Localhost Settings

[nickanderson1982]

I can setup for all my servers except for local nagios server. How would I monitor the following items on the local server?
define host {
host_name XXXXXX
address 172.24.x.x
check_command check_ncpa!-t xxxxx -P 5693 -M system/agent_version
max_check_attempts 5
check_interval 5
retry_interval 1
check_period 24x7
contact_groups admins communication
notification_interval 60
notification_period 24x7
notifications_enabled 1
icon_image ncpa.png
statusmap_image ncpa.png
register 1
}

define service {
host_name xxxxxxxx
service_description CPU Usage
check_command check_ncpa!-t xxxxxxx -P 5693 -M cpu/percent -w 20 -c 40 -q 'aggregate=avg'
max_check_attempts 5
check_interval 5
retry_interval 1
check_period 24x7
notification_interval 60
notification_period 24x7
contact_groups admins communication
register 1
}

define service {
host_name xxxxxxx
service_description Memory Usage
check_command check_ncpa!-t xxxxxxxx -P 5693 -M memory/virtual -w 50 -c 80 -u G
max_check_attempts 5
check_interval 5
retry_interval 1
check_period 24x7
notification_interval 60
notification_period 24x7
contact_groups admins communication
register 1
}

define service {
host_name xxxxxxxxxxx
service_description Event Processor [Test]
check_command check_ncpa!-t xxxxxx -P 5693 -M 'services' -q 'service=Event Processor [Test],status=running'
max_check_attempts 5
check_interval 5
retry_interval 1
check_period 24x7
notification_interval 60
notification_period 24x7
contact_groups admins
register 1
}

[cdienger]

What isn't working exactly? I imported the config and besides having to make sure that the host_name in all the definitions are the same and configuring the address line to use localhost instead of 172.24.x.x , this loaded okay for me. Is NCPA installed on the localhost?

[nickanderson1982]

I tested out the following:
define service {
host_name localhost
service_description CPU Usage
check_command check_ncpa!-t xxxxxxx -P 5693 -M cpu/percent -w 20 -c 40 -q 'aggregate=avg'
max_check_attempts 5
check_interval 5
retry_interval 1
check_period 24x7
notification_interval 60
notification_period 24x7
contact_groups admins communication
register 1
}

In the Nagios GUI I get the following:
UNKNOWN: Error occurred while running the plugin. Use the verbose flag for more details.

What would I need to change to make the cpu check work?

[mbellerue]

That definition looks good. Is NCPA installed and running on the system that is running Nagios? Does the NCPA interface come up when you browse to https://<IPAddress>:5693

What do you get when you go to the command line and run,

Code: Select all

/usr/local/nagios/libexec/check_ncpa.py -H <IPAddress> -t <NCPAToken> -P 5693 -M cpu/percent -w 20 -c 40 -q 'aggregate=avg' -v

[nickanderson1982]

So you have to install NCPA on the local host that is running Nagios?

[nickanderson1982]

I went and installed NCPA on the localhost that runs Nagios.
rpm -Uvh https://repo.nagios.com/nagios/8/nagios ... noarch.rpm
yum install ncpa -y

Edit the NCPA Configuration
/usr/local/ncpa/etc/ncpa.cfg
change the community_string = xxxxx (used our token)

After saving the change, restart the NCPA listener service:
/etc/init.d/ncpa_listener restart

Added the following edit to the localhost.cfg:
define service {
host_name localhost
service_description CPU Usage
check_command check_ncpa!-t xxxxxxx -P 5693 -M cpu/percent -w 20 -c 40 -q 'aggregate=avg'
max_check_attempts 5
check_interval 5
retry_interval 1
check_period 24x7
notification_interval 60
notification_period 24x7
contact_groups admins communication
register 1
}

define service {
host_name localhost
service_description Memory Usage
check_command check_ncpa!-t xxxxxx -P 5693 -M memory/virtual -w 50 -c 80 -u G
max_check_attempts 5
check_interval 5
retry_interval 1
check_period 24x7
notification_interval 60
notification_period 24x7
contact_groups admins communication
register 1
}

define service {
host_name localhost
service_description Wazuh
check_command check_ncpa!-t xxxxxxx -P 5693 -M 'services' -q 'service=wazuh-agent.service,status=running'
max_check_attempts 5
check_interval 5
retry_interval 1
check_period 24x7
notification_interval 60
notification_period 24x7
contact_groups admins
register 1
}

------------
The results were not good.
All the new services are unknown: UNKNOWN: Error occurred while running the plugin. Use the verbose flag for more details.

/usr/local/nagios/libexec/check_ncpa.py -H 172.24.xx.xx -t xxxxxxx -P 5693 -M cpu/percent -w 20 -c 40 -q 'aggregate=avg' -v
Connecting to: https://172.24.xx.xx:5693/api/cpu/perce ... regate=avg
An error occurred:<urlopen error [Errno 111] Connection refused>

Shouldn't I be editing the localhost.cfg in a different manner? How would I check for cpu, memory and disk? I'm confused as to what to do for the localhost running nagios

[nickanderson1982]

Does anyone know how I am supposed to configure the localhost file for the linux server that runs nagios? What would be the syntax to monitor the cpu, mem and harddrive locally?

[mbellerue]

Apologies for the delay in response.

Okay, so there was a misunderstanding. I didn't see that the host you were trying to monitor was the localhost that runs Nagios. NCPA is an agent that allows Nagios to talk to a remote server.

To monitor the machine that is running Nagios, you can use the plugins that are already installed locally. If you check the /usr/local/nagios/etc/objects/commands.cfg file, you should see some commands defined, and the definition shows you how to use them in your localhost.cfg file. For example,
commands.cfg

Code: Select all

define command {
    command_name    check_ping
    command_line    $USER1$/check_ping -H $HOSTADDRESS$ -w $ARG1$ -c $ARG2$ -p 5
}

localhost.cfg

Code: Select all

define service {
    host_name              localhost
    service_description    PING
    use                    local-service
    check_command          check_ping!100.0,20%!500.0,60%
    register               1
}

The check_command launches the check_ping command defined in commands.cfg, and then we have 2 arguments separated by !. Looking at the commands.cfg definition, you can see that the argument 100.0,20% is used for the -w flag (in almost every plugin, this will be your warning threshold flag), and then 500.0,60% is used for the -c flag (in almost every plugin, this is the critical threshold flag).

If you want to know what potential values or flags could be used, you can find all of the plugins at /usr/local/nagios/libexec/. So you can run /usr/local/nagios/libexec/check_ping --help and find out how to use the plugin.

More information can be had at the following link. I highly recommend keeping this link handy, especially the Object Definitions section, as Nagios configuration files can be pretty overwhelming.
https://assets.nagios.com/downloads/nag ... n/toc.html

[nickanderson1982]

I just reviewed the file. Nothing stands out for me to setup memory, cpu and hard drive checks. How would I manually set those up in the commands.cfg?

[Box293]

Just taking a step backwards. There's nothing wrong with using NCPA installed on the Nagios Server to monitor it, doing it this way saves re-inventing the wheel.

An error occurred:<urlopen error [Errno 111] Connection refused>

I suspect you simply need to add a firewall rule to your Nagios server to allow 5693 to connect, even though it is a local connection. Check page 15 of this guide that has firewall port opening commands:

https://assets.nagios.com/downloads/ncp ... g-NCPA.pdf