Show TOC

Function documentationAlerting Framework Locate this document in the navigation structure

 

You monitor the status of the alerting infrastructure. In detail, this is the following components:

  • Alert Consumer Connector (ACC)

    The Alert Consumer Connector is the interface for all infrastructures that use alert information, such as the Alert Inbox, mail, or third-party applications. Depending on the relevant configuration settings, the alert information is transferred using a push mechanism or a pull mechanism.

  • Data Provider Connector (DPC)

    The Data Provider Connector is responsible for transferring monitoring data to the end-to-end Monitoring and Alerting Infrastructure (MAI). The required information can be transferred using a push mechanism or a pull mechanism; various data sources are supported, such as the local CCMS, NetWeaver Management agents, or CA Wily IntroScope.

  • Event Calculation Engine (ECE)

    The Event Calculation Engine uses the reported metric values to determine the status, events, and alerts of the administered objects on the basis of configuration settings in the Alerting Directory.

Features

  • On the Status tab, you can:

    • In the Data Provider Connector tab, you can display the extractors used for alerting. You can display error messages and activate, release or restart extractors. For further information, see Identify Extractor Errors.

    • On the Alert Calculation Engine tab page, you can check the results of traces that you start, if, for example, production metrics were collected, but no alerts were raised, or the expected status is not returned. For more information, see Display Alert Calculation Engine Traces.

  • On the Configuration tab page, you can set the value of two configuration parameters that monitor the HTTP resources and, if necessary, restrict the number of work processes available for this. In this way, you ensure that the processing of HTTP requests works correctly, without causing bottlenecks when processing other requests. This is especially important with regard to receiving data from diagnostics agents. For more information, see Activating and Deactivating a Restriction of Parallel HTTP Requests.

  • On the Performance tab page, you can display the values over time of the most important performance metrics for the Data Provider Connector, the Event Calculation Engine, and the Alert Consumer Calculator for a specified period. You can, for example, determine whether runtimes are longer at certain times of day, and analyze the causes.

    You can view the following values for the following metric types by choosing the Display button.

    Data Provider Connector Metrics

    Description

    Number of metrics provided by DPC Pull RFC per technical system

    This metric specifies the number of metrics that were generated using the DPC pull collection process for a particular combination of technical system, client, and data provider source.

    The data provider sources are typcically software components.

    Average pull RFC reaction time per technical system in ms

    This metric specifies the response time of the DPC pull collection process for a particular combination of technical system, client, and data provider source.

    DPC pull average runtime per technical system in ms

    This metric specifies the runtime of the DPC pull collection process for a particular combination of technical system, client, and data provider source.

    Number of metrics provided by DPC push per agent

    This metric specifies how many metrics were generated using the DPC push collection process for a particular diagnostic agent.

    Average DPC push runtime per Web service call in ms

    This metric specifies the runtime of the DPC push mechanism for a particular diagnostic agent.

    Event Calculation Engine Metrics

    Description

    ACE calculation operations per second

    This metric specifies the throughput of the Event Calculation Engine in calcuation operations per second.

    Size of Metric and Event Store

    This metric specifies the number of entries in the Metric and Event Store.

    Total ACC Runtime

    This metric specifies the runtime of the Event Calculation Engine.

    Alert Consumer Connector Metrics

    Description

    ACC housekeeping runtime

    This metric specifies the runtime of the administration job of the Alert Consumer Connector. This job deletes alert groups and the associated data after the corresponding period has ended.

    ACC housekeeping throughput

    This metric relates to the same issue as the ACC housekeeping runtime described above. However, instead of the runtime, the number of processed objects per second is specified.

    ACC incident runtime

    This metric specifies the runtime of the incident management process of the ACC. Depending on the configuration settings, when alerts occur, this process generates corresponding incidents and calls the responsible components for additional processing of these incidents.

    ACC incident throughput

    This metric relates to the same issue as the ACC incident runtime described above. However, instead of the runtime, the number of processed objects per second is specified.

    ACC notification runtime

    This metric specifies the runtime of the notification process of the ACC. Depending on the configuration settings, when alerts occur, this process generates corresponding notifications and calls the responsible components for additional processing of these incidents.

    ACC notification throughput

    This metric relates to the same issue as the ACC notification runtime described above. However, instead of the runtime, the number of processed objects per second is specified.

    ACC reaction runtime

    This metric specifies the runtime of the reaction management process of the ACC. This process forwards the required alert data to the relevant auto-reaction methods and the registered third-parties.

    ACC reaction throughput

    This metric relates to the same issue as the ACC reaction runtime described above. However, instead of the runtime, the number of processed objects per second is specified.

  • In the Exceptions tab, you can access a graphical interface to the application log (transaction SLG1). A problem overview is displayed. To show error messages, select the problem.