I have asked questions before about a hardware watchdog timer, and looking at the issues list for 0.5.x and 0.6.x - it seems that nothing is proposed - is this true ???.
In my opinion we really need an integrated watchdog solution, to guard agains any form of software/firmware failure/lockup.
I have a Photon which has been running perfectly since mid-Feb, controlling my pool heater. On Wednesday, something went awry and it was no longer controlling the heat pump. The ONLY way I could tell (apart from guests saying the pool was like a sauna was that it had stopped sending out some messages (both UDP status reporting and messages being forwarded to Librato). The cloud seemed to believe that the Photon was all OK - said it was ‘breathing’ cyan) - but somehow my own code loop had failed (its 5000 miles away so I can’t debug it easily ;-).
A watchdog timer being refreshed from within my main loop would have rebooted the photon and recovered within minutes (depending on timeout). I have successfully used a watchdog on numerous PIC-based projects/products with excellent success.
So - if its not likely to become part of the Photon OS - then I guess I’ll have to ‘roll my own’ - which is probably going to be a simple external hardware option - as I worry about tinkering down in the hardware timers when the OS might be changed in the future
Of course - if I’ve missed it somewhere (as I really don’t know the roll-out schedule) then I apologise for asking again…