On Sat, 2010-12-18 at 16:37 +0800, Ian Chapman wrote:
You might also when to check the head load cycle count with smartctl
smartctl -a /dev/sdX | grep Load_Cycle_Count
They have stupidly aggressive power saving which as I understand it
cause many unnecessary head parks, depending on the workload of your
system. For example one of my drives hit over 240,000 cycles in 6
months before I became aware of it. By comparison another 4 year old
drive is currently only showing 70,000 cycles. There's a tool from WD
which lets you alter how aggressively it parks the heads.
Unfortunately it's a vendor specific setting the firmware so you can't
use a standard linux util to change it.
I had the same issue with a laptop. The internal drive parking itself
after a few seconds, and Linux waking it up every few seconds, rinse
lather repeat. Luckily, I could issue a command to change the power
saving on the drive.
Unfortunately, I have a similar problem with an external drive, but
thankfully the timeout period is much longer, since the drive
enclosure's interface blocks any attempt to directly configure the
drive. One of those Seagate Desktop Expansion drives, so I can't really
recommend them.
--
[tim@localhost ~]$ uname -r
2.6.27.25-78.2.56.fc9.i686
Don't send private replies to my address, the mailbox is ignored. I
read messages from the public lists.