Normal Datapoints

You should define normal datapoints if the data you'd like to monitor can be extracted from the raw output collected from your devices.  The configuration fields for normal datapoints depend on the DataSource's collection method, but the following fields are common to normal datapoint configurations for most DataSource collection types (not all):

Metric Type

Each datapoint must be a Gauge, a Counter, or a Derive.

  • Gauge: stores the reported value directly.  e.g. data whose raw values are 100, 130, 90 would be stored as 100, 130, 90.
  • Counter: interprets collected values at a rate of occurrence and stores data as a rate per second. Counters correct for counter wraps in collected data.  
  • Derive: similar to counters, except they do not correct for counter wraps. They can return a negative value.

Counter Example

If an interface counter that is sampled once every 120 seconds reports values of 600, 1800, 2400, 3600, it would then store these as 10 ((1800-600)/120); 5 ((2400-1800)/120) and 10 ((3600-2400)/120).

Counters also account for counter wraps. Consider the following example: if one sample is 4294967290, and the next, one second later, is 6, the counter will store a value of 12, and assume the counter wrapped on the 32 bit limit.  This behavior can lead to incorrect spikes in data if a system is restarted, and counters start again at zero - the counter data type may assume a very rapid rate caused a counter wrap, so will store a huge value.  For this reason, it is advisable to set Max values with Counter data types.  More significantly, it is rare that you need to use a counter type. It is almost always better to use a derive type.  Unless you are dealing with a datapoint that will wrap frequently (e.g. for a gigabit interface and using a 32 bit counter), you should use a derive instead of counter.

Valid Value Range

Designates the range of datapoint values for which a Collector will return data.  Values outside this range will be reported as "No Data."

This functions as a data normalization field to filter outliers and incorrect datapoint calculations.

Raw Metric

This field name changes depending on the collection method (to contain the JMX Mbean if the JMX collector is used, or the SNMP OID if the snmp collector is used, etc).  For all collection methods, this field specifies what raw output should be collected. (e.g. use the HTTP response codes, or the body of the HTML returned.)  For detailed information about the raw metric field for a specific data collection method, see the help page for that collection method.

Post Processor

How the data should be extracted from the raw output.  The options available are specific to the DataSource's collection method.  For detailed information about each post processor method supported, see this page.

Script-based Datapoints

If you are using a script collector, you will need to choose a source for your datapoint from the following options:

  1. Exit code of the script. Usually, 0 means the script finished correctly. Non-zero means the script finished with errors: this option will display whether or not your script ran and returned data successfully.
  2. Execution time of script in milliseconds: displays how long your script ran. Exceptionally long execution times can indicate the script placed excessive load on your device or an inefficient script.
  3. Content the script writes to the standard output: the most common source for datapoints collected via script. Selecting this option will cause several configuration fields to appear, all of which are used to determine how LogicMonitor will interpret the raw data written to the standard output.