0,252,252,OK (Always passing),000000000000,0,Enabled
194,Disk Temperature,0,64,51,OK (Always passing),0031000A0018,0,Enabled
195,Hardware ECC Recovered,0,100,100,OK (Always passing),000000000000,0,Enabled
196,Reallocation Event Count,0,252,252,OK (Always passing),000000000000,0,Enabled
197,Current Pending Sector Count,0,252,252,OK (Always passing),000000000000,0,Enabled
198,Off-Line Uncorrectable Sector Count,0,252,252,OK (Always passing),000000000000,0,Enabled
199,Ultra ATA CRC Error Count,0,200,200,OK (Always passing),000000000000,0,Enabled
200,Write Error Rate,0,100,100,OK (Always passing),000000000137,0,Enabled
223,Load/Unload Retry Count,0,96,96,OK (Always passing),00000000101B,0,Enabled
225,Load/Unload Cycle Count,0,98,98,OK (Always passing),00000000727E,0,Enabled

Will migrate files and maybe repurpose as off site backup if a format doesn't resolve it. I have checked the cables using another drive but probably worth trying a different port or power connection.

[satrow]

Looks pretty good to me, though I prefer screenshots ^^

Check the temperature tab, what are the Lifetime and Measured high temps?

Cable errors would usually be detected under 199,Ultra ATA CRC Error Count but that looks clean.

[jcdenton]

Temperatures seem ok...

[satrow]

Yes they do, difficult to tell from the base data: 194,Disk Temperature,0,64,51,OK (Always passing),0031000A0018,0,Enabled

Well, I don't see anything obvious, yet it doesn't feel like a soft error either. Have you run chkdsk /f /r on it yet?

[Agrippa]

I had a few of these drives myself and I can't remember experiencing anything like your issue.

How old is the drive? If memory serves the F3 was introduced in 2009 and if yours is getting on in years it might be time to think of a new one regardless of the fact that everything looks good. Sometimes there's just no kind of warning forthcoming from all manner of diagnostics before a drives goes and dies on you.

[watercooled]

Most of the Samsung drives I have have a fairly distinctive spin-up sound, but if it's definitely a new sound for your drive then I agree it's a possible cause for concern.

However there are quite a few other things which can cause disk access issues. E.g. were there any software or hardware changes leading up to this issue? New Windows update, new storage drivers, new power cables, moved power cables, etc?

Two of the most bizarre drive issues I've had include cheap-rubbish SATA power cables causing a HDD to appear as if it were failing (spinning down, read errors, SMART errors, etc), and another was when I had an optical drive on a secondary SATA controller - every time I accessed it, Explorer would hang for ages. In that case I just disabled that controller in BIOS and moved the drive to the chipset SATA ports but I suspect it was just a buggy driver.

[jcdenton]

Just checked and the drive was bought in 2011, so can't complain too much I suppose. I would say the problem has been happening for about 5 months, its strange as it doesn't feel slow or have any errors but the stuttering/delay can happen every few minutes or more (even when not seeming to access files on this drive).

Will try checkdisk, will also try formatting after migrating files.

Thanks for the help.

[watercooled]

You could also use something like SeaTools to perform a drive SMART self-test, where the drive basically runs a few internal tests, both electrical and mechanical.

[jcdenton]

Have now run checkdisk on restart, which took over an hour at 16% but then went straight to 100% ok.

SeaTools has passed the 'long generic test' as well so seems like the problem is not detected by any software.

[stevie lee]

have you checked whether its just the Disk indexing kicking in every now and again.

my drives that have it turned on, randomly spin up and sometimes lock the system for about a second.
they do it a lot less after I took all the data off, formatted and put the data back on again.

[Agrippa]

Disk indexing is certainly a possible culprit. I turned it off for this very reason, but that was so long ago I never even thought of it.

Sounds to me you've done just about every possible check for problems and that the drive has passed with flying colours, so the choice is to be paranoid (usually my action of choice) and replace the drive, or trust the results, put your feet up and relax (probably the more reasonable choice).

[watercooled]

Indexing shouldn't cause issues like this as it runs with low priority, but it's possible it's running more than usual if you have an unusually high number of files in indexed locations and/or something has gone wrong with indexing. You could try rebuilding the index, instructions here: http://windows.microsoft.com/en-gb/w...dexing-options

I wouldn't normally recommend just turning it of because, as I say, it shouldn't cause problems if it's working properly and is often extremely useful, allowing near-instant search results for files and their contents. This applies equally to HDDs and SSDs, as there was a myth going around that it was 'unnecessary' on SSDs, which is nonsense. But each to his own; if you know you definitely don't use it, and it seems like it's causing the problem and you don't want to try resolving it then turning it off is an option.

Back to the HDD itself: It's good that the long generic test passed, but did you run chkdsk with the /f flag or the /r flag? /r implies /f but also includes a full surface scan which would be why it took so long. But if it took that long with /f I'd be interested to see the log from Event Viewer as it should only take a few minutes on most drives.