While it does not happen often, last night at 9:00 pm HA and its addons went unresponsive. I woke up to all of my Gosund switches which have been flashed with Tasmota blinking red. Most times this means that a network issue exists and when I was using Asus it was common, but I have since switched to Unifi Access Points with an Edgemax router and network is rock solid. It took me a bit to think through but eventually I came to the conclusion that MQTT was not responding and so I rebooted the Rpi4 and all the switches went back to normal. My wife was less than pleased. What are you all using to notify or even restart an unresponsive HA instance? I was considering setting up a zabbix network monitor on my proxmox server but wonder if that is overkill.
I believe there is some kind of a daemon that can be uitilized when the CPU has kernel panic (forgot what it was called), but I agree, that is probably not the best way… You can set up a repeating ping that would send you an alert when it does not get a response after x time it could tell HA to either restart or even restart the host itself, but purists will tell you that is not the best approach, the best approach is to resolve the core issue/s, but I would take the other side of that argument in some cases, with increasingly complex home networks sometimes there are combinatiolns of issues that cannot be anticipated - !