IBM Tivoli Composite Application Manager for Application Diagnostics, Version 7.1.0.1
Request throughput spikes when Managing Server is restarted
The problem: When many heavily loaded data collectors are communicating with the Managing Server and then the Managing Server is restarted, the Enterprise Overview shows a spike in the request throughput.
The reason: This may happen if many heavily loaded data collectors are communicating with the Managing Server. A data collector automatically detects the status of Publishing Server and drops events when the Publishing Server is disconnected. But some events may remain in the queue. So when the Managing Server restarts, these events may show up as a spike in throughput.
Parent topic:
Run the Managing Server