NRDS Config Manger - Non-x86 devices

[derekb]

We deploy a RaspberryPi or other ARM-based device at customer premise either running NRPE or NRDP.
In the XI NRDS Config Manager when you click CREATE CONFIG and select Linux as the OS, this is the default template you are presented with:

Code: Select all

command[__HOST__]=/usr/local/nagios/libexec/check_ping -H localhost -w 200.0,40% -c 400.0,80% -p 1
command[Check Users]=/usr/local/nagios/libexec/check_users -w 5 -c 10 
command[Check Load]=/usr/local/nagios/libexec/check_load -w 15,10,5 -c 30,25,20
command[Check Disk]=/usr/local/nagios/libexec/check_disk -w 20% -c 10% -p /
command[Check Zombie Procs]=/usr/local/nagios/libexec/check_procs -w 5 -c 10 -s Z
command[Check Total Procs]=/usr/local/nagios/libexec/check_procs -w 150 -c 200

On the RaspberryPi running Raspbian (fork of Debian), Nagios plugins are actually stored in /usr/lib/nagios/plugins, which makes me have to modify each line in the NRDS config template. Knowing this, I just do it each time. On an x64 install of CentOS 6, my plugins dir is /use/lib64/nagios/plugins. As you can probably deduce, neither of these scenarios using Debian or CentOS have the Nagios Plugins directory set as /usr/locla/nagios/libexec which the default NDRS config manager implies.

Is there any way to change this, or add a new template for Linux 32-Bit and Linux 64-Bit so that we don't have to manually change the path in the NRDS config manager every time?

[tgriep]

If you go in to NRDS Config Manager and create a new config with the paths adjusted to your environment, and then use that config for your new installs, would that work for you?

[derekb]

If you go in to NRDS Config Manager and create a new config with the paths adjusted to your environment, and then use that config for your new installs, would that work for you?

I'm not sure what you mean? Currently when I make a new config, I adjust the paths in the "commands" section. Changing the paths in the "additional settings" section didn't seem to do anything.

If there was a "duplicate config" option, I'd be perfectly find just duplicating a working config with the correct paths, but that's not the case here.

[tgriep]

There isn't a way to duplicate a config but you could create a new config, edit the paths and save the config with a different name and when you need to install it, click on the floppy icon for that name and use that config to install the agent. That Client Config will have the changes already done and the client's paths will be in the correct location.
You can create multiple configs on that page for all of your different servers.
Does that help?

[derekb]

There isn't a way to duplicate a config but you could create a new config, edit the paths and save the config with a different name and when you need to install it, click on the floppy icon for that name and use that config to install the agent. That Client Config will have the changes already done and the client's paths will be in the correct location.
You can create multiple configs on that page for all of your different servers.
Does that help?

Maybe I'm not understanding you...
Say I have 100 remote devices at different sites that I want to deploy NRDP to, but they all are doing passive checks for different devices. If I make 1 NRDS config and use the DISK icon to install, all the devices are going to get the same script which is going to say "run this command against ip x.x.x.x" -- that's not going to work for each device.

I'm aware that I can make multiple configs, but I don't want to have to copy/paste a working scrip that has the correct paths. My issue is in the COMMAND section where it sells the remote device where to run ITS script from -- that path is not that same as it is on the XI server.

[tgriep]

The path to the commands are for the remote system and not for the XI system so it should work for you.
You could create a config called Centos6 and have the path set to /use/lib64/nagios/plugins. You would use that config for those systems.

[derekb]

The path to the commands are for the remote system and not for the XI system so it should work for you.
You could create a config called Centos6 and have the path set to /use/lib64/nagios/plugins. You would use that config for those systems.

When you keep saying "the path" can you send me a screen of what section of the config manager you're actually talking about? I think we're looking at different things here...

[tgriep]

This is the screen I am looking at. Is it the same as yours?

[derekb]

This is the screen I am looking at. Is it the same as yours?

Yeah that's where I've been changing the path. But the config that I make there isn't going to be the same for every device I deploy -- the commands will be different based on the device.

I'm use this to monitor IP cameras at a customer site. So one remote device running NRDP might have 10 check_ping commands to various internal IPs.
The next remote device would have multiple check_ping commands, but to different IPs.

So I have to make a config file for each device.

I'm just asking that when we select "Linux", that there be an option that changes the default path that is specified in that 'commands' section so I don't have to change it every time.

[jdalrymple]

You could potentially modify PHP in the component to achieve what you want. I could think of a far better way to handle this if your solution doesn't need to be passive.

Create an nrpe.cfg for all of your remote monitoring hosts with a command something like:

Code: Select all

[check_ipcam]=/path/to/your/check_ping -H $ARG1$

then create a service to apply to all of your cameras (defined as new hosts) using a command that expands like:

Code: Select all

check_nrpe -H <remote monitoring host> -c check_ipcam -a '$HOSTADDRESS$'

To make it a step slicker remote monitoring host could be a host custom object. This would be more flexible for future additional checks in case you wanted to monitor some other metrics that the IP Cams had to offer.