Setting up Linux init.d service

First, write a server configuration file, which would be placed inside /etc/init.d directory. Note, that your service name should (must) end with letter 'd'. Example: xmrigd. Configuration file (/etc/init.d/myserviced) should look as follows (newlines matter):

#!/bin/sh

### BEGIN INIT INFO
# Provides:             myservice
# X-Start-Before:
# X-Start-After:
# Default-Start:        2 3 4 5
# Default-Stop:         0 1 6
# Description:          Runs myservice service
# Short-Description:    Runs myservice service
### END INIT INFO

export PATH=$PATH:/usr/bin:/bin

case "$1" in
start)
    (/usr/bin/myprogram &)  # Manually fork here, it is important
    ;;
stop)
    ps aux | grep -i myprogram | awk '{ print $2 }' | xargs kill -TERM
    ;;
*)
    echo "Usage: /etc/init.d/myserviced {start|stop}"
    exit 1
    ;;
esac

exit 0

• Provides. Name of the server without letter 'd' (again, it is not necessary, but just for good look at least ...)
• Description. Description of the server. You can type anything you want here, but you should avoid newlines.
• Short-Description. Short description of the server. You can type anything you want here, but you should avoid newlines.

In this example, service script just executes /usr/bin/myprogram and calls fork. Sometimes you may find out that the service does not see this executable (even if it is located in /bin directory). This is normal, just set the PATH variable in your service script (take a look at stackoverflow solutions).

Since this script is not a systemd service, we should convert it to one. This can be done with update-rc.d command:

update-rc.d myserviced defaults

Now, you can start your service by running:

systemctl start xmrigd.service

Make your service run on startup:

systemctl enable xmrigd.service

Verify that your service is running withour unwanted errors (press 'q' button to exit out of interactive menu):

systemctl status xmrigd.service