In this article, I am going to explain how to implement a monitoring on your system, using an agentless solution (without Nagios’ nrpe agents), and I will underline some advantages of this useful solution. The target machine will be Linux or Unix Operating Systems.
The need can arise from the fact that some of the Operating Systems do not provide the Nagios’ agent or the customer does not intend to install software of third parties, as the system is considered too critical or the customer wants to avoid discussions with the assistance offered by the system’s vendor.
The basic idea is very similar to the one utilized by Nagios’ agent. Having recourse to the features available in the SNMP protocol (Net-SNMP package) installed by default on the main Operating Systems “Unix-like”, it is possible to define a script and run it with a specific OID via the SNMP protocol.
In this article, I will use Oracle Solaris 11 as the target Operating System. However, with some simple configuration changes it is possible to adapt the configuration also to other systems.
The configuration is quite simple. Here are the main steps:
The following is a screenshot that shows some of the steps:
There are no limits to the potentialities of these checks, except for the ability to write its own scripts.
This package will allow you to increase your checks and with some simple keywords you will be able to expand the features available and make recourse to further controls provided by the package. For instance, you can verify a partition of your target system, simply by adding the following configuration in the snmpd.conf file. For example, for the partition var you must add:
The integration with NetEye is very easy as well, thanks to the plugin check_snmp, that is already installed on our system.
All you need now is to try this opportunity!