Investigating Agent Manager diagnostics
You monitor the state of your Agent Manager instances and the related Management Server adapters using the FglAM Performance Overview dashboard. Use this dashboard to better find out how these components perform over time and to look for any indicators that may predict potential bottlenecks. 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, choose Dashboards > Management Server > Diagnostic > Agent Manager.
The information appearing on this dashboard appears on two major tabs, Adapter Details and Client Details, each consisting of several sub-tabs. For more information about the data appearing on this tab, see the following topics:
Exploring the Adapter Details tab
The Adapter Details tab allows you to investigate the performance of the Management Server adapters that communicate with the connected hosts. Use it to see the overall processing rates and to estimate your system’s load. Higher processing rates may lead to decreased performance, requiring further investigation.
For more information about the data appearing on this tab, see the following topics:
Message Processing tab
The Message Processing tab contains graphs that tell you how well the adapter is handling incoming messages. It displays the numbers of connected hosts and agents, the number of incoming queue threads, and the numbers of pushed and pending messages over time. It also shows the minimum and maximum polling intervals, the rate of incoming messages, and the percentage of time the adapter spends on message processing.
XML Serialization tab
The XML Serialization tab shows graphs indicating the rates and times of message serialization to and from XML used by the adapter over the selected time range. High peaks in the graphs can indicate signs of performance decrease and may need to be investigated.
