Diagnosing unexpected HAOS restarts

[daernsinstantfortress]

For months and months now, I've been having issues with constant, periodic (but unpredictable) restarts of HAOS. The box simply appears to reset and there is no useful diagnostics in journalctl -b -1 that would indicate what caused it - the logs just stop pretty arbitrarily. Right now, it's happening at least once per week, with the last one just under two days ago. I've replaced the RAM in the box and soak tested it for a few hours, with no errors found, but still it continues to happen. The box is running on a UPS and nothing else connected to it has any issues, so I don't think that power is the problem.

Can anyone suggest any next steps to try to diagnose this issue within HAOS? Even a kernel panic would be a step forward for me, as it would give a hint as to the problem! Are there any debug options I can enable that might allow me to dig deeper into this?

Platform:

• Dell Optiplex 3060
• i5-8500T
• 16GB
• NVMe SSD storage.
• HAOS 15.0, bare metal

Linux homeassistant 6.12.18-haos #1 SMP PREEMPT_DYNAMIC Mon Mar 17 11:16:39 UTC 2025 x86_64 HAOS

[bcutter]

Just for the usual obvious part: what SSD are you using and how is it connected (directly, USB enclosure, ...)?