Monitor output from a script

You can define a data source to collect data from a script or external program. Use it when application data is not available through a standard management interface or when you need to provide a summary of multi-row data in a single row. The agent runs the script and collects its output. Each line in the script output is parsed into a row of the resulting data set.

Data can be collected from either a local or remote system. The output of the script or program must contain only values for each attribute within the attribute group. To return multiple rows of data, the data for each row must be separated by a line break. The attributes in each row of data are separated by the separators you define. For more information about separators, see Script parsing and separators

The command can use environment and configuration variables within the user-created script, executable file, query, or system command. The command cannot use environment or configuration variables on the command-line invocation of the command, with only the following exceptions available: AGENT_BIN_DIR, AGENT_ETC_DIR, AGENT_LIB_DIR, CANDLE_HOME, and CANDLEHOME.

The agent monitors script output that is written by using the same locale and code page that the agent runs in.