Starting from results of a job execution, an anomalies list can be created to highlight where the performance problems are located. Some recommendation are given on each anomaly for fast bottleneck ressolution..
Anomaly profiles are used to generate anomaly lists starting from job results.
Create a new anomalies list
-
Right-click on a job result folder and select the "Generate Anomalies..." menu in the contextual menu,
or
Select the Generate Anomalies icon 
on the toolbar.
The following dialog is opened:
The top left part of the dialog box allows a job result folder to be selected (a result folder is already selected if you chose "Generate Anomalies" from the result tab contextual menu).
The job characteristics related to the selected result folder appear in the top right:
- Collector Tasks groups
- Scripts task groups
- Number of virtual users
- ...
-
Select an anomaly profile name from the "Anomaly Profile" drop-down list
-
Specify a name for the anomalies list in the "Anomalies List name" field,
-
Click OK
-
A new anomalies list is created, added to the results tree and displayed in the job management window.
Description of an Anomalies list view
The anomalies list view is composed of two parts:
-
The top view which offers two ways to display the list of anomalies (Click the button located at the top right part of the list to change the view):
-
arranged by module: Anomalies are displayed in an arborescent way as follows:
Module (ex: UNIX, Oracle ...)
|_ Collector Name, Anomaly type, Number of anomalies (ex: MyOracleCollector, Buffer cache hit ratio: 3)
|_ Anomaly 1
|_ Anomaly 2
|_ Anomaly 3
-
a raw list: Click on a column header to sort the list.
-
Select an entry in the list; at the bottom of the view, a chart shows the statistics and the time periods on which an anomaly has been raised.
Seven anomalies with different severities can be seen in the following screenshot.
