The device publishes a Particle event named spark/device/diagnostics/update in response to a Device Cloud request to refresh device vitals. The scenario of a wake from sleep (stop) appears to cause my devices to get into an odd state which is eventually picked up by the watchdog. Device OS 2.2.0/Photon.
If the device is put into sleep mode and then woken on an interrupt pin sometimes (and this seems to depend on the time the device has been in sleep), then request to send the diagnostics update event arrives at the same time the device is sending an event to confirm it has been woken this results in a hang. Is there anything defensively that can be put in the application to control the accepting of the diagnostics update request?