[LAU] Special Consideration for WD Green HDDs

Hans Wilmers hanswil at notam02.no
Thu Mar 14 22:43:20 UTC 2013


On 03/14/2013 04:06 PM, Ralf Mardorf wrote:
> On Thu, 2013-03-14 at 00:37 +0100, Ralf Mardorf wrote:
>> Until now I didn't test it myself, but since several people are annoyed
>> by WD's drive killer, it might be interesting for this list.
>>
>> In a German Arch forum,
>> https://bbs.archlinux.de/viewtopic.php?pid=300934#p300934 , I get the
>> hint that there is idle3ctl -d, see the English Arch Wiki,
>> https://wiki.archlinux.org/index.php/Advanced_Format#Special_Consideration_for_WD_Green_HDDs .
>
> The WD support didn't reply, but the moderator of the German WD forums
> replied.
>
> http://community.wdc.com/t5/Externe-Laufwerke/Sinnloses-Rauf-und-Runterfahren-der-WD-Elements/m-p/549265/highlight/false#M772
>
> It's said, that the EU forced that external drives have to spin down.
> It's said, that once they were spin down, they keep parked as long as
> there's no access.
>
> On all of my Linux the drive does spin down and up again and again, even
> if no partition is mounted, only GNOME 2 or Xfce 4 is running no
> application is running and I'm even not touching my computer.
>
> So IIUC this does mean that
> - Linux does touch in some kind the external drive
> - or if Linux doesn't touch the drive, it's broken and I should prove
>    the warranty claim
>
> I bought the drive 2 weeks ago and it happens since the first time I
> used it, with Suse 11.2 64-bit GNOME 2, Ubuntu Quantal 64-bit Xfce 4 and
> an updated Arch Linux 64-bit, with kernels from 2.6.x to 3.7.x, default
> kernels and rt patched kernels.
>
> Is it a Linux bug or a broken WD drive? How can I test it? Any ideas?

It might be the commit interval of the ext4 journal. To test, try 
remounting with option commit=60. Does it wake up the disk once a minute 
then? If yes, experiment with bigger commit intervals, and make it 
permanent in fstab once you are happy with it.

/ Hans


More information about the Linux-audio-user mailing list