Hard Drive Response Delays - Windows Explorer hanging

[rcanino]

Hi folks -

Win7 Pro 64-bit / MSI Big Bang XPower X58 / 2 SSDs + 5 HDDs (1 is external USB) / HX850 PSU

The HDDs on the SATA II ports don't always respond immediately when I click on a directory or a file. It takes anywhere from a few seconds to ~20 secs. and then everything is fine. A few times (over 6 years) the system hangs indefinitely and I've been forced to reboot.

SSDs (SATA III ports) are unaffected; the external HDD is fine.
Problem HDDs are both on SATA II ports - two are WD Black, two are Samsung HJ 103SJ.

The trouble was always with just one HDD so I replaced it but it's still happening. Lately the problem seems to be spreading: my 2-month old WD Black FZ5001FZWX is the worst offender.

I don't defrag on a regular basis but I don't ignore it altogether. These delays are beyond a defrag issue; they're clearly caused by Windows Explorer hanging.

I configured new a power plan where everything is set to "never" turn off and that didn't help. It doesn't appear to be a hardware issue, unless the SATA II ports are slowly dying (though I've never heard of that happening).

I'm pretty much convinced it's a Windows Explorer problem - that is the service that hangs - but I haven't a clue how to address it.

Well...thanks for any thoughts / ideas you care to share. And thanks for a great website!

-Richard

 

[samuria]

Are the drives sata3 running on sata2 ports? its normal on first run as windows has to read all the disk content if you do it again it will have cached data so should be quicker

 

[rcanino]

thanks for your reply samuria -

Good thought, only this happens randomly and fairly often - I don't think loading the cache explains it. If it happened just once at the start of the day then maybe but this happens all the time. Not every time but often.

Seems to me something is causing Windows Explorer to hang.

 

[samuria]

Its a lot of info so any cache is short lived you didnt answer are they sata3 running on sata2 that wouldnt help some need a jumper to reduce speed

 

[rcanino]

thanks S - sorry i've been away for a few days -

the drives are SATA III capable - if that's what you're asking me - and they're running on SATA II ports, yes. I've never seen delays like this when running the exact same setup, i.e. SATA3 drives on SATA2 ports.

AFAIK you only notice slow downs with SATA3 SSD drives on SATA2 ports - mechanical drives like the ones I'm talking about are too slow to saturate even the SATA2 bus lets alone SATA3.

very strange behavior - it happens so randomly - I highly doubt it's cache-related.

Anyway thanks again for your replies - I appreciate it!

 

[GEWB]

Some drives go into a power saving type of sleep mode after a period of inactivity - they take a few seconds to wake and spin up when accessed in that mode. I have several spinning disks that behave just like that.

That's not to say it couldn't be a Windows related issue.

Try using a live Linux DVD to access your drives and see is the problem remains.

Regards,
GEWB

 

[rcanino]

Try using a live Linux DVD to access your drives and see is the problem remains.

Interesting. I have zero experience with Linux but I can't imagine how using a different OS will evaluate a hard disk(s).

Some drives go into a power saving type of sleep mode after a period of inactivity - they take a few seconds to wake and spin up when accessed in that mode. I have several spinning disks that behave just like that.
GEWB

this is certainly how it appears - as if the disks are sleeping. In fact I can duplicate the problem by using a power scheme that puts the disks to sleep after x-minutes. After x-minutes the symptoms are similar, except for the randomly-occurring part.

I've begun to wonder whether the High Performance scheme (or the HP scheme I configured, which is nearly the same as the stock HP scheme) is actually being implemented.

I ran a powercfg /energy command for 2 minutes and found nothing unexpected. Don't know what's going on but I'm sure these WD Caviar Black drives shouldn't automatically sleep. If it was "a few seconds" before they awoke I wouldn't be concerned but 5, 10 even 20 seconds surely isn't normal.

Well thanks for your reply G, I appreciate your help!

 

[GEWB]

You posted "I can't imagine how using a different OS will evaluate a hard disk(s)" - that is to work on eliminating the Windows OS as a cause rather than test the HDs. If under Linux the condition improves then I would start looking into Windows as a contributing factor.

Regards,
GEWB

 

[rcanino]

Ah - of course - thanks for the explanation. I think I will look into that idea GWEB - thanks again!