Random Unexpected Restarts

[boreal tapir]

My HA install is unexpectedly restarting and only HA. All my other containers seem to be fine.

Installation method: Home Assistant Container
Core: 2025.11.3
Frontend: 20251105.1

OS: Ubuntu 24.04.3 LTS
Memory: 64GB
CPU: Intel(R) Core(TM) i7-6700T CPU @ 2.80GHz

• No Memory leak from what I can tell
• No CPU usage spike
• Plenty of storage available

I've enabled Debug Mode and Asyncio Debug Mode and checked the logs surrounding a few different restarts and can't seem to find anything that would be directly linked to a restart.

There doesn't seem to be any pattern to it or at least I haven't found it yet. The restarts usually occur within 48 hours of the previous restart though.
I'll add the log file from before the restart and directly after as a reply to this thread.

I don't know exactly when this started but now that I'm thinking about it it might have been around the time I upgraded to 2025.11.

Not necessarily looking for a solution but threads to pull and investigate

Logs from before the restart: https://pastebin.com/Vk95eGRz

Logs from after the restart: https://pastebin.com/UjMCGEUP

[worldly bay]

There are quite a lot of errors in that log. I suggest starting by renaming the custom_components directory by adding a .safe suffix or similar, or maybe temporarily remove this one:

2025-12-01 05:45:51.114 ERROR (MainThread) [custom_components.monitor_docker.helpers] [Docker]: Can not connect to Docker API (Missing valid docker_host.Either DOCKER_HOST or local sockets are not available.)

[boreal tapir]

Thanks for the suggestion, what would that accomplish?

I will remove that docker thing though so I don't use it. I'll also remove all the other custom components and integrations I don't use as a start. The docker one is complinaing because I stopped passing the docker socket to the HA container.

[worldly bay]

the log is full of errors, so it's hard to know what might be fatal to HA

basically, telling HA not to load any custom components is the first step for stuff like this

[boreal tapir]

telling HA not to load any custom components is the first step for stuff like this

Gotcha. I was trying to figure out what renaming the directory would accomplish but I see that now

I'll work on cleaning up those template errors and such as well to alleviate some noise

[boreal tapir]

Okay, so I've resolved as many of the Errors from the logs I could.

There was an issue with one of my manual sensors that was trying to convert the string "33.9 psi" to a float.
There was an issue with a blueprint I was using where a key was not present that was expected to be.
The Monitor Docker custom integration from HACS could not connect to the Docker.sock and that was intended since I removed it. Not using that Integration anymore so I completely removed it. I also removed all other HACS integrations or custom components I wasn't using either.
I use a local DNS server and Home Assistant was trying to send analytics and was unable to reach the server due to my DNS server filtering the DNS query. Allowed that one through the gates, I'd give HA my first born if it asked.
There was an error in an automation that I don't use, deleted the automation.

Considering the restarts can take quite a while to happen I'll just wait to see if clearning any of that up solved this 🤞

[boreal tapir]

Got a restart today.
It restarted at 15:25 local time (CST) logs are in UTC. Logs follow

Before: https://pastebin.com/BdeDCwhC
After: https://pastebin.com/PcGf1uPd

The last log before the restart was almost a whole 6 minutes beforehand. There doesn't seem to be anything in there that would point to why things are restarting.

I do see that my weather event sensors are taking a while to update due to the API taking a long time to respond. I have the scan interval set for that sensor set to 1 minute. I suppose if the timeout on the request is >= 1 minute I could get a backlog of those requests. Increasing that interval to 10 minutes should solve that issue if it exists.

Again, no CPU or memory spikes. CPU sits around 25% all the time, memory around 10-15%

[worldly bay]

It seems like you're having some issues with Google that could be affecting HA's web component, but I suggest running docker compose logs -f homeassistant after a failure (or even now) to see what happened in the last iteration from the container standpoint