Home assistant rebooting every 3 minutes or so

[mossy plover]

Suddenly today my **Home Assistant Blue ** keeps rebooting. I recently installed ZHA which is the only change. I have not updated today, for example.

I am able to log in long enough to look at the logs, and I currently have "Home Assistant Core" and "Supervisor" logs I can provide.

In an effort to rule out the Zigbee usb stick causing the issue, I unplugged it. I still get the reboot cycle.

I left it alone after it "died" last time, and it eventually came back.

The most relevant log entries from "Supervisor" log seem to be the following, but I can provide more logs if that is helpful.

2024-12-06 15:43:53.165 ERROR (MainThread) [supervisor.homeassistant.api] Timeout on call http://172.30.32.1:8123/api/core/state.
2024-12-06 15:43:53.166 WARNING (MainThread) [supervisor.misc.tasks] Watchdog missed an Home Assistant Core API response.
2024-12-06 15:46:24.155 ERROR (MainThread) [supervisor.homeassistant.api] Timeout on call http://172.30.32.1:8123/api/core/state.
2024-12-06 15:46:24.156 ERROR (MainThread) [supervisor.misc.tasks] Watchdog missed 2 Home Assistant Core API responses in a row. Restarting Home Assistant Core!
2024-12-06 15:46:24.164 INFO (SyncWorker_4) [supervisor.docker.manager] Restarting homeassistant
2024-12-06 15:46:45.074 ERROR (MainThread) [asyncio] Task exception was never retrieved
future:

I also just found the following in the Home Assistant Core log

RuntimeError: Detected that custom integration 'hass_agent' calls device_registry.async_update_device from a thread other than the event loop, which may cause Home Assistant to crash or data to corrupt

which makes me think maybe I should turn off that custom integration. It is strange though - I have been using that integration for several weeks.

[midnight kestrel]

similar thing is happening to me a few threads down but with different errors.

[summer tendon]

I'm having a very similar issue. I updated to HaOS this morning and the system never came back up. I'm running an old NUC system (DN2820FYKH).

I finally plugged a monitor into it see what was going on. The system would boot and typically 5-10 seconds are starting the Home Assistant CLI, the cursor would freeze and a few moments later the system would reboot.

I tried switching between boot slots as one of them looked like it had the older HaOS on it, but even that slot experienced the same thing. I unplugged everything from the system except power, keyboard, and monitor. Nothing changed. The system would come up on the off occasion, but my curiosity got the best of me and I wasn't hearing anything from anyone in the house about things not working.

I updated both Core and OS to the latest while it was up hoping there was some fix in there, but neither helped. The last thing I did to get it working before I decided to stop investigating was this:

1. Turn off system
2. Connected everything EXCEPT my Zigbee USB. For me this was Ethernet and a USB cable to the UPS.
3. Once Home Assistant CLI started on the console I quickly typed "core stop" and pressed enter.
4. I gave the system a few minutes to see if it was still up and it was so I went ahead and started Core by running the command "core start"
5. Again, gave it a minute or two to see if it was still up and since it was, I plugged by Zigbee USB in and all the Zigbee devices came back online.

I'm not sure if leaving the Zigbee USB unplugged was necessary, but it seemed to be more likely to stay up if that device was unplugged.

Unlike you two, I'm not seeing any glaring errors in the logs as to why this is happening. The only warning I'm seeing is regarding Pulse Audio.