I run home assistant on an Optiplex 3040 micro with a bare metal install. About a day and a half ago it started crashing it seems like about every 12 hours. It completely drops off the network and it requires a reboot to get it back up and running, it then runs fine for another 12 hours or so. It’s done it about 3 times now.
I took a look at the logs but I’m honestly not really sure what I should be looking for. How would you go about diagnosing the problem? Home assistant is up to date including the OS update that came out recently.
In Host logs there’s an option hidden under the three dots to show previous boots. Can you see some logs from the time when it became unavailable? If not, it may be possible the kernel crashed with a panic, or the disk became unavailable. In that case start with confirming it’s a regression by downgrading the OS, i.e. run ha os update --version 14.2 in the console. If you confirm it’s stable then, try another update to the latest version to verify the problem is reproducible. If it is, getting the logs might be a bit more complicated because you will need physical access to the device to check what’s shown on the display when it crashes.
Sorry Tom, but putting it like that that is unnecessary FUD and misinterpretation of the feedback. At this point we’re approximately around at least 100k of installs without any issues. Absolute majority of the issues has been identified shortly, quite some of them were unrelated to the update itself. The loudest group is RPi 5 users having issues when the device is running an old bootloader, then there is a known problem with AMD SoCs (used usually in some thin clients repurposed for HAOS). But almost all problems are limited to specific hardware or configuration.