Monitoring a BusinessEvents Cluster : Overview of BusinessEvents Monitoring and Management

Overview of BusinessEvents Monitoring and Management
BusinessEvents Monitoring and Management features are available when Cache Manager (cache-based OM) is used. They are not available for in-memory or persistence OM types.
All BusinessEvents cluster members — agents, JVMs (processing units) and machines — generate system metrics. They are made available over the network for use by the Monitoring and Management (MM) web-based user interface.
After the Monitoring and Management component is configured (as explained in Chapter 6, BusinessEvents Monitoring and Management Configuration), you start the MM engine (which is a BusinessEvents application). You can then connect to the MM server using MM Console, a browser-based UI. Using MM Console, you can examine all aspects of the cluster, including objects in the cache.
At each level of the cluster hierarchy, various panes display metrics relating to that level graphically. You can rearrange and remove panes, and promote panes of special interest to the cluster overview to create a custom dashboard.
Reference to the Metrics
For each level of the cluster hierarchy, various metrics are provided using charts and tables. Later sections in this chapter provide a reference to the metrics available at each level.
The charts and tables appear in different panes on the panel or panels (tabs) relating to one level of the cluster hierarchy.
The agent level can have two panels. One shows the same information for all types of agents. For inference and query agents, a second panel shows information specific to the agent type. Cache agents do not have a second panel.
Note the following:
Charting Period  All charts display data for the previous ten minutes (or less if, for example MM or a JVM has been running less than ten minutes).
Updates  Monitoring and Management polls for updates and refreshes the display periodically. The interval is configurable. See the updateInterval property in Configuring MM Console Properties.
Health Indicators and Alerts
You can configure health indictor thresholds that define whether a value indicates normal functioning, a potential problem (warning), or a critical situation.
You can also configure alerts to bring specific problem situations to the attention of system users.
These health indicators and alerts are shown in the Cluster Overview, providing a dashboard where you can read the health of the entire cluster at a glance.
See Chapter 6, BusinessEvents Monitoring and Management Configuration for configuration details