From: Tejun Heo <tj@kernel.org>
To: Pascal Vandeputte <pascal@asmodeus.be>
Cc: linux-ide@vger.kernel.org
Subject: Re: frequent disk head unloads under Linux: additional hardware
Date: Wed, 01 Oct 2008 04:28:13 +0900 [thread overview]
Message-ID: <48E27DCD.2040904@kernel.org> (raw)
In-Reply-To: <48E144F7.7070303@asmodeus.be>
Pascal Vandeputte wrote:
> Tejun Heo wrote:
>> Hello,
>>
>> Pascal Vandeputte wrote:
>>
>>> Hi,
>>>
>>> I've been highly frustrated for more than a year about this problem with
>>> my HP nx7400 laptop and its Seagate Momentus 5400.2 disk (which is
>>> otherwise a really fine piece of equipment). By switching from Ubuntu to
>>> openSUSE today, I had to look for the Suse way of fixing this up
>>> (instead of editing /etc/hdparm.conf on Ubuntu), and I stumbled upon the
>>> storage-fixup script which is really nice!
>>>
>>> As my harddisk isn't included in by default yet, I've attached the
>>> required information to this e-mail.
>>>
>>> Adding this to /etc/storage-fixup.conf solves it:
>>>
>>> # Reported drive model: ST9100824AS
>>> rule hp-nx7400
>>> dmi system-manufacturer Hewlett-Packard
>>> dmi system-product-name HP Compaq nx7400*
>>> ata model ST9*AS
>>> act hdparm -B 255 $DEV
>>>
>> How much does the Load_Cycle_Count increase under idle usage per hour
>> without the adjustment?
>>
> About thrice per minute:
>
> # while : ; do smartctl --all /dev/sda | grep ^193; sleep 60; done
> 193 Load_Cycle_Count 0x0032 037 037 000 Old_age
> Always - 126856
> 193 Load_Cycle_Count 0x0032 037 037 000 Old_age
> Always - 126859
> 193 Load_Cycle_Count 0x0032 037 037 000 Old_age
> Always - 126862
> 193 Load_Cycle_Count 0x0032 037 037 000 Old_age
> Always - 126865
> 193 Load_Cycle_Count 0x0032 037 037 000 Old_age
> Always - 126867
> 193 Load_Cycle_Count 0x0032 037 037 000 Old_age
> Always - 126869
>
> I reached the first 120000 cycles within 10 months of moderate usage
> (it's my personal laptop, not used for my day job).
>
> It occurs in Windows XP as well, but not as frequently.
>
>>> An unfortunate side-effect of meddling with the power saving on my
>>> system occurs at shutdown: for some reason I can clearly hear the drive
>>> spin down and then suddenly spin up and initialize again before the
>>> power is killed... This doesn't happen if I leave everything at default
>>> settings. Restoring the default setting (128) manually before shutting
>>> down doesn't prevent this either. It's weird and I've been working
>>> around it by rebooting and using the power button when I see GRUB appear.
>>>
>> That's a known problem of certain HP laptops. Please take a look at the
>> following bug report.
>>
>> http://bugzilla.kernel.org/show_bug.cgi?id=8855
>>
>> Thanks.
>>
> Thanks for the pointer, I have already submitted my own data in that
> topic in the mean time. I hope that someday I can install the latest
> distro without worrying about hdparm tweaks :-)
>
> Greetings,
>
> Pascal
added
--
tejun
prev parent reply other threads:[~2008-09-30 19:29 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-09-07 20:54 frequent disk head unloads under Linux: additional hardware Pascal Vandeputte
2008-09-29 7:27 ` Tejun Heo
2008-09-29 21:13 ` Pascal Vandeputte
2008-09-30 19:28 ` Tejun Heo [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=48E27DCD.2040904@kernel.org \
--to=tj@kernel.org \
--cc=linux-ide@vger.kernel.org \
--cc=pascal@asmodeus.be \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).