Why, after a cloud flash (and prior successful cloud flashes), is a device doing this?
Feel like it would be even more rapid but buffered by rate-limit of Particle.publish in OS.
And I unplugged within the hour to avoid data overrun.
And I don't believe the signal is a problem, albeit somewhat poor -
Oh and we reverted to a prior program that was 'working alright' but came online and still did the OTA flipping...
Which version of Device OS?
Are you using sleep modes in your code?
Do you call System.disableUpdates() or System.enableUpdates()in your code? Where?
Were using electron 3.0.0
And were not using sleep mode or calling either of those functions
For more context me and eric just deployed firmware to 20 or so devices and 2 of them are erroring afterward, 1 of those 2 we are working on is flashing the hard fault SOS. It seems to be booting up then crashing repeatedly. As far as we can tell there is no invalid pointers or double emptying memory, however even if there was, why would it only be affecting certain devices? Any help is appericated thanks
Ok we just recovered it using the 'hard reset' via super useful post: Particle debugger to recover electron? - #11 by ak71vie
In short -
*'Rolled back' to tinker@2.2.0, came online fine.
*Then local flashed with an 'old' firmware, also came online fine.
*Then separately (repeating above) cloud and local flashed to 'new' firmware (running fine on other ~18 devices), and hard faults again...
Curious what to look for, since in our minds faults here but not on other devices, all things apparently the same.
Interestingly think there was an EEPROM mismatch and a 'hard rest' (above) along with an EEPROM erase seemed to solve it.
