Check_mk documentation clarification

In the Check_mk setup instructions here: https://docs.librenms.org/Extensions/Agent-Setup/
The instructions say to copy the service file to /etc/xinetd.d

`cp check_mk_xinetd /etc/xinetd.d/check_mk`

Then the instructions say to start the service in init.d, which cannot be done, because it is not there.

`/etc/init.d/xinetd restart`

Should the service go in init.d, or is there a reason it has it’s own folder and the startup instruction needs to be corrected?

What OS/distro is this on? The idea is to restart the whole xinetd service so that it reconfigures and then knows about the check_mk sub-service that you added to it.

I am using Ubuntu. I understand your answer, but it still does not answer my question. From what you are saying xinetd.d (not xinetd) should have it’s own directory? If that is the case, then the restart command is incorrect in the instructions. IF the restart command is correct, then the copy command is incorrect in the instructions. Now with all of this said, init.d is symbolically linked to rc, so shouldn’t check_mk go into the init.d directory?? What am I missing here?

What I’m saying is that xinetd looks in the xinetd.d directory and uses any of the files in that directory to create sub-services.

So when you copy the check_mk file to /etc/xinetd.d/check_mk, you instruct the main xinetd service to now run a check_mk sub-service under itself. So instructions look okay to me.

That said, if you’re running a recent version of Ubuntu should probably use the systemd method as I don’t think xinetd is installed by default.

So what is wrong with using init.d??

Okay, so I did some more reading and realized this was the Check_mk agent, which does not work for Windows anyway. This is not what I wanted.