IT Infrastructure Monitor with Nagios

Difficulty Level : 4 out of 5 

Ever heard of Nagios, Zabbix, Cacti or even Icinga? If I name it, you probably heard it. These are the famous monitoring tools for IT Infrastructure today and, majority of them are built for one intention, to monitor your infrastructure be it Servers, Network Switches and Routers or even printers. They are really famous out there because they are actually free or licensed under GPL/GNU GPL which means you can actually study, share or even improve it should you have the capability to do so.

The topic I cover today would be Nagios Core IV, do not get confused with another product Nagios XI, in fact both of them are by the same developer and one is intended for enterprise that pays and it is not cheap though. It is ever important skill for an IT professionals to learn about the implementation of these tool into their existing system, remember that the monitoring can always be automated in any day now but the implementation will be not at any time now.

 Have a look at the link below should you want to know more about what Nagios XI offers. https://www.nagios.org/downloads/nagios-core/

 I would say Nagios Core itself is suffice for majority of the system administrator out there and Nagios XI would be the icing on top of the cake as it provides better interfaces and report functionality and of course the support.

I have made this guide into 7 steps, the topic will covers the installation of Nagios CORE, Nagios Plugin (the library of feature for the CORE) and NRPE plugins (the plugin used for monitoring).

Step 1 :

Prepare your server you want Nagios CORE to be installed with. Make sure the server come ready with LAMP (Linux, Apache, MySQL, PHP). You can install of of them by using yum install for Red-Hat/Centos or apt-get for Debian/Ubuntu.

i. Create a user 'nagios'

- sudo useradd nagios

ii. Create a group 'nagcmd'

- sudo groupadd nagcmd

iii. Make the user part of the group created

- sudo usermod -a -G nagcmd nagios

* For learners, I suggest you to build the LAMP stack on your own local machine. Should you need any guidance in it please write in to me and I will try to help you out. You may run it in virtual machine as well with VMWare Workstation Player.

Step 2 :

Before we go for Nagios installation, make sure we have all the dependencies needed for NagiosCore (NC). The usual requirement for NC is as follow :

build-essential libgd2-xpm-dev openssl libssl-dev xinetd apache2-utils unzip

- You can use yum or apt-get to install them

* Some people may experience problem running yum or apt-get. Do not worry, it is likely that your network is blocked with a firewall or even your school or company is blocking them. Try to use proxy instead and do Google it out. 

Step 3 :

Installing NagiosCORE. Start by downloading NagiosCORE from the link below :

https://assets.nagios.com/downloads/nagioscore/releases/xxxxxxx 

Just get any latest version you can find and of course like all Linux admin does, choose the [Stable] one.

Extract the NagiosCore in a directory (temporary for installation use)

- tar xvf nagios-*.tar.gz

Change to the directory

- cd nagios-*

Set up Email server before using Nagios

You can use PostFix / EXIM or SMTP  servers. I recommend you use PostFix (/usr/bin/mail) for the ease of mind. Same thing you can install it by using yum install mailutils or apt-get install mailutils.

Lets build the configuration file by running the command below, you can see I added a parameter '--with-mail=/usr/sbin/sendmail' because i wanted to use PostFix, remove it or change it if you wanna use the other one.

- ./configure --with-nagios-group=nagios --with-command-group=nagcmd --with-mail=/usr/sbin/sendmail

After configuration is done the script actually generates some file inside this directory with all the configuration specified. You can run this next to compile them.

- make all

This will compile the NagiosCore. Next we need to initiate all the Nagios script by running the command below :

sudo make install

sudo make install-commandmode

sudo make install-init

sudo make install-config

sudo /usr/bin/install -c -m 644 sample-config/httpd.conf /etc/apache2/sites-available/nagios.conf

Depending on your apache group and user. You will next need to set the apache user to the group of nagios by running the following :

- sudo usermod -G nagcmd

www-datawww-data is for my vanilla Apache2 ... It might be different in your environment and do check them out. 

Step 4 :

Installing Nagios Plugin

Just like step 3, we download the Nagios Plugin this time round. You can use curl -L -O or wget. Its up to you.

Link : http://nagios-plugins.org/download/nagios-plugins-xxxxx.tar.gz <-- get latest version

Extract it

- tar xvf nagios-plugins-*.tar.gz

Change to the directory

- cd nagios-plugins-*

Configure the plugin

- ./configure --with-nagios-user=nagios --with-nagios-group=nagios --with-openssl

(This will configure the nagios plugin with the nagios user and their group)

Compile it

- make

Install It

- sudo make install 

Its done for plugin, PLUGIN is not for monitoring yet. Plugin is just the library for the NagiosCORE. Next we talk about NRPE or the monitoring part..

Step 5

Installing NRPE

NRPE is the part used by NagiosCore for monitoring the server. In this scenario, we will setup the NC to monitor ourself (localhost). Just like Step 3 and Step 4, download the NRPE from the link in a temporary directory u want.

Link : https://sourceforge.net/projects/nagios/files/nrpe-2.x/ 

 Same thing, lets get the latest and stable

Extract it

- tar xvf nrpe-*.tar.gz

Change to Directory

- cd nrpe-*

Configure it

- ./configure --enable-command-args --with-nagios-user=nagios --with-nagios-group=nagios --with-ssl=/usr/bin/openssl --with-ssl-lib=/usr/lib/x86_64-linux-gnu

Build and Compile

make all

sudo make install

sudo make install-xinetd

sudo make install-daemon-config

Let's modify the xinetd file after that, you might need to install xinetd yourself in case you dont see one. Can be done with yum or apt-get. Xinetd should be at the following path :

sudo nano /etc/xinetd.d/nrpe

Look for the portion "only_from = [IP Address]

modify it by adding your own IP address into this section wiith (comma ,) if you want more IP.

Remember to Save and Exit it.

Restart xinetd service now with sudo service xinetd restart (systemctl might be needed if your on RHEL 7.x)

Now Nagios 4 Core , Plugin, NRPE is installed !

Step 6

Configuring Nagios now with our server environment.

Open your Nagios Configuration file for editing.

- sudo nano /usr/local/nagios/etc/nagios.cfg

Locate this section

#cfg_dir=/usr/local/nagios/etc/servers

* This is actually the path to your configuration file for Nagios to know where can I read the server configuration. Uncomment it so that nagios will be ready to use this config file. Remember to Save and Exit.

Now create a directory based on the config file path you set above. In this case :

- sudo mkdir /usr/local/nagios/etc/servers

* Remember how Exabytes set each server a config file, its for this purpose.

Next, you will configure the Contact Email for your NC

- sudo nano /usr/local/nagios/etc/objects/contacts.cfg

* Look for the section email and change it to your Email Address (default is nagios@localhost)

Next, configure the check_nrpe.

This is useful in order to check your NRPE version (the monitoring plugin). One of the service we want to monitor. You can of course ignore it if you want to.

Edit the file : - sudo nano /usr/local/nagios/etc/objects/commands.cfg

Add in the following at the end :

define command{       

command_name check_nrpe       

command_line $USER1$/check_nrpe -H $HOSTADDRESS$ -c $ARG1$

}

Save and Exit

Configure Apache

Yes, its lengthy process but Apache will need to be setup as well since we are using the web interface to view Nagios. (eg. www.pikachuben.com/nagios). Make sure Apache rewrite and cgi is enabled.

- sudo a2enmod rewrite

- sudo a2enmod cgi

Create a user you want to use for nagios monitoring, in my case nagiosadmin

- sudo htpasswd -c /usr/local/nagios/etc/htpasswd.users nagiosadmin

(Put a password you can remember when it prompts you and please remember it)

 Take note if you want to use other user, you will need to modify every section in the file /usr/local/nagios/etc/cgi.cfg)

Create a symbolic link for nagios.conf to the site-enabled directory. ( This is for debian/ubuntu since the Apache works a bit different) Skip this part if your on Redhat/Centos

- sudo ln -s /etc/apache2/sites-available/nagios.conf /etc/apache2/sites-enabled/

Lets start nagios and restart apache to take effect !

- sudo service nagios start

- sudo service apache2 restart

To setup NAgios to start when boot

_ sudo ln -s /etc/init.d/nagios /etc/rcS.d/S99nagios

Basically your Nagios is ready to kick now, you can access to pikachuben.com/nagios and try to login now !

For Redhat / Centos

- You do not need to make site-available symbolic link actually since Apache works differently in Centos/Redhat. Just run the following :

- sudo systemctl daemon-reload

- sudo systemctl start nagios.service

- sudo systemctl restart httpd.service

To make nagios start during boot

- sudo chkconfig nagios on

Done for RedHat !

Step 7

Monitoring a remote Host

Go to the server  you want to monitor. Install both the following with yum or apt-get

- nagios-plugins

- nagios-nrpe-server

Modify the following file after the installation above.

sudo nano /etc/nagios/nrpe.cfg

Find this section "allowed_host"

Add in the IP address of your remote server like this allowed_hosts=127.0.0.1,[Your_NAGIOSCORE_IP_ADDRESS]

Save and Exit

Now lets try to monitor the disk !

type this

- df -h /

You will see your disk name most likely /dev/vda or /dev/sda or others.

Open up this file /etc/nagios/nrpe.cfg

Look for these section : server_address / allowed_hosts / command[check_hda1]

Change the /dev/hda1 to the disk name above where u see ur actual disk name is. Do not change the front part [check_hda1].

Change the server_address to your own server IP address (your current server IP not the nagios core server)

Change the allowed_hosts to your NAGIOS CORE server IP address.

 After that, restart the NRPE by using

sudo service nagios-nre-server restart

You probably ask what are you doing now. To make it simple to understand, you are actually creating a NRPE plugin for the server right now. This command can then be called from the main NAGIOS core server which will be explained next. Some feature do not need these custom call at all. PING is special you don't need to set it up. Others yes like HTTPD, EXIM etc etc.

NOW you will need to go to your NAGIOS Core server and setup the monitoring based on what you have done in the host server (host = the server you want to monitor and the NRPE you just setup there)

- Go to the following path :

/usr/local/nagios/etc/servers/

* Remember this path we created early on for the preparation of any future server? Now you will use it.

Create a cfg file here, use the server name which you are monitoring for more convenient, you can use anything though. In my case, theorionsbelt.cfg

Add the following into it

define host {

use                             linux-server       
host_name                       [Your_remoteserver_name]     
alias                           [Just a name for your remote server]     
address                         [Remote server IP Address]       
max_check_attempts              5       
check_period                    24x7       
notification_interval           30       
notification_period             24x7

}

This portion is needed because it will define which is the remote host you wanna monitor. Look at the parameter. You need to modify it based on what the remote server is.

Take note that the notification interval and period controls how frequent the notification is being sent out. 

If you want to monitor Ping, you can add the following line on the bottom of the same file

define service {       

use                             generic-service       
host_name                       yourhost       
service_description             PING       
check_command                   check_ping!100.0,20%!500.0,60%

}

This will monitor Ping

Remember the NRPE we made just now ? You can build the same thing here creating a new function based on it.

define service {       

use                             generic-service      
host_name                       yourhost      
service_description             PING      
check_command                   check_nrpe!checkhda1

}

Notice the check_command section follow by check_nrpe

Do not modify check_command.

check_nrpe is used to initiate to tell core to go and look for the nrpe.cfg file in the remote server, then check_hda to tell the core to look at the command for check_hda1 which I tell you not to change it earlier on.

To take effect after the change, you just restart it with sudo service nagios reload

Login to your web interface next and VOILA ... Ping and DISK monitoring is there !!!!

MISSION ACCOMPLISHED !

For windows monitoring you need NSClient++

The post have no image yet, will most likely upload them soon. I used to work with a company running on Nagios system and the reliability of this software is definitely good. Majority of my work today covers in Compute Virtualization. As a system administrator or system architect, it is always crucial for you to setup monitoring tools for your subordinates and the team.

Prepared and written by Ruben.Rubin.Ruban.Reuben.Rubun(?).