This is a very simple utility intended to help with troubleshooting performance problems. All it does is capture the top N CPU hungry processes on a linux system every N seconds, and outputs that information to a log file with timestamps.
Run install.sh as root. This will do the following:
- Copy the primary bash script to
/simple-process-monitor.sh - Copy the systemd file to
/etc/systemd/system/simple-process-monitor.service - Copy a logrotate configuration file to
/etc/logrotate.d/simple-process-monitor.conf - Create a logging directory:
/var/log/simple-process-monitor/ - Reload systemd services
- Enable a new service called
simple-process-monitor - Start the new service called
simple-process-monitor
The log containing process information is located here: /var/log/simple-process-monitor/simple-process-monitor.log
If the script outputs anything to StandardOutput, that will be located here: /var/log/simple-process-monitor/StandardOutput.log
If the script outputs anything to StandardError, that will be located here /var/log/simple-process-monitor/StandardError.log
A logrotate configuration is setup to rotate the files within /var/log/simple-process-monitor/ on a weekly basis, if the files reach 10M in size.
To enable on boot: systemctl enable simple-process-monitor
To disable on boot: systemctl disable simple-process-monitor
To start: systemctl start simple-process-monitor
To stop: systemctl stop simple-process-monitor
To restart: systemctl restart simple-process-monitor
Get Status: systemctl status simple-process-monitor -l
All configuration exists within the bash script simple-process-monitor.sh towards the top of the script. The installed location is /simple-process-monitor.sh so you would need to edit it there after installation. If you change the configuration when the utility is already running, you need to either restart the service or reboot.