I've got a WD10EACS-00Z V01.0 firmware 01.01B01 1TB drive on a Centos box, and the

hdparm -B 255 /dev/sda

command failed with the same IOFAIL error that others have gotten.

I discovered in another forum that:

[After, of course, yum install hdparm]

hdparm -S 242 /dev/sda

which sets the spindown timeout to 1 hour
now:

smartctl -a -d ata /dev/sda

shows Load_Cycle_Count stopped at 435228

Seems to survive reboots, so no need to muck with wdidle, which may or may not work on this drive ("We demand rigidly defined areas of doubt and uncertainty!").

[Edit: After a couple of days I'm up to 425500, so it's not completely stopped. I redid the hdparm command and will refrain from rebooting (I hope) and see if it's really stopped. Maybe it doesn't survive a reboot... Still, at only a hundred a day that's much better than the couple of thousand per day I was getting before.]

Wow! Either the head parking is disabled (noticed some people said theirs came predisabled) or you are giving it plenty of workload. Maybe with the AACS, you could disable the advanced power management altogether from smart tools. Or maybe, the idle timer isn't set at something as agressive as 8 seconds.. so the smart polling keeps the drive from parking (that's a guess).

Either way, that's way better than both my WD Blacks combined. Could also be WD masked the reading.. being a spcr reviewer, surely your PC is silent! You ever hear the "CLUNK" from head parking? That count may be the total amount of times you restarted that pc/nas.

In IDE mode the bios screen halts for a second or two on the line it detects my hdd. I can hear it do the unusual noises, maybe spinning up twice?

Anyone else seeing this behaviour?

Sent a question to WD support on the 18-th of December. Still no answer. The status of my question was elevated but if it was set automatically or not I don't know.

I got 4 x WD15EADS-00P8B0 all manufactured from Malaysia
for my HP Mediasmart EX495
1 x WD15EADS-00P8B0 Manufacturer Date Nov. 2009
3 x WD15EADS-00P8B0 Manufacturer Date Oct. 2009

I used the Nov. 2009 one as the main WHS drive
and on it the Load_Cycle_Count only change by 1 when I turn off the EX495 and turn on. ( right now 57 hours on and LCC is only 64)
but for the other 3 drive the Load_Cycle_Count changes like crazy
36hour on LCC 510
36hour on LCC 400
36hour on LCC 413

So I just used wdidle3.exe and disabled the head park on the 3 x WD15EADS-00P8B0 with LCC going up non stop. So now the LCC on them only changes by 1 when I turn off and turn on my EX495

all 4 x WD15EADS-00P8B0 had the head park set to 8000 millisecond by default from WD
I don't know why the one manufacture from Nov. 2009 wasn't giving me high LCC.

are the 3 x WD15EADS-00P8B0 from Oct. 2009 defective????
or the one from Nov.2009 head park is not working properly??

should I get the 3 x WD15EADS-00P8B0 from Oct. 2009 exchanged to Nov. 2009 or just leave it with head park disable?

Great thread guys! I just registered to say thanks - you have saved my sanity and my silent PC!

I just bought a WD Scorpio Blue 500Gb 2.5" drive to run in my silent HTPC. The drive is perfect, almost totally silent, with your ear right up to the drive itself you can only hear a faint fan-like noise when transferring data. Move a couple of feet away and it's so silent it could almost be a SSD.

...at least that was what I thought while I set it up, installed Windows 7 and loaded my iTunes collection onto the drive...

When I came to use the PC 'for real', I started noticing a loud clicking noise every 5-6 seconds, coming from the drive.

I'll skip the hours of searching the net I carried out to try to find the reason for this noise, followed by fruitless attempts to get a response form WD support (email reply within 24hours? not a chance...so far been 8 days and nothing!). Also, note I hadn't spotted the Load_Cycle_Count (which was up at 1250 for only 19 hours use ) until I found this thread - it's the noise that bothered me, but seeing that count would have made me return the drive if I hadn't found a solution.

Anyway I found this thread and tried the WDIDLE3.exe utility, both versions v1.00 and v1.03. Ran them off a USB DOS boot disk. I needed to disable AHCI in the bios to get them to work, otherwise they just popped up with drive not found error (v1.00) or hung the system (v1.03). With AHCI off, booting DOS and running the PC in IDE mode, the WDIDLE3.exe loaded and reported the drive as follows:

WDIDLE3 Version 1.00
Copyright (C) 2005-2008 Western Digital Corp.
Configure Idle3.
Model: WDC WD5000BEVT-22A0RT0
Serial: WD-xxxxxxxxxxxx
Idle3 Timer is enabled and set to 4000 milliseconds.

Wtf? The APM is set at 4 seconds on this drive! That's insane!!!

No wonder it's ticking like a clock!

I then tried to disable using "WDILDE3 /D". With v1.00, this seemed to work and both v1.00 and v1.03 report timer disabled. ...however when I reboot and load up Windows 7, there was absolutely no difference, it was still happily clicking away every few seconds. Using the same command with v1.03 does not report timer disabled - instead it says the timer is still set to some very high number... but again no effect in Windows 7. Setting the timer to various other values also had no effect.

So I turned to hdpram. As with WDIDLE3, I found hdpram only runs with AHCI disabled in BIOS, but it's a small sacrifice for silence. Then a simple:

hdpram -B 255 hda

...stops the clicking!!!

I then followed the excellent write-up by swivelguy2 on page 7.

Only needed to make a couple of changes to that method because I don't use a log-in password on my Win7 HTPC (it's a single user machine) and Win 7 complains that you can't set a scheduled task without password. If you're in the same situation, it's an easy fix - just don't tick the "run whether the user is logged in or not", and set the task to run "at login" not at startup.

And that's it, a perfectly silent HTPC is possible with this drive and Windows 7. All this headache to fix a 'feature' that makes a decent drive incompatible with modern operating systems...

Steve

Steve3000, Welcome to SPCR!


Thanks for posting your trials an tribulations, and bringing swivelguy2's post to my attention.
I just got a WD2500 2.5 blue and the "squirting" was driving mad. I had forgotten about this thread...now the noise is gone and cycle count is staying at 935. it got to that count in 4 days.



We need a more general thread on how to disable noisy WD park cycles, as its not GP's that are doing this.

Why is this even a feature?

I have just installed a new WD15EADS from Nov 2009 in my system, and the LCC count just went up to 2000 in 3 days.

I will look into the solutions (namely hdparm and wdidle from USB), but so far starting Speedfan with /NOSMARTSCAN seems to be solved the case. I will use it with /NOSMARTSCAN for a week and report back what's the result.

It's not just the WD Green that has this problem, but probably all the other 3.5" "Green" drives and pretty much every modern day notebook drive. I used to be worried about the head parking issue, but then realized in all this time, there is not even a SINGLE test/study done on whether the issue will have any measurable affect on HD lifespan. Everything has been just speculation so far.

I think it's a good feature, and as 99.999% of users doesn't use Speedfan for example, it works well for the average users.

And for all the NAS boxes/linux users/etc. It is easy to turn it off from wdidle 1.03, so it's quite safe.

I hope this question will not be too unwelcome as I have the same problem with a WD Scorpio Blue, but with a Playstation 3 rather than with a PC.

The insane frequency of the head parking causes playback skips when using video streaming services and I believe also causes games to hang and occasionally crash.

My question was if any of these fixes to disable the "power saving" head parking feature were permanent to the drive so that perhaps I could apply the fix to the drive hooked up to a PC and it would retain it's new settings when it is transferred back into the PS3?

I supposed I was thinking some sort of firmware change... I have looked high and low on the internet regarding this problem but it seems that only very few are aware of it and have no other fix than to force the harddrive to constantly run (with a silent mp3 playing in the background no less!) to stop the parking... it's so ridiculous.

At the very least, if there is no way to fix the WD drive, is anyone aware of a 2.5 laptop HDD that does not have this issue?

Thanks in advance, and apologies if my questions are not welcome.

The wdidle utility is supposed to do exactly just that. I read somewhere that it voids the warranty because it modifies the firmware by replacing a module. I don't know if its true or not though. You only need to run it once if it works. So doing it on another computer should work.

Thank you for the reply, I did a bit more reading on PS3 forums last night and discovered others had used WDIdle3 to change the firmware and it had indeed solved the problem for them, so I will give that a try. Many thanks!

Edit: Could you also tell me if backing up the harddrive is necessary? Does changing the firmware with WDIdle3 mean it will need to be reformatted?

If you would like to use wdidle, that's the best version/writeup I found on the net:
http://www.synology.com/support/faq_sho ... u&q_id=407

I've been keenly following the WD Green head park issue on the web and nowhere have I found a definitive result.

I want to buy a WD10EARS drive but this issue has me confused.

Shall I opt for the drive above or go for a Seagate LP HDD?

The 7200.11 firmware fiasco has left a bitter taste in my mouth as I was one of the many affected.

Can someone please help me choose.

I'm going to use the drive for storage only and speed is not important. Reliability is the biggest criterion.

They are absolutely reliable, IF the problem appears, you can either disable SMART monitoring, or switch your drive using wdidle.exe.

I had sent an email to WD Support regarding the LCC issue and this is their reply....

Subject
Head parking/unparking + Load Cycle Count Issues

Discussion Thread
Response (Paul B.)02/16/2010 07:11 AM
Dear xxxxxx,

Thank you for contacting Western Digital Customer Service and Support. My name is Paul B.

I apologize for the inconvenience this has caused you; because this is a Green Power Drive it will do more load and unloads than a normal drive for power savings. Although the counter of the drive will increase incrementally with each load and unload, this is only a counter and does not affect the drive. To date, we have had no reported failures due to high cycle and load times with any of our drives.

(Obligatory disclaimer: long time lurker, first time poster)

I've not read the entire thread thoroughly, so someone might already have posted a solution to this problem, but I just want to share the solution I found that worked for me. All credit goes to the user JunkMail over at the xtreamer.net forums where I found this solution after some googling.
The downloads total under a megabyte, so it's easy on bandwidth-time too



This worked wonders for me, the rate of the increasing LCC has dramatically been reduced after I did this, and the guide only took a few minutes to follow, and worked like charm, unlike some of the other guides I've tried.

Hope this post is a help to some of you.[/code][/url]

I have a WD15EARS drive. It's also has this head park and increasing cycle count problem after several days of usage.

Should I use WDidle3 to increase the park head wait time at this drive?

So is there any consensus on what to do if you have the problem drives and want to run Speedfan?

I have 4 WD10eacs drives for my HTPC and I am using speedfan to control fan noise. Can I turn off the smart feature on the drives? Should I? Any other fix?

Hi,

I joined just to post on this thread. I have 2 x WD20EADS that seem to have the high LLC problem too, I'm using them in a QNAP TS-239 Pro II NAS. They are used as single volumes, not in a RAID config.

Should I use WDIDLE on them? If so...

1. Is it recommended to set them to 300secs, or disable head-parking entirely?
2. Are there any disadvantages to having the drives not head-park in terms of reliability?

Thanks....

Hello,
First time poster here.

First I just wanted to say thank you to everyone that has contributed to this thread. It has been very useful in troubleshooting what was going wrong with my fileserver.

From the thread entries thus far it seems
1) confirmed that the Western Digital Green Power drives display very high Load_Cycle_Counts, especially when used with linux.
2) some posters agree that the Load_Cycle_Counts are high but believe that there is no conclusive evidence that this actually causes problems with the hard drives.

I have a couple of data points to add to this.
Both myself and a friend of mine purchased Western Digital MyBook World Network Attached Storage Devices. These devices use WD GP hard drives (although you probably wouldn't know this unless you opened one up, and opening it up voids your warranty).

Both I and my friend opened the units up in order to remove the stock OS and reinstall with Debian as described here:
www<dot>ismprofessional<dot>net/pascucci/documenti/mybook/mybook-hacking-en.html

Both units were running Debian & acting as media/fileservers, low volume webservers.

Both units began to experience intermittent problems after about 1 year. I did some diagnostics on my system & saw cryptic messages in the kernel log prior to system freeze ups. After reading this message thread, I tried digging into the smart messages for the hard drive with smartctl.

The Load_Cycle_Count for my drive was around 420,000 -- exceeding the manufacturers expected lifetime. I checked out my buddy's drive too. His Load_Cycle_Count was over 2 million.

I took my drive out, installed ubuntu on it and stuck it in another machine to act as a server. The intermittent problems were back & the OS locked up in under a day. I put the same ubuntu install on a spare hard drive (not a WD GP) and stuck it back in the testbed server. Smooth operation for weeks, no freeze up.

At this point I'm convinced that our WD Green Power hard drives are hosed, and I believe that the high Load_Cycle_Count is the culprit.

A couple of other things may be relevant:
1) The Mybook World units that we got these drives in use a Linux OS (although this is concealed from the user in most cases). It is possible that a lot of units in the wild are failing because of this problem (WD GP + Linux + high Load_Cycle_Count = shortened lifespan).
2) If the MyBook units are failing, most people probably wouldn't know why - the NAS's are supposed to be pretty much a blackbox to them. Most home users would not think of troubleshooting the hard drive to read SMART data & most wouldn't know how.
3) Western Digital may be aware of the potential for this problem in the Mybook's and may have avoided it by using a custom Linux kernel. This could be a non-issue if people are using the units as they are 'supposed' to.
4) Me and my buddy can't really go and complain to Western Digital because opening up the Mybooks voids the warranty, and we knew this when we did it (hey, what can I say, we like hacking).
5) If this is an issue for customers out there using the Mybook who did not hack their boxes, it might be impossible for them to isolate the problem without voiding their warranty. So they might be hosed & have no way to prove it.
(in fairness, they might be able to root their boxes as instructed here:
mybookworld<dot>wikidot<dot>com/first-steps-with-mbwe#toc0
then get smartmontools here:
ipkg<dot>nslu2-linux<dot>org/feeds/optware/mbwe-bluering/cross/unstable/
and get the diagnostic tools they need without physically opening the units, but I'm not sure if just rooting your box voids the warranty by itself)

Anyway, I just finished building a new server. When I was shopping for a new hard drive to replace the crippled WD GP, I deliberately avoided any Western Digital products. I got a Hitachi drive instead. Of course, it's possible that I'm wrong (I only have access to a sample size of 2), but I'm convinced that it is lousy firmware on the WD GP that killed my hard drive off long before its time. I shouldn't have to be conversant with the quirks of the firmware of a certain hard drive model; that is the kind of thing that should just work.

(shoot - won't let me post - I've got to go back & pull out my links)

you can try disabling head parking - maybe your drives will work that way.

at least, now we know the answer to the question raised by the title of this topic...

When i bought my two WD20EARS drives i set 1 hour spindown via hdparm -S 242 /dev/sdX, because i really wanted 1 hour spindown.

The result is :




In my case, it looks pretty normal.

One WD20EADS and three WD10EACS in my fileserver running Ubuntu for about a month now. The WD20EADS is running constant (seeding torrents) and the others are usually sleeping. No problems yet and I don't expect any.

The WD20EADS was my storage HDD in my primary system (running Windows) for a while, the others have been used less often with an eSATA docking station.

Could it be too fast headparking problem that make a WD20EADS disappear from a RAID5 on 9690sa. ?

It become visible again only untill that drive get power-cycled.

I've set them all to TLER 7sec wdidle3 at 4.5min , Should I set them to /d 64min ?

Unload idle is not same as TLER (Time-Limited Error Recovery). You need to use TLER application (if you can find one, and if it works on new WDs, or buy a RAID edition WD that has TLER disabled at factory. It has nothing to do with unload timer setting configurable through WDIDLE3.

It is the opposite, TLER is required to be enabled for hardware RAID.

I'm considering buying the WD10EARS. How audible is the sound it makes when it parks its heads?

I'm looking for a quiet drive but I've seen some people say these drives can make an annoying clicking sound during parking. I read through most of this thread but I didn't get an answer.

The head parking is inaudible. You won't hear it as the drive head moves through air to a rest position.

If you're worried about "clicking noises", don't be, those noises have nothing to do with the parking issue at hand. If you do run into lots of clicking noises in the usual running of your HDD, RMA/return it immediately. Usually, it's a possible sign of internal problems as it's a emergency retraction (this is if you've ruled out power being an issue).

I've been following this and many other threads of the same ilk. So far as I can concur:

1. The head park/unpark (aka Intellipark) affects the desktop line of Green label WD drives.

2. According to a WD response, Intellipark has no affect on the performance or lifespan of the hard drives. This was also confirmed in other threads by users whose drives had counts into the millions.

3. This park/unpark does not make any "clicking" noises and thus is unrelated to some reported clicking noises. (Like I said above, bad sign of possible drive failure, RMA immediately if you run into frequent clicking noises from your HDD.)

4. The desktop WD drives should not be used in any other RAID configs other than 0 and 1 (read: not 0 + 1, only to be used in RAID 0 or RAID 1) as they have the crucial TLER feature disabled. TLER is apparently required for RAID configs to function properly.

Given the potentially massive amounts of data involved in RAID configs, you should be investing in Enterprise level drives which do have the TLER feature enabled. Sure they cost more, but would you rather lose your data?

Given all that I now know, I'm not worried about the new WD15EARS I just ordered on Friday. I'm hoping though I won't get any frequent clicking noises as I have and do get from my current 7200.11 1.5 TB Seagate and my 1.5 TB Seagate Expansion drives. The fact that 400+ GBs of important data is severely at risk of being lost scares the bejeezes out of me. Add onto that fact that Seagate is being hush-hush about this is worse as I have not seeing any direct responses from Seagate engineers/tech support over at the Seagate official forums on this issue at all.

not having TLER doesn't mean one will lose his data - it means OS will freez while waiting for read from bad sector. yes, you'll have to reboot. no, you'll not lose your data - you'll be able to run degraded array or replace faulty drive.

btw, there's a way to enable TLER on most desktop drives, too:
http://forums.storagereview.com/index.p ... tler-cctl/