Metric Duration Reporting
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
‎01-05-2011 07:49 AM
Does anyone know why there are two duration fields calculated when the metrics definition is setup on a certain table? For example, we are collecting metrics on our problem table and we have two duration fields, mi_duration and prb_calendar_duration. Each field has a different value. The mi_duration field seems to have the correct metric for status duration of a particular problem, but I'm not sure what prb_calendar_duration is collecting or how we would use it.
Thanks for any help.
Rick
- Labels:
-
Service Mapping
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
‎01-05-2011 10:04 AM
I am assuming that you are looking at the "Problem Metric" (problem_metric) table. This is a database view based on the problem, metric definition, and metric instance table. In essence it is one large table of all those tables put together. What happens is that if a column on one table has the same label name as another table, the Database View has to distinguish them somehow. In this case the problem table actually has a column called "calendar_duration" with it's label being "duration". Likewise the MEtric Definition table has a field called "duration" with its label being " Duration". In order to distinguish these in the larger Database view, the view prefaces these names with whatever the variable_prefix contains in the Database view. See System Definition --> Database Views and look at problem_metrics view. Each View Table has a variable prefix. In the end the view calls the Duration field from the problem table prb_calendar_duration, and the duration field from the Metric Instance mi_duration.
The reason why the mi_duration "looks right" to you, is because it is the one that gets populated by the metric instance.