To troubleshoot requests that suffer errors or high latency during stress testing, you can analyze their traces to identify the root causes.
Prerequisites
A stress testing scenario is created. For more information, see Create a stress testing scenario.
Application Real-Time Monitoring Service (ARMS) is activated. For more information, see Activate ARMS. An ARMS agent is installed, or Managed Service for OpenTelemetry is connected. For more information, see Agent management.
Tracing analysis is configured for the stress testing scenario. For more information, see Configure end-to-end monitoring.
Analyze the root cause of error requests
On the page for an ongoing stress test or on the Report Details page, click the Call Trace Analysis tab.
In the Filter: section on the left side of the page, set the Status filter condition to Error to search for the traces of all error requests.
In the List section, click Details in the Actions column of a trace. In the panel that appears, view the trace details and the time consumed at each stage in the trace.
Click a span to view the detailed method stack of the application.
You can also view the proportion of time consumed by each method and the exception information on the Analysis Overview tab.
Analyze the root cause of slow requests
On the page for an ongoing stress test or on the Report Details page, click the Call Trace Analysis tab.
Search for the slow requests by specifying the duration filter condition. Set the unit to ms.
In the List section, click Details in the Actions column of a trace. In the panel that appears, view the trace details and the time consumed at each stage in the trace.
Click a span to view the detailed method stack of the application.
You can also view the proportion of time consumed by each method and the exception information on the Analysis Overview tab.