Yesterday, I discovered why all my devices where acting strange, Home Assistant restarted every minute or so, without any notices or errors.
I checked the logs and nothing fatal showed up, except for ERROR (updater) [telegram.ext.updater] home assistant
.
After a while the logs showed the client disconnected message and then hass got restarted, clearing the logs.
The strange thing is, that when I try to stop hass from restarting a crashed Home Assistant instance, it wouldn’t allow me to do so, so I had to use some brute force (sudo killall hass) to kill all Home Assistant instanced (since there were multiple running somehow).
When I dropped into the shell as the homeassistant user and I executed /srv/homeassistant/bin/hass -c "/home/homeassistant/.homeassistant"
everything was fine, and the server has been running non-stop for 10 hours straight.
When I try to start Home Assistant with systemctl start [email protected], it keeps rebooting and not telling why.
I am running custom components (Toon components from https://github.com/cyberjunky/home-assistant-custom-components) but they are all working as they should.
I did get a few database locked warning, but they are gone too, since I removed (renamed) the database file.
Any ideas why Home Assistant keeps crashing when it’s started by systemctl and not when it’s started with the exact same command line?
/etc/systemd/system/[email protected]:
[Unit]
Description=Home Assistant
After=network.target
[Service]
Type=simple
User=homeassistant
ExecStart=/srv/homeassistant/bin/hass -v -c "/home/homeassistant/.homeassistant"
[Install]
WantedBy=multi-user.target
I am running Home Assistant on Ubuntu 16.04.6 Server LTS with an Intel® Core™ i5 CPU 750 @ 2.67GHz (quad core) with around 10GB’s of system ram.
Home Assistant has been running smoothly for around a year now.