eG Monitoring
 

Measures reported by SiebelAppMgrStsTest

Application Object Managers (AOMs) host the Business Objects layer and Data Objects layer of the Siebel architecture. The Web clients host the Siebel application user interface layer. The AOM is used primarily to support Siebel Web client connections. To do this, the AOM handles multiple users simultaneously by making requests to the Siebel Server on their behalf.

AOMs are hosted as components in the Siebel Server and run on the application server (the machine that hosts the Siebel Server). The Siebel Server provides the infrastructure for an AOM to serve multiple Siebel Web client users. Multiple AOM components can run on a single Siebel Server installation. AOM components can be configured to run as multithreaded processes in the Siebel Server. Like other Siebel Server components, you can administer AOM components using the Siebel Server Manager.

AOMs communicate with clients using the TCP/IP protocol through a Web server that contains the Siebel Web Server Extension plug-in (SWSE). Communication between the Web server and the AOM can be compressed and encrypted. An independent session is established to serve incoming connect requests from each client. Subsequent requests from clients are directed to the same AOM tasks until the sessions are terminated. After startup, AOMs do not achieve their full run-time environments until after the first connect, therefore, leading to possible delays during the first connection.

One of the most important types of server components is the Application Object Manager (AOM). These server components always run in interactive mode. They process user requests and are application- or service-specific. For example, the Siebel Call Center component group contains the Call Center Object Manager, one for each language deployed on the Siebel Server. This AOM provides the session environment in which this application runs.

Internally, each AOM also contains a data manager and the Siebel Web Engine. When an AOM receives a user request to start an application, the AOM follows this procedure:

  • The business object layer starts an application user session, processes any required business logic, and sends a data request to the data manager.

  • The data manager creates an SQL query and forwards it to the Siebel Database.

  • The data manager receives the data from the database and forwards it to the business object layer for additional processing.

  • The business object layer forwards the result to the Siebel Web Engine, which helps create the UI for the data. The Siebel Web Engine then forwards the Web pages to the Siebel Web Server Extension on the Web server.

Whenever users start complaining of delays encountered in receiving response to requests, it is the onus of the administrators to figure out the exact cause of the delays in request processing – is the delay due to an unresponsive database/Application Object Manager or a lengthy request or due to errors encountered by the Application Object Manager? To ensure optimal request processing, such issues should be rapidly detected and remedial steps should be taken accordingly. This is where the SiebelAppMgrStsTest test helps!

This test helps you to figure out the errors encountered by the Application Object Manager, the size of the request and reply messages and the requests received per session. In addition, this test provides insight on the response time of the database and also the rate at which the request and reply data were processed in the Application Object Manager.

The measures made by this test are as follows:

Measurement Description Measurement Unit Interpretation
Object_manager_errors Indicates the number of errors encountered by this Object Manager. Number Ideally, the value of this measure should be zero.
Request_messages Indicates the number of request messages received by this Object Manager per second. Requests/Sec  
Reply_messages Indicates the number of reply messages sent from this Object Manager per second. Responses/Sec  
Average_connect_time Indicates the average time taken to connect to this Object Manager session. Secs A low value is desired for this measure. A high value indicates connection bottlenecks.
Average_Request_size Indicates the average size of the request messages received by this Object Manager. KB  
Average_Reply_size Indicates the average size of the reply messages sent from this Object Manager. KB  
Average_Requests Indicates the number of requests received by this Object Manager per session. Requests/Session  
Average_Response_Time Indicates the average time taken by this Object Manager to respond to requests. Secs A low value is desired for this measure. A very high value indicates that the component responds slowly to requests. Response time issues can be caused by high CPU utilization or heavy load on the components.
Average_think_time Indicates the average end user think time between requests received by this Object Manager. Secs  
Total_database_response Indicates the time taken by the database to process and respond to the requests sent through this Object Manager. Secs A sudden/gradual increase in the value of this measure is a cause of concern as this may indicate network connectivity issues, performance bottleneck of the server etc.
Total_request_size Indicates the amount of request data processed per second in this Object Manager. KB/Sec  
Total_reply_size Indicates the rate at which the reply data of this Object Manager is processed. KB/Sec