.. This file is source for HDB++ diagnostics tool documentation 


Diagnostics
==============

* This is tool is used to control the behaviour of your HDB++ system.
* It controls your 'HdbConfiguratorManager' and 'HdbEventSubscriber' devices.
* It can be launched by **View / Diagnostics** menu in configurator GUI.
* Or directly by the command: **java org.tango.hdb_configurator.diagnostics.HdbDiagnostics**

Main window
------------
    * The main window display a table where:
        - rows are archivers (last row is the manager).
        - columns are the number of attribute faulty, started, paused, stopped and statistics.
           And finaly the archiver context.
      
      .. image:: images/Diagnostics.png
      
    * By a double click on a cell (or using right click menu) you can have a detailed display.
    
    
    

    
Faulty attributes
------------------
    * If you click on a **Faulty** cell, you can see:
        - The list of faulty attributes
        - The error message
    * By a right click on **Selection** Column, you can stop or remove selected attributes.
    
    .. image:: images/FaultyAttributes.png
    



    
Attribute statistics
---------------------
    * Some statistics are computed (Events and Failures by period) by the **HdbEventSubscriber** device.
    * For instance if the property **StatisticsTimeWindow** is 1, you have the number of events and failures by seconds.
    * If you click on a statistics cell you will have statistics by attribute since last reset.
    
    .. image:: images/AttributeStatistics.png


    
    
Change context
---------------
    * If you double click on l;ast column (Context), you can change the specified **HdbEventSubscriber** context.
    * The better way is to do it on **HdbConfiguratorManager** (last row) to change context on all archivers.
    

    
    
Distribution
-------------
Distribution window is available using **View / Distribution** menu.

    * The first tab gives a chart displaying the distribution:
        - Number of faulty attributes by archiver.
        - Number of *OK* attributes by archiver.
        - Number of events received by archiver since last reset.
    
    .. image:: images/Distribution-1.png


|
    * The second tab gives a chart displaying the maximum:        
         - Process time (from event reception to storage)         
         - Store time (time to store value)                       
         - Pending (number of attributes waiting to be archived)  

    .. image:: images/Distribution-2.png
    

|
    * The third tab gives the same information as numeric values in a table.
    
    .. image:: images/Distribution-3.png