I am not sure if this is happening with the latest MDot production libmdot version 2.0.17 or used to happen earlier.
We have a custom firmware compiled against the libmdot library mentioned earlier. On MDot wake cycle, the application reads sensor data transmits it to conduit if the data has changed, or just sends some kind of heart beat to indicate the mdot is active and then goes into deep sleep mode for a minute and wakes up using RTC WAKE mode. This cycle repeats.
Although the time interval when it doesn’t come out of sleep seems random, if we keep the mdot running all the time, the mdot suddenly won’t wake up even though the 1 minute sleep cycle has elapsed. Sometimes it 10 hrs into the wake/sleep cycles, sometimes 24 hrs and the longest so far we have seen is 30 hrs into the wake/sleep cycle. Also we have enabled debug full trace, so we know that the mdot went into sleep and the session info was stored successfully in the flash and then it never seems to wake up. So it almost seems like the mdot has entered some idle cpu mode.
So currently we send data via the USB serial port and all of a sudden the mdot seems active again and goes back into its regular wake/sleep cycles, until it hits the sleep mode scenario described above.
Any thoughts on why this could be happening over long period of time?
Thanks,
Ajay