From: ekgnkb3d <ac...@gm...> - 2011-08-23 20:46:31
|
Paul Alfille-2 wrote: > > Can you describe more about the circumstances around the slave resets? > A simple temporary disconnect of the cable, it totally confuses some of the parts. For example a DS18B20 is almost not effected, it simply will be polled again after the cable reconnected and responded a new value - fine, in this case you can call it "stateless". But the problem starts if it had been alarms configured, they got lost and therefore the DS18B20 will never trigger the alarm after the cable was disconnected/reconnected until the configuration is send again (I guess, I'm not sure with the DS18B20) so it behaves stateful in volatile memory. With the BAE0910 I'm sure: Once the cable was disconnected only 1s, the complete PWM configuration is gone. A later writing to the Duty-Cycle registers to modify the PWM will not raise an error, but the output pins are on default mode, which is not working PWM... so it have to be reconfigured. In my case it can happen, that the 1-Wire bus is disconnected for a short moment. I would like to continue operation like before, once it is reconnected.... Is this somehow possible? Thanks Achim -- View this message in context: http://old.nabble.com/How-to-detect-lost-configuration-%28e.g.-in-BAE0910%29-tp32319775p32322059.html Sent from the OWFS - Dev mailing list archive at Nabble.com. |