Nagios : Quick install and configuration guide

Version 1
April 14, 2016
Abstract
Nagios is an efficient tool for network and system monitoring. However, first install and configuration is pretty intricate since it requires to understand a major part of the official documentation (250
pages). This tutorial is for ones who want to quickly build a working configuration so as to monitor
simple networks including NFSv4 clients and servers. This guide goes through installation and basic
configuration of Nagios.

Contents
1 Install 2
1.1 Copy distribution and plug in . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 
1.2 Create user . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 
1.3 Configuration and compilation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2 Configuration 4
2.1 Time periods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 
2.2 Contacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 
2.3 Hosts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.4 Hostgroups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 
2.5 Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 
2.6 Web server configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 
3 Run Nagios 9

1 Install

1.1 Copydistributionandplugin
First, you must get the latest Nagios release (nagios-x.y.tar.gz) and the official Nagios plugins (nagios-plugins-x.y.z) from http://www.nagios.org/download/. I also advise to get the official Nagios documentation(http://www.nagios.org/docs/)ifyouplantotunetheconfigurationexplainedhere.

Unpack the tarball :

tar xzf nagios-2.0.tar.gz tar xzf nagios-plugins-1.4.2.tar.gz 1.2

1.2 Createuser
First, we need to create the user Nagios will run under :
adduser nagios mkdir /usr/local/nagios                  chown nagios.nagios /usr/local/nagios

You will probably want to issue external commands from the web interface. So you need to identify the user your web-server run as. For Apache, you can get it with :
grep “^User” /etc/httpd/conf/httpd.conf
By default, the user is Apache. See /etc/passwd file or report to you web-server installation if you are running another web server. Also note that the path to httpd.conf can vary depending on your system.

We need to create a group whose members include the user your web server is running as and the user Nagios is running as.

/usr/sbin/groupadd nagcmd    /usr/sbin/usermod -G nagcmd apache    /usr/sbin/usermod -G nagcmd nagios
Recall to replace apache by the user name you obtained at last step.

1.3 Configurationandcompilation
We can now configure and install Nagios. Go to the directory where you untared the distribution and run :
./configure –prefix=/usr/local/nagios –with-nagios-user=nagios –with-nagios-group=nagios –with-command-group=nagios
Depending on your system, you should get a configuration like the following :

Configuration summary for nagios 2.0 02-07-2006 ***:
    General Options:

————————

Nagios executable: nagios

Nagios user/group: nagios,nagios

Command user/group: nagios,nagios

Embedded Perl: no

Event Broker: yes

Install ${prefix}: /usr/local/nagios

Lock file: ${prefix}/var/nagios.lock

Init directory: /etc/rc.d/init.d

Host OS: linux-gnu Web Interface Options:

————————

HTML URL: http://localhost/nagios/

CGI URL: http://localhost/nagios/cgi-bin/

Traceroute (used by WAP): /bin/traceroute

Now let’s compile and install :
make all

make install

make install-init

make install-config

make install-init allows to run nagios at startup (/etc/rc.d/init.d/nagios). This does not always work (I had problems on Fedora Core 4) but we see another way to run it at startup later on. make install-config installs the sample configuration files. This will be useful to setup a basic configuration.

In order for nagios to be of any use to you, you need to install some plugins. First, install the official plugins. Go to the directory where you untared nagios-plugin-x.y.z and run :
./configure make make install

2 Configuration

Normally, you should already have some configuration sample files in /usr/local/nagios/etc. Else, try to run again make install-config. Files should be called xxxx.cfg-sample, so you need to rename them :
cd /usr/local/nagios/etc

mv bigger.cfg-sample bigger.cfg (…)

The entry point is through nagios.cfg. This file is run first when nagios starts. You may not need a lot of modifications in this file.
log_file=/usr/local/nagios/var/nagios.log

indicates where the logs will go. Set the appropriate owner of the log file :
chown nagios.nagios /usr/local/nagios/var/nagios.log

Other configuration files are called from nagios.cfg, for instance at lines :
cfg_file=/usr/local/nagios/etc/checkcommands.cfg

cfg_file=/usr/local/nagios/etc/misccommands.cfg

Those files contain commands definition.

Andfinally,ourownconfigurationwillbewritteninanewfile(ma_conf.cfg forinstance)soweneed to replace :
cfg_file=/usr/local/nagios/etc/minimal.cfg
By :
cfg_file=/usr/local/nagios/etc/ma_conf.cfg
To create our configuration, we will take minimal.cfg as the basis :
cp minimal.cfg ma_conf.cfg
And now let’s edit it to setup a configuration for a simple network with one monitoring station and one NFSv4 server.

2.1 Timeperiods

Time periods are used to define some monitoring periods. For the time being we can keep 24*7 but you may need to change it -for instance not to receive alarms if your NFS servers are down during night.

2.2 Contacts

You will certainly need to add some contacts,which are persons in charge of monitoring the network. By default they will receive notification in 24*7 by e-mail.

define contact{

contact_name adminNFS
alias Jonathan Lyard

service_notification_period 24×7

host_notification_period 24×7 s

ervice_notification_options w,u,c,r

host_notification_options d,r

service_notification_commands notify-by-email

host_notification_commands host-notify-by-email

    email adminNFS@localhost
    }

Now let’s create a contact group. This one will be used to indicate which contacts should be notified for each issue. Here is only one group, but feel free to add all groups you need (for instance if the ones in charge of NFS administration are different from network administrators).

define contactgroup{
contactgroup_name admins
alias Nagios Administrators
members nagios-admin,adminNFS
}

2.3 Hosts
In addition to localhost, you must specify all hosts your are monitoring. I assume they are all on-link (in the same LAN as your monitoring station). Else, see Nagios official documentation to get it working through routers (use parents option). Here is what is required in order to monitor a machine called nfs2 with IP address 192.168.0.5. Alarms from this host will be reported to the “admins” group.

define host{
use generic-host ; Name of host template to use

host_name nfs2 alias nfs2

address 192.168.0.5

check_command check-host-alive

max_check_attempts 10 notification_interval 120

notification_period 24×7

notification_options d,r

contact_groups admins

}

2.4 Hostgroups

In this basic configuration, we will only have one hostgroup. You can define several hostgroups to group hosts together in Nagios interface.

define hostgroup{

hostgroup_name nfsv4config

lias NFSv4 test configuration

members localhost,nfs2

}

2.5 Services

Finally, one can find services definition. Services do not necessarily mean real services that run on host (FTP, POP, HTTP…) but also some other types of metric associated with the host (response to ping, number of logged-in user, free disk space…). For instance, to monitor connectivity for localhost and our nfs2 server, we need the following service definition :

define service{
use generic-service ; Name of service template to use
host_name localhost,nfs2
service_description PING
is_volatile 0
check_period 24×7
max_check_attempts 4
normal_check_interval 5
retry_check_interval 1
contact_groups admins
notification_options w,u,c,r
notification_interval 960
notification_period 24×7
check_command check_ping!100.0,20%!500.0,60%
}

Later on, we will also define a service to monitor NFSv4.

2.6 Webserverconfiguration

In this step, you will learn how to configure the web-server. I assume you are using Apache, but the configuration may be similar for other web-servers. Check your own web-server’s documentation to put it at work.

Add the following lines in httpd.conf (in general /etc/httpd/conf/httpd.conf).

ScriptAlias /nagios/cgi-bin /usr/local/nagios/sbin

Directory /usr/local/nagios/sbin

#       SSLRequireSSL
       Options ExecCGI

       AllowOverride None

       Order allow,deny

       Allow from all

       # Order deny,allow

   # Deny from all

   # Allow from 127.0.0.1

   AuthName Nagios Access

   AuthType Basic

   AuthUserFile /usr/local/nagios/etc/htpasswd.users Require valid-user /Directory

Alias /nagios /usr/local/nagios/share

    Directory /usr/local/nagios/share
    #     SSLRequireSSL
    Options None

    AllowOverride None

Order allow,deny

Allow from all

#     Order deny,allow

#     Deny from all

#     Allow from 127.0.0.1

     AuthName Nagios Access

     AuthType Basic

     AuthUserFile /usr/local/nagios/etc/htpasswd.users

     Require valid-user

     /Directory

This configuration enable authentication. The first directory entry is for CGI scripts, the second for html pages. Of course, you can tune this configuration to fit your requirements.

You can also disable authentication by setting use_authentication=0in/usr/local/nagios/etc/cgi.cfg but then, you will not be allowed to run remote commands. Use it only if you have a problem with authentication.

You now need to create one or more users being able to authenticate. For the first one (-c to create the file htpasswd.users) :

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

and for (possible) next users :

htpasswd /usr/local/nagios/etc/htpasswd.users nagiosadmin2

Then, you need to give some rights to the users.

You can do it by editing the file /usr/local/nagios/etc/cgi.cfg.

For instance :
authorized_for_system_information=nagiosadmin,nagiosadmin2

And so on for all line you want to grant privilege to nagiosadmin and nagiosadmin2. I have uncommented all authorized line and give access to nagiosadmin. I advised to at least do the same if want to be able to configure everything from the web interface.

For more details on the access rights, read the chapter”Authentication and Authorization in the CGIs”in Nagios official documentation.

Restart your web browser to take modifications into account :
/usr/sbin/httpd -k restart

You should be able to point http://localhost in your web browser provingt hat your web serveris up.

You should also be able to point http://localhost/nagios/ and authenticate yourself with one of the user created previously (htpasswd). At this time, the CGI will not works since Nagios is not running.
3 RunNagios
To run Nagios, you can use the following command to make it start as a foreground process:
/usr/local/nagios/bin/nagios      /usr/local/nagios/etc/nagios.cfg &

If make install-init failed to make Nagios run at startup you can add this command to/etc/rc.d/rc.local (here on FC4 but path and filename can vary depending on distribution). To check that all configuration files are alright, you can issue :

/usr/local/nagios/bin/nagios -v     /usr/local/nagios/etc/nagios.cfg

Finally, a useful command to restart Nagios and therefore consider modifications in the configuration files :

killall -SIGHUP nagios

Enjoy monitoring systems with Nagios!
JonathanLyard{jonathan.lyard@bull.net}