Foglight for Databases 5.9.5 - Administration and Configuration Guide

Manage Foglight Database Performance

In Foglight, retention policies define time periods during which the collected data is sampled, persisted into the database, aggregated, or purged from your system. The Object Cleanup dashboard is useful for inspecting, purging, and deleting data objects, and particularly for cleaning up objects that are no longer needed. For instance, if a set of agents is no longer required, the objects created by those agents are still visible in the browser interface. Removing these objects can have a positive impact on performance.

NOTE: In previous releases, the Data Management dashboard allowed you to manage both retention policies and object cleanup. Beginning with Foglight 5.9.x, there are now two separate dashboards: • Retention Policies • Object Cleanup

Using these dashboards you can manually tune the size and performance of your environment. Tuning performance manually allows you to:

To access this dashboard, from the navigation panel, click Dashboards > Management Server > Servers. Then select either Object Cleanup or Retention Policies.

Another way to control the retention policies on a per-object basis is using the Manage Retention Policies dashboard. For more information, see Manage Data Retention.

In addition to the Object Cleanup and Retention Policies dashboards that allow you to inspect and tune the overall performance, the Database Overview dashboard summarizes the database activities such as data row operations, database buffer pool, and any inserts, deletes, and updates. To access this dashboard, from the navigation panel, click Dashboards > Management Server > Servers > Database Overview.

For more information, see the following topics:

Monitor Server Performance

Foglight manages host data sent from agents, and evaluates rule conditions and metric calculations. It also provides browser interface access to remotely monitored servers. The browser interface includes a set of dashboards that allow you to monitor the server state and prevent potential bottlenecks.

The Performance dashboard contains at-a-glance view of Foglight diagnostics. It shows the rate of database inserts, data processing activity, JVM memory usage, server load, and other combinations of views. Certain types of metric patterns displayed on this dashboard can be useful in troubleshooting specific performance problems. For example, a sudden increase in free memory utilization is a good indicator that the amount of incoming data exceeds typical thresholds. To access this dashboard, from the navigation panel, click Dashboards > Management Server > Diagnostic > Performance.

The Management Server View dashboard is useful for examining server performance. Use it to look for root causes of server-related performance problems. To access this dashboard, from the navigation panel, click Dashboards > Management Server > Servers > Management Server View.

Server log files are another source of information that can help you diagnose the root cause of performance-related bottlenecks. They contain information about known events and error conditions as well as verbose or informational messages. The Log Analyzer dashboard allows you to analyze generated log files or download a selected log file to a desired location. To access this dashboard, from the navigation panel, click Dashboards > Management Server > Diagnostic > Log Analyzer.

For more information, see the following topics:

Monitor Agent Manager Performance

Foglight uses the Agent Manager to communicate with monitored hosts. The embedded Agent Manager can be used to monitor the host on which the Management Server is installed. Your monitoring environment typically includes a number of Agent Manager instances that are installed on different hosts. You can monitor their state using the Performance Overview dashboard. For example, an unusually high number of pending messages in the queue indicates a potential performance bottleneck. To access this dashboard, from the navigation panel, click Dashboards > Management Server > Diagnostic > Agent Manager.

For more information, see the following topics:

Associate Service Objects with Groups and Tiers

Foglight monitors specific parts of your environment based on the concept of services. A service is a collection of monitored objects. An object group is a mechanism that assists in service creation and monitoring. It is a logical way of grouping objects that are of interest to an individual user (for example, an Oracle database administrator), or to multiple users of a system (for example, Oracle databases).

Object groups can be associated with another logical component, tiers. Each tier is a logical representation of a service component. A Foglight service can have one or more tiers. By default, Foglight organizes data into default tiers, including User, Web, Application, Database, Host, and Agent. Tiers allow you to structure services in a way that best represents your monitored environment. This type of logical structure helps you isolate performance problems associated with a specific service tier. For example, to investigate the state of the monitored hosts within the Host tier for a service, drill down on the Host tier and investigate the hosts that are related to it. For more information about services, see the Foglight User Guide.

The object groups that are needed for most service monitoring already ship with Foglight. You can create additional object groups and associate groups with tiers using the Object Groups and Tier Definitions dashboards.

To access the Object Groups dashboard, from the navigation panel, click Dashboards > Services > Object Groups.

To access the Tier Definition dashboard, from the navigation panel, click Dashboards > Services > Tier Definition.

For more information, see the following topics: