Chapter 3. Monitoring Using Nagios

To change the default password for the user nagiosadmin, execute:

# htpasswd -c /etc/nagios/passwd nagiosadmin

# htpasswd /etc/nagios/passwd newUserName

Update the nagiosadmin email address in /etc/nagios/objects/contacts.cfg:

define contact{
    contact_name   nagiosadmin            ; Short name of user
    [...snip...]
    email          yourName@example.com   ; << CHANGE THIS
    }

Verify that the basic configuration is working:

# nagios -v /etc/nagios/nagios.cfg

If errors occur, check the parameters set in /etc/nagios/nagios.cfg.

Ensure that Nagios is started automatically when the system boots:

# chkconfig --add nagios
# chkconfig nagios on

Start up Nagios and restart httpd:

# service httpd restart
# service nagios start

Check your Nagios access by using the following URL in your browser, and using the nagiosadmin user and the password that was set in Step 2:

http://nagiosHostURL/nagios

Write a short script for the item to be monitored (for example, whether a service is running), and place it in the /usr/lib64/nagios/plugins directory.

For example, the following script checks the number of Compute instances, and is stored in a file named nova-list:

#!/bin/env bash
export OS_USERNAME=userName
export OS_TENANT_NAME=tenantName
export OS_PASSWORD=password
export OS_AUTH_URL=http://identityURL:35357/v2.0/

data=$(nova list  2>&1)
rv=$?

if [ "$rv" != "0" ] ; then
    echo $data
    exit $rv
fi

echo "$data" | grep -v -e '--------' -e '| Status |' -e '^$' | wc -l

Ensure the script is executable:

# chmod a+x nova-list

In the /etc/nagios/objects/commands.cfg file, specify a command section for each new script:

define command {
        command_line                   /usr/lib64/nagios/plugins/nova-list
        command_name                   nova-list
}

In the /etc/nagios/objects/localhost.cfg file, define a service for each new item, using the defined command. For example:

define service {
        check_command   nova-list
        host_name       localURL
        name            nova-list
        normal_check_interval   5
        service_description     Number of nova vm instances
        use             generic-service
        }

Restart nagios using:

# service nagios restart

In the /etc/nagios/nrpe.cfg file, add the central Nagios server IP address in the allowed_hosts line:

allowed_hosts=127.0.0.1, NagiosServerIP

In the /etc/nagios/nrpe.cfg file, add any commands to be used to monitor the OpenStack services. For example:

command[keystone]=/usr/lib64/nagios/plugins/check_procs -c 1: -w 3: -C keystone-all

Each defined command can then be specified in the services.cfg file on the Nagios monitoring server.

Start the NRPE service:

# service nrpe start

In the file, specify a host section for each machine on which an OpenStack service is running and should be monitored:

define host{
    use linux-server
    host_name remoteHostName
    alias remoteHostAlias
    address remoteAddress
}

where:

For example:

define host{
  host_name     Server-ABC
  alias     OS-ImageServices
  address   192.168.1.254
}

In the /etc/nagios/nagios.cfg file, under the OBJECT CONFIGURATION FILES section, specify the following line:

cfg_file=/etc/nagios/objects/hosts.cfg

In the /etc/nagios/objects/commands.cfg file, specify the following to handle the use of the check_nrpe plugin with remote scripts or plugins:

define command{
        command_name    check_nrpe
        command_line    $USER1$/check_nrpe -H $HOSTADDRESS$ -c $ARG1$
 }

In the file, specify the following service sections for each remote OpenStack host to be monitored:

##Basic remote checks#############
##Remember that remoteHostName is defined in the hosts.cfg file.

define service{
    use generic-service
    host_name remoteHostName
    service_description PING
    check_command check_ping!100.0,20%!500.0,60%
}

define service{
    use generic-service
    host_name remoteHostName
    service_description Load Average
    check_command check_nrpe!check_load
}

##OpenStack Service Checks#######
define service{
    use generic-service
    host_name remoteHostName
    service_description Identity Service
    check_command check_nrpe!keystone
}

The above sections ensure that a server heartbeat, load check, and the OpenStack Identity service status are reported back to the Nagios server. All OpenStack services can be reported, just ensure that a matching command is specified in the remote server’s nrpe.cfg file.

In the /etc/nagios/nagios.cfg file, under the OBJECT CONFIGURATION FILES section, specify the following line:

cfg_file=/etc/nagios/objects/services.cfg

Verify that the updated configuration is working:

# nagios -v /etc/nagios/nagios.cfg

If errors occur, check the parameters set in /etc/nagios/nagios.cfg, /etc/nagios/services.cfg, and /etc/nagios/hosts.cfg.

Restart Nagios:

# service nagios restart

Log in to the Nagios dashboard again by using the following URL in your browser, and using the nagiosadmin user and the password that was set in the beginning:

http://nagiosHostURL/nagios