From there it just loops endlessly, trying to reconnect, but the problem is not detected by HA. Restarting the add on usually solves the problem.
My feature request is that it would be better if the add on failed after a certain number of retries, so that the HA watchdog or some other automation could restart it. Is that doable?
Do you know if the deconz app has a health/alive API? If so I can plug that in as the watchdog URL so it can monitor it. Basically just looking for an API that returns a 200 if its alive and anything besides a 200 otherwise.
As I said, we don’t develop the addon. Furthermore, the docker version doesn’t seem to have this issue. As the ha addon is also using docker in some extend, there should be other ways.
Thanks for a quick response. I’m trying to navigate here with little success. You’re pointing towards the addon developers, and they’re pointing at the deCONZ developers so I’ma little lost.
My understanding is that the addon wraps around the underlying deCONZ code. Isn’t that correct?
Is the log output in my OP generated by the ceCONZ code or the addon?