Photon losses connection with cloud (sometimes) after several months deployed in field

I have a couple of photons deployed in the “wild” and after several months of proper operation, they lose connectivity to the cloud and my MQTT server. Most of the time, when this happens, I use Particle Workbench to re-upload the program, and things are working again. However, this last time, I get a message “Device is running debug firmware which does not support OTA. Reflash Device OS using Workbench or CLI (particle update)”. Per console, my device has 1.4.4 loaded.

Help?

More data, from my console.particle.io, my remote photon reports every minute the following sequence (basically a “online” status, followed 10-20 seconds later by “unreachable”).

spark/status - Published by xxxxxxxxxxxxxxxx83931 on 5/30/20 at 4:28:29 pm - online

then

spark/device/diagnostics/update - Published by xxxxxxxxxxxxxxxxx83931 on 5/30/20 at 4:28:54 pm - {“service”:{“device”:{“status”:“unreachable”},“cloud”:{“uptime”:24,“publish”:{“sent”:2}},“coap”:{“round_trip”:null}}}

I would raise a ticket with Particle. I haven’t seen this but there have been changes made on the Cloud side that are having subtle affects on APIs. I guess you might first want to try a dfu mode complete update of the OS including bootloader.

I’ll try the “ticket” route, but the DFU/usb programing solution requires 350 miles each way to make it happen…