Prev: rsync mystery
Next: lsof, pid, which Desktop/VT are running appY ?
From: Jeremy Nicoll - news posts on 17 Feb 2010 19:52
Aragorn <aragorn(a)chatfactory.invalid> wrote:

> ... The kernel does contain some routines to put certain USB devices in
> "ACPI sleep state", but for all I know that could simply be some code that
> tells the kernel not to poll the devices after a given time-out until some
> process is trying to access them again.

Would it be stupid to write a small shell script that - say - echos a
timestamp into a file on the disks concerned, and use cron to run that
script every so-many minutes?

--
Jeremy C B Nicoll - my opinions are my own.

Email sent to my from-address will be deleted. Instead, please reply
to newsreplynnn(a)wingsandbeaks.org.uk replacing "nnn" by "284".
From: Aragorn on 18 Feb 2010 01:31
On Thursday 18 February 2010 01:52 in comp.os.linux.misc, somebody
identifying as Jeremy Nicoll - news posts wrote...

> Aragorn <aragorn(a)chatfactory.invalid> wrote:
>
>> ... The kernel does contain some routines to put certain USB devices
>> in "ACPI sleep state", but for all I know that could simply be some
>> code that tells the kernel not to poll the devices after a given
>> time-out until some process is trying to access them again.
>
> Would it be stupid to write a small shell script that - say - echos a
> timestamp into a file on the disks concerned, and use cron to run that
> script every so-many minutes?

Stupid? No, definitely not. But it would be an ugly hack. There ought
to be better mechanisms to prevent devices from going into an unwanted
sleep state, and I suspect that the kernel provides for some mechanisms
for that. Provided of course that the sleep state is not induced by
the device's built-in logic itself, because it would be trickier to
override that, and in such a case, your script might come in handy.

--
*Aragorn*
(registered GNU/Linux user #223157)
First  |  Prev  | 
Pages: 1 2 3 4 5 6
Prev: rsync mystery
Next: lsof, pid, which Desktop/VT are running appY ?