Nagios is one of the best open source server and network monitoring solutions available. Using the flexible nagios framework, you can monitor pretty much anything (including database and custom application). This article, using 4 simple steps, explains how to setup contact definitions who will get notification when a host or service has any issues.
1. Define Generic Contact Template in templates.cfg
Nagios installation gives a default generic contact template that can be used as a reference to build your contacts. Please note that all the directives mentioned in the generic-contact template below are mandatory. So, if you’ve decided not to use the generic-contact template definition in your contacts, you should define all these mandatory definitions inside your contacts yourself.
The following generic-contact is already available under /usr/local/nagios/etc/objects/templates.cfg. Also, the templates.cfg is included in the nagios.cfg by default as shown below.
Please note that any of these directives mentioned in the templates.cfg can be overridden when you define a real contact using this generic-template.
# grep templates /usr/local/nagios/etc/nagios.cfg cfg_file=/usr/local/nagios/etc/objects/templates.cfg Note: generic-contact is available under /usr/local/nagios/etc/objects/templates.cfg define contact{ name generic-contact service_notification_period 24x7 host_notification_period 24x7 service_notification_options w,u,c,r,f,s host_notification_options d,u,r,f,s service_notification_commands notify-service-by-email host_notification_commands notify-host-by-email register 0 }
- Name – This defines the name of the contact template (generic-contact).
- service_notification_period – This defines when nagios can send notification about services issues (for example, Apache down). By default this is 24×7 timeperiod, which is defined under /usr/local/nagios/etc/objects/timeperiods.cfg
- host_notification_period – This defines when nagios can send notification about host issues (for example, server crashed). By default, this is 24×7 timeperiod.
- service_notification_options – This defines the type of service notification that can be sent out. By default this defines all possible service states including flapping events. This also includes the scheduled service downtime activities.
- host_notification_options – This defines the type of host notifications that can be sent out. By default this defines all possible host states including flapping events. This also includes the scheduled host downtime activities.
- service_notification_commands – By default this defines that the contact should get notification about service issues (for example, database down) via email. You can also define additional commands and add it to this directive. For example, you can define your own notify-service-by-sms command.
- host_notification_commands – By default this defines that the contact should get notification about host issues (for example, host down) via email. You can also define additional commands and add it to this directive. For example, you can define your own notify-host-by-sms command.
2. Define Individual Contacts in contacts.cfg
One you’ve confirmed that the generic-contact templates is defined properly, you can start defining individual contacts definition for all the people in your organization who would ever receive any notifications from nagios. Please note that just by defining a contact doesn’t mean that they’ll get notification. Later you have to associate this contact to either a service or host definition as shown in the later sections below. So, feel free to define all possible contacts here. (for example, Developers, DBAs, Sysadmins, IT-Manager, Customer Service Manager, Top Management etc.)
Note: Define these contacts in /usr/local/nagios/etc/objects/contacts.cfg define contact{ contact_name sgupta use generic-contact alias Sanjay Gupta (Developer) email [email protected] pager [email protected] } define contact{ contact_name jbourne use generic-contact alias Jason Bourne (Sysadmin) email [email protected] }
3. Define Contact Groups with Multiple Contacts in contacts.cfg
Once you’ve defined the individual contacts, you can also group them together to send the appropriate notifications. For example, only DBAs needs to be notified about the database down service definition. So, a db-admins group may be required. Also, may be only Unix system administrators needs to be notified when Apache goes down. So, a unix-admins group may be required. Feel free to define as many groups as you think is required. Later you can use these groups in the individual service and host definitions.
Note: Define contact groups in /usr/local/nagios/etc/objects/contacts.cfg define contactgroup{ contactgroup_name db-admins alias Database Administrators members jsmith, jdoe, mraj } define contactgroup{ contactgroup_name unix-admins alias Linux System Administrator members jbourne, dpatel, mshankar }
4. Attach Contact Groups or Individual Contacts to Service and Host Definitions
Once you’ve defined the individual contacts and contact groups, it is time to start attaching them to a specific host or service definition as shown below.
Note: Following host is defined under /usr/local/nagios/etc/objects/servers/email-server.cfg. This can be any host definition file. define host{ use linux-server host_name email-server alias Corporate Email Server address 192.168.1.14 contact_groups unix-admins } Note: Following is defined under /usr/local/nagios/etc/objects/servers/db-server.cfg. This can be any host definition file. define service{ use generic-service host_name prod-db service_description CPU Load contact_groups unix-admins check_command check_nrpe!check_load } define service{ use generic-service host_name prod-db service_description MySQL Database Status contact_groups db-admins check_command check_mysql_db }