This is the primary function of the monitoring agent: submit the readings to the main server. Each sensor produces two values—the application return code and a single floating-point value that represents this particular reading. If there are multiple values to be returned, they must be split into two separate checks and each check must be called separately.
The agent receives an instruction to run the check, and each instruction contains two parameters: the sensor name and the options string. The sensor name is used to find the sensor code; the directory containing the sensor application must have the same name as the sensor. In addition to this convention, the sensor application name must match the name defined in the client configuration file. When the agent receives the instruction, it starts the sensor application and passes the option string to it as additional parameters.
Was this article helpful?