Wirrunna;433927 Wrote: 
> Gordon, another point to add in the help files (if you haven't already
> done so) is that SCPowerTool keeps the Server log open, so in XP if you
> want to delete the log to start fresh with a new release, you have to
> close SCPowerTool with the task manager before you can delete the log.
Huh.  That's surprising.  While SCPowerTool.exe does write to the
system events log, it doesn't touch the SC server.log.  I wouldn't have
expected this and it makes me wonder if SqueezeCenter isn't formally
closing the log file and is depending on the OS to close it once it sees
that the SC process has terminated.

Perhpas this could be solved by using StartHidden.exe to execute
SCPowerTool.  By necessity, I'm using StartHidden to launch SCPowerTool
as a fully detached process for the "Restart SqueezeCenter" action on SC
7.3.x.  Perhaps I ought to do this for every invocation of SCPowerTool.

I'll try to do some tests on this before the next "release".  Thanks
for pointing this out.

Also, there ought to be an easier way to kill instances of SCPowerTool
without having to resort to the task manager.  Perhaps a
"--cancel-wakeup" parameter which would have the effect of killing any
instances that have pending wakeup timers.  I have a feeling that this
is easier said than done in via actual code, though....especially since,
as now written, SCPowerTool doesn't have a real message loop.

Anyway...this sort of problem just points out how weirdly the whole
system wakeup issue is handled by windows, as compared to linux and osx.
With the latter OSs, the kernel supplies a driver that completely
abstracts the programming of the bios rtc wake alarm.  With these, after
a simple call, the system will wake itself from any power state,
including stone cold off.  With windows, afaik there is no way to
directly program that rtc wake alarm.  Wakeup on windows is really a
side effect of asking the OS to service a timer in a dormant process
that's hanging around in the background waiting for that call...and
nothing else...and, oh yes, please wake the system up if you need to to
get this done, Mr. windows kernel.  I think of myself as being more
adept at writing C code for windows that at writing bash scripts.  But
this is one area where windows seems really primitive.  It seems strange
to me that windows fails offer what seems like such a basic facility. 
Maybe I should rethink this whole approach and instead of using waitable
timers to do this, try programming the system task scheduler instead.


-- 
gharris999
------------------------------------------------------------------------
gharris999's Profile: http://forums.slimdevices.com/member.php?userid=115
View this thread: http://forums.slimdevices.com/showthread.php?t=48521

_______________________________________________
plugins mailing list
[email protected]
http://lists.slimdevices.com/mailman/listinfo/plugins

Reply via email to