Getting familiar with the Monitoring view



Monitoring aspect overview:

The monitoring view part of the Petals Webconsole provides a

technical supervision of the exchanges transited in Petals ESB.

This aspect is implemented by the monitoring tab of the Petals

Webconsole, through it you have access to the following functions:

  • Monitoring filter creation.
  • Monitoring filter management.
  • Detailed vision of exchange using charts and diagrams.
  • Detailed overview on each stored exchange.


Filter management tab

Filter creation form

This form allows you to create different monitoring filter by playing with the

different form parameters.

The form provides several fields which may allow you to create different

monitoring filter configuration:

  • Filter on the endpoint name.
  • Filter on the interface name.
  • Filter on the service name.
  • Filter on the operation name.

  

Each previous parameter may not be used, in this case null value is assigned to the parameter

For example if you define a filter with null/null/null/null value for endpoint/service/interface/operation

parameters, upon activation of this filter all exchanges will be monitored by this filter.

Warning: pay very attention with this king of filter, because monitoring all message exchanges

can seriously affect the speed, performance and overload on the Petals node where the filter is

activated.

Another important form parameter is the storage checkbox which allows you to declare a filter to

archive each message exchange caught.

Filter management table

This useful table allows you as this name mean to manage each declared filter. Each of them is

represented by a table row, which display several informations on the current filter:

Displayed informations are partly the same as those that you served in the monitoring creation filter form:

  • Endpoint name.
  • Interface name.
  • Service name.
  • Operation name.
  • Stored exchange value

Other informations displayed is the current server where is declared the monitoring filter, and the monitoring

state of the filter enable/disabled, alterable by the following icons:

You can click on each icons to switch the current filter state, for example:

If a filter is disabled click on icon to active current filter on the associated

server, after that the monitoring enabled state icon will be displayed .

Finally, the following icon: allows you to delete the current filter, if is enabled

on the associated server it's firstly disabled and then delete, instead if is

disabled it is directly delete from the monitoring filter table.

  

Two buttons are available on the top of filter table, the first button:

allows you to export all filters of the list into CSV file. Example is displayed bellow:

EndpointInterfaceOperationServiceStorageServerStateChange stateDelete
edpt-945EBEF356D7737C3E228882447520291{http://petals.ow2.org}Clock{http://petals.ow2.org}printTime{http://petals.ow2.org}ClockServicetrue0EnabledEnabledDelete
endpointPojo{http://petals.ow2.org/}Pojonull{http://petals.ow2.org/}servicePojofalse0DisabledDisabledDelete

The second button: allows you to import a previous

  

exported filter list. All filter will be in state with which they were imported.

  

Several examples:

  • If filter-1 already exsist and is disabled on server but, it is enabled in csv file,
    it will be enabled at the end of csv importation file.
     
  • If filter-2 already exsist and is enabled on server but, it is disabled in csv file,
    it will be disabled at the end of csv importation file.
     
  • If filter-3 don't exsist on current server and it is enabled in csv file,
    it will be enabled at the end of csv importation file.
     
  • If filter-4 don't exsist on current server and it is disabled in csv file,
    it will be disabled at the end of csv importation file.
Container owner for monitoring filter

Take attention, during an importation process only monitoring filters declared for the current container will be imported.

For example if you try to import on container "0" a monitoring filter declared for container "1", the importation process don't import it.

The container owner is the sixth column parameter, the title of column is "Server".

Monitored filters tab

Monitored exchange filter

In this tab an table displays all message exchange caught by each declared filter on the current server.

Each row represent a declared filter and displays its linked informations like:

  • Endpoint name.
  • Interface name.
  • Service name.
  • Operation name.
  • Storage value.

  


Information: Each row own the following icon that allows you display more informations, on the current monitoring
filter.

  

Many different informations are displayed like the number of monitored exchange for the filter, message exchanges

caught by the current filter.

  

The following form allows to set starting/ending dates, in order to define a time range where

the displayed exchanges will match the defined range.

  

The table which display message exchanges consists of three columns. The first

shows the message exchange id beginning with petals:uid:, the second column

correspond to the message exchange pattern and finally the last column contain

the following icon , which will server to show you the detailed view of each

message exchange (a king of overview).

  

Idea: another useful point in each monitoring filter row is the following icon , which allows

to display charts and diagrams representing exchange flows, exchange status, exchange

number ... of the current filter.

Charts & diagrams view

When you click on the icon a page linked to the current filter is displayed, in the top

of the page a form allows you to modify the window duration and the total duration of the

current monitoring view (int value representing second value).

  

The first displayed chart present the different status of message exchanges caught by the

current filter:

Information: You can spend on each chart part to know the percentage of the associated status,

this feature can be useful to report to you if the service correctly processes message exchanges.

  

The diagram located on the bottom of the chart monitoring view page, present the exchanges flows

by an number exchange average for a specified time scale:

Four curves are displayed thank to this diagram:

  • A blue curve that represent message exchanges with done status.
  • A green curve that represent message exchanges with active status.
  • A yellow curve that represent message exchanges in error status.
  • A red curve that represent all message exchanges without taking into
    account the different status.

  

The X,Y axis values are automatically adapted to the parameter values of the

monitoring view form.

  

Information: This diagram can be considered as a sliding horizon, that mean your diagram

move to the left at each view update.

Exchanges overview page

This page is the most useful part of the monitoring, because it permits you to display all

informations of the current exchange that was caught by the monitoring filter. This large

page is composed of tree different parts, the first on top represent all global exchange

informations:

  • The message exchange id beginning by petals:uid.
  • The message exchange pattern.
  • The exchange status.
  • The endpoint name.
  • The interface name.
  • The service name.
  • The operation name.
  • The total exchange duration in millisecond of the exchange.
  • The provider message processing duration in millisecond.
  • The consumer message processing duration in millisecond.
  • The beginning date of the exchange.

  

Also displayed in this part is the properties table of the current exchange with

two columns, in the first hand one for the property key and in other hand one

for property value:

  

The second page part is the most representative for the current exchange, which

consists to model the exchange sketch:

This draft displays the consumer component and the provider component, you can also know

the container holder just by passing you mouse over each component. If an unknown component

plays a role in the exchange, then an image of generic component will be displayed.

After that, several message types can be drawn according to the message exchange pattern:

Message
Meaning
This message type represent the IN message
sent to begin the message exchange.
Message representing the OUT message
sent during the message exchange.
DONE message type that represent the
correctly finishing of the exchange.
This allow to model a FAULT message
transited during the exchange.
That allows to display if an ERROR occurred
during the exchange.

  

The last overview part is located on the page bottom, this part displays each message

appeared during the exchange. Each displaying message have its own part which present

its associated informations:

Enter labels to add to this page:
Please wait 
Looking for a label? Just start typing.