Home Assistant daily crashing


I’ve been running Home Assistant on a Raspberry Pi 4B with 2GB RAM for about 18 months. Its in an Argon ONE V2 case with an Argon ONE power supply. I am running Z-Wave via a Z-Wave.me USB stick, ZHA via a Popp ZB Shield, have a SSM-D (replacement D2 here SSM-D2)and am running it from a 240G Kingston USB SSD. I have the following integrations:

  • AccuWeather
  • Apple TV
  • BMW Connected Drive
  • Broadlink
  • Fully Kiosk Browser
  • Generic Camera
  • Google Nest
  • HACS
  • VELUX Gateway as a HomeKit accessory
  • HomeKit
  • Logitech Harmony Hub
  • Mosquitto
  • Nest Protect
  • NUT
  • OpenUV
  • SmartThings
  • Sonos
  • Sony Bravia TV
  • Speedtest
  • MariaDB
  • Tasmota
  • Thermal Comfort
  • Z-Wave JS
  • ZHA

And the following Add-ons:

  • AppDaemon
  • ArgonOne Active Cooling
  • Duck DNS
  • File Editor
  • Grafana
  • InfluxDB
  • Lets Encrypt
  • MariaDB
  • Mosquito Broker
  • php-nginx
  • Portainer
  • rest980 docker image
  • Samba Share
  • Studio code server
  • Terminal and SSH
  • WireGuard
  • Z-Wave JS

Until 6 months ago, this all worked fine, and then I started to see periods where Home Assistant would hang and nothing would respond. The only way of solving this was to wait (sometimes up to 6 hours), or restart the Pi. The frequency of occurrences has increased over the last 6 months to the point where it will do it every day. It seems to happened most often at about 0500 BST and then when we put the clocks back last weekend, I noticed that it was at 0400 GMT, so is not changed by daylight savings. I’m mitigating it by running a script that restarts the Pi at 0400 every day but can’t find the cause. I have tried disabling all the Integrations at some point or another and it still happens. I suspect it may be hardware, but how do I find out? I’ve looked at the logs and can’t see what’s likely to be responsible.



It could be backup or database purge that is running around that time and with the hardware you have connected it could be an underrated power supply.

I would probably try to move some of the USB devices to a power hub to see if that helps.

I would try to some ideas like:
install Glances and check CPU usage, IO etc. Try to looks data into Grafana about the same when you have a hangs.
keep an open terminal session with tail -f homeassistant.log
Put off add-on one by one and see.
Put off integrations in same approach
Execute service call to clean the database.
check dmesg, core info, supervisor , all possible logs
Complete shutdown + check with HA cli commands
Disk space usage

I hope it help a bit

Thanks. I’ve tried that but it didn’t work. Weirdly I couldn’t get it to boot when I moved the SSD to the powered USB hub.

When you move an USB then you change its path, so your RPi could not find its drive with important files.
Only move other USB devices. :slight_smile: