W dniu 30.04.2014 13:02, Lennart Poettering pisze:
On Wed, 30.04.14 01:02, Marcin Kocur ([email protected]) wrote:

Hello all,

I wrote a service /etc/systemd/system/disable-2nd-hdd.service. I
have 2 HDD's where one of them doesn't need to work all the time. So
I thought it'd be cool to shut it down after 30 minutes of
inactivity (the disc isn't mounted when system starts):
/usr/bin/hdparm -S 241 /dev/sdb

Then I wrote a service:

[Unit]
Description=Disable sdb
Requires=local-fs.target
After=local-fs.target
These two lines are implied and hence can be removed.

My intention was to be sure that the disc is ready to be spun down.

Note that devices are probed asynchronously, and a device not needed at
boot is not waited for. THis means that your service might run before
the harddisk showed up in the kernel, or possibly that your hdparm
command is invoked before the initializating/probing/identifying of the
hdd or its filesystems is compelted, thus possibly negating the effect
of your command.

YOu probably want to invoke your commandn with RUN from a udev rule so
that it is run when the hw is found.

Thanks, I expected such situation, namely that disc is somehow probed or activated AFTER I issue hdparm command which causes it wake up again. Right. But I really expected that systemd is able to determine when the disc is ready and I can somehow pass this in Requires= or After= keys. My intention was to learn systemd a bit, not udev's hackish syntax ;p

--
Pozdrawiam / Greetings
Marcin Kocur █
Dane kontaktowe / Contact details: http://koci.net.pl/email

_______________________________________________
systemd-devel mailing list
[email protected]
http://lists.freedesktop.org/mailman/listinfo/systemd-devel

Reply via email to