Home-assistant_v2.db corrupt after each reboot

Installation
1

Hi together,
I do have my homeassistant installed in a docker on my raspberry pi5 running on a SSD card.
I’m a bit lost - copilot as well doesn’t have a solution.
More or less after each reboot, home-assistant_v2.db is corrupt. I can delete it but homeassistant restores it in a corrupt way again. That’s the log:
13:54:58] INFO: Home Assistant Core finish process exit code 256
[13:54:58] INFO: Home Assistant Core finish process received signal 11
s6-linux-init-hpr: fatal: unable to talk to shutdownd: Operation not permitted
execlineb: fatal: unable to exec /package/admin/s6-linux-init/command/s6-linux-init-shutdownd: Exec format error
execlineb: fatal: unable to exec /package/admin/s6-linux-init/command/s6-linux-init-shutdownd: Exec format error
execlineb: fatal: unable to exec /package/admin/s6-linux-init/command/s6-linux-init-shutdownd: Exec format error
execlineb: fatal: unable to exec /package/admin/s6-linux-init/command/s6-linux-init-shutdownd: Exec format error
2025-10-28 13:55:02.284 WARNING (Recorder) [homeassistant.components.recorder.util] The system could not validate that the sqlite3 database at //config/home-assistant_v2.db was shutdown cleanly
2025-10-28 13:55:02.363 WARNING (Recorder) [homeassistant.components.recorder.util] Ended unfinished session (id=736 from 2025-10-28 12:54:37.880482)

Only solution so far ist removing homeassistant docker, create new and restore backup.
Why is homeassistant restoring a corrupt db?

Thanks for your time!

2

If you mean these two lines, these are warnings. I also get these on every restart and i just ignore them and home assistant is working normally. Do you have other symptoms that let you think that the database is corrupt?
(3 instances of HA in docker containers)

3

Hi Jürgen - yes - don’t miss these:
13:54:58] INFO: Home Assistant Core finish process exit code 256
[13:54:58] INFO: Home Assistant Core finish process received signal 11

It’s an endless loop and you can’t access the website - moreover the PI5 is running at roughly 50% CPU all the time

4

Try to find the reason for these messages. It is probably dependent on the installed linux distribution.

5

I think you are making a connection that do not really exist.
It looks like the DB is shutdown (with the standard warning about a clean shutdown), but something else made HA fail.
I do not run a Docker installation, so I am unsure of the requirements. Check if you are missing something, like a host network access or a system service.

6

nothing missing - when I stop homeassistant via docker stop homeassistant before shutdown/rebooting all is fine. Meanwhe I changed my PI5 to HAT M.2 SSD and I’m instaling a shutdown script that should stop homeassistant ( and others ) before shutdown/reboot/halt and I’m testing this service

7

unfortunately the script runs manually but not via shutdown/reboot/halt

8

I had this very early on when I had a docker installation (since moved to HAOS) and got warnings about the database not being shutdown cleanly, but add a

–stop-timeout 180

to your docker run command (or option when setting up the docker, or use the docker stop command) to increase the timeout see docker container stop | Docker Docs as mentioned in the docker documentation it defaults just to 10 seconds to allow it to shutdown cleanly. I increased it to 90 seconds but don’t think it takes that long to do it.

Test it to see if that stops that error from happening!

Alex

1 Like
9

Hi Alex - thanks - I do have already 90s break in it.
What finally helped was moving to M.2 SSD. This is fast enough to write and close everything before shutdown. but now - how wonder - I have another error in the logs:
execlineb: fatal: unable to exec /package/admin/s6-linux-init/command/s6-linux-init-shutdownd: Exec format error

Seems that something is missing:
alfred@raspberrypi5:~ $ docker run -it --rm Package home-assistant · GitHub sh
ls -l /package/admin/s6-linux-init/command/
=> shows nothing

10

I run docker and require no special command

History/recorder is hard on HDD/SD. I had this issue when running pi with SD. Eventually I disable recorder/history and only reenable after move back to x86/hdd.

DB never have issue for me even if pull power plug and bad shutdown.

Could you have hardware issue?

11

Hi, no - it’s the 3rd PI5 now.
I guess slow SSD created the problems. After moving to HAT M.2 SSD problem is gone - but now I do have an s6related prob.

12

Hi Alfred,

Unfortunately I don’t have the docker image to be able to check if I can see if that is the same. I take it the execlineb is from the home-assistant.log or from docker logs home-assistant?

Does your architecture match up as Exec format error seems to indicate that?

On the docker host try

uname -m

then check the docker image that is used

docker image inspect home-assistant/home-assistant --format=‘{{.Architecture}}’

the image name may need to be changed (e.g.

docker image ls

)

Hopefully you have aarch64 for the docker host and possibly the same in the docker image?

13

Hi Alex,

yes you’re right - logs are out of container “docker logs homeassistant”.
I tried a lot of things with the installation.
At the end I destroyed home assistant container completely, started from scratch and resored latest backup :slight_smile:
Since that time it’s running as expected - no issues at all.Thanks for your time!

1 Like