Commands I find useful
Check the 200 latest rows in the log (good for knowing why something went wrong)
Increase number for more rows, or skip to get the whole log. If log have rotated since incident add
.1 after syslog.
sudo tail -n 200 /var/log/syslog
Check ram memory
Check how much ram and cpu docker containers use
sudo docker stats
Subscribe to all messages published on Mosquitto mqtt broker
monitor/# to follow all messages on "monitor" topic, or other topic you'd like to follow.
mosquitto_sub -u username -P password -v -t \#
Rotate logs in /var/log
Good if logs grow too big. After running, delete all files with ".1".
sudo logrotate -f /etc/logrotate.conf
Check what's using cpu
Set up cron jobs to for example restart service, reboot or other stuff
sudo crontab - e
55 * * * * systemctl restart monitor.service
The above cron expression runs at hh.55 every hour. Cron expression examples