Power savings on USB Hard drives
Laurent LEVIER
llevier-lists at argosnet.com
Mon Dec 3 12:00:21 PST 2007
Hi Everyone,
Since my post, I found some new interesting stuff digging & playing around.
I got here 2 USB2 HD:
One based on a Western Digital "Mybook Essential Edition" 500 GB HD
(black book design with a circle green light as indicator)
Another one based on a basic Textorm alloy case, with fan + switch.
When I plug the Textorm to my M$ Windows PC, HD does not spin down at
disconnect time. However, WD does (in fact, it switches off completely).
So I connected my WD on my FreeBSD 6.2p8 and the "camcontrol stop
da0", or start or eject (except it ejects nothing), are working perfectly.
When I try this with Textorm case, I receive as answer an illegal request:
# camcontrol stop da0 -v
Error received from stop unit command
(pass0:umass-sim0:0:0:0): STOP START UNIT. CDB: 1b 0 0 0 0 0
(pass0:umass-sim0:0:0:0): CAM Status: SCSI Status Error
(pass0:umass-sim0:0:0:0): SCSI Status: Check Condition
(pass0:umass-sim0:0:0:0): ILLEGAL REQUEST asc:20,0
(pass0:umass-sim0:0:0:0): Invalid command operation code
So, after this first analysis: definitely the USB case makes the difference.
The USB->IDE Interface must be able to understand the SCSI start/stop
commands so camcontrol works properly.
About the timeouts now...
Of course, the Textorm plugged on FreeBSD keeps running infinitely,
even unmounted.
The WD case spins down by itself (I supposed) after a while.
Final test, WD mounted, after a while, guess what??
External HD normall spins down after 15mn, that is the delay to power
save in my bios!!!
So as a summ, definitely the case makes possible or not to have power
save on USB hard drives.
The point is: how to know the case you are purchasing can (or not) do this?
All I can say is the one I purchased from WD is ok.
Hope this will help everyone
Brgrds
More information about the freebsd-acpi
mailing list