Disk IO going to 100% and killing Home Assistant after upgrade to 2022.12

Hi all,

I have HassOS installed on an Intel NUC. After upgrading to 2022.12.1 it seems that after running for a few hours Home Assistant stops responding and I’m unable to access it at all.
When I look at the Disk IO activity light on the NUC it is constantly on during this time and I am only able to restore Hass after rebooting the NUC.

This has happened twice after which I restored a backup from 2022.11 and the problem went away.
Any ideas / suggestions on how to diagnose what’s causing this ?

I noticed there are some (heavy) database updates, which run in the background; i suspect that was the case here…(it should be possible to check in the logs)

How long does this happen after installation ?
I can’t access the logs as Hass is inaccessible when this happens. After reboot logs are wiped clean so I don’t have a history pre-reboot.

How long the updates last depends a bit on the hardware used.
I know some months ago it would take at least one hour before the updates where finished (using MySQL with 6 months history).
If you are using SQLite, and are not intrested in the history, you can simply delete the home-assistant_v2.db file (and it will create a new one)

1 Like

Thanks will give it a try and see what happens.

The log from home assistant’s previous run will have .1 add to the end of its filename.

Thanks. So I ended up deleting the DB file and upgrading.
So far so good.

1 Like