Hi all
I’m sharing an experience I had today (and well, most of yesterday). I’m trying to resolve issues with a HA server and Zigbee setup that has migrated through 3-4 devices due to crashes and other issues, so I decided to start clean slate.
Ubuntu VM, Docker, Portainer, Mosquitto and Z2M; then HA;
On startup, HA auto added (not just discovered but added/configured) some Sonos speakers and some Chromecasts. Then the problems started. The logs indicate one or more of these devices became almost immediately unreachable which caused an error. These unreachable errors seemed to stack up to a point where HA Core decided to shutdown (ie a controlled shutdown) and, because my container is restart-always, I ended up in a restart loop.
Worse, deleting the devices / integrations from the UI didn’t seem to persist or fix the issue. I ended up having to delete the .storage directory and change the config.yaml to ignore auto discovery for Sonos and Cast.
That seems to have fixed my issue temporarily.
But my question for you smart people is this:
I’ve never had an issue before with these services, and I can’t see how some unreachable devices should cause a shutdown - previously they would have just shown up as needing to be reconfigured (which is also annoying, and I’m going through the exercise of static IPs and manually changing NIC settings to keep them alive, etc).
I’ll pop in the detailed logs later / if someone specifically wants to see the chain of exceptions.
Any thoughts?
