Linux Networking Cookbook

13.13. Monitoring Name Services

Problem

You want Nagios to monitor your DNS and DHCP servers.

Solution

Add the DNS and DHCP command definitions to commands.cfg if they do not already exist, then create new host and service definitions just like we did in Recipe 13.9, Recipe 13.10, Recipe 13.11 through Recipe 13.12.

DNS uses an ordinary command definition:

	# commands.cfg
	# 'check_dns' command definition
	define command{
	        command_name    check_dns
	        command_line    $USER1$/check_dns -H $ARG1$
	        }

Then, define your DNS query parameters in the service definition, specifying a domain or hostname to use for testing the server:

	check_command     check_dns!host.domain.com

DHCP is bit more work to set up because the check_dhcp plug-in requires root privileges to get full access to the network interface. Give it the SUID bit, owned by root, in the nagios group:

	# chown root:nagios check_dhcp
	# chmod 4750 check_dhcp

To query DHCP functionality on the network, don't specify any options:

	# 'check_dhcp' command definition
	define command{
	        command_name    check_dhcp
	        command_line    $USER1$/check_dhcp
	        }

Add the -s option to specify a server to check:

	$USER1$/check_dns -s $ARG1$

Then, specify the server in the service definition:

	check_command      check_dhcp!12.34.56.78

Discussion

check_dns requires that you specify a hostname to check. This can be a local host or a remote host that you can reasonably expect to be up, such as Google, Yahoo, or your ISP. Be nice—don't bombard other people's servers. By default, check_dns queries the ...