Skip to content
Anonymous edited this page Dec 12, 2013 · 2 revisions

Overview

Selector engine aims to perform aggregation operation on many event, providing this way a simple aggregated information.

Goals

Allow user to define it's own aggregation information. For now, selector allows to compute the worst state on an event set.


##Synchronous consume dispatch method

This method listen to the rabbitmq bus and each time an event becomes available, a crecord is retrieved from this event. The crecord provide infomation about witch event implied in the information aggregation. information are:

  • a list of event id to include
  • a list of event id to exclude
  • a cfilter describing a mongo style database query which allows custom event selection

These information make selector compute a set of distinct status values for these event. From these distinct values, the worst state appears in a natural way and then, a new event is produced with the selector information and some other information relative to the selector description made through the UI.

Clone this wiki locally