On my system the power control loop checks againsts a value set by the
system state loop (srvPowerControl participates in defining the system
state);
If there is no value to be read by the control loop, or the value is
not polled by the control loop between 2 updates of the state loop,
something is wrong (since by design the control loop runs light and
fast, and the state loop slow with low priority.) 
When no polling has occured, my system commands reboot (after a few
reads, to "be sure" --e.g. take services startup order into account)

I suggest having a marker saying the system state value has been read
or written, optionnaly by whom. That can serve IHMO avoiding cases where
one system waits on updates from the other, while in fact it is out of
service. I'd be wary about using a timestamp, because of
suspend/hibernate. A variable or boolean would do a more robust job.  
(this reflex of rebooting is just mine; a more lenient fallback policy
will probably be better for most systems)


-- 
epoch1970
------------------------------------------------------------------------
epoch1970's Profile: http://forums.slimdevices.com/member.php?userid=16711
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