Windows CE Stress Tool Harness (Windows CE 5.0)
The harness for the Windows CE Stress tool consists of an sServer.exe server application that runs on the development workstation and an sClient.exe application that runs on the target device.
On the development workstation, sServer.exe provides a dialog box that you can use to specify options for the test and view test results. The sServer.exe application sends options that you specify to the sClient.exe application on the target device and receives test results from the sClient.exe application.
The sClient.exe application on the target device runs the test modules, manages the life span of the test modules, and records the results of tests. The sClient.exe application also monitors statistics of the target device, such as memory use. The sClient.exe application reports the state of the target device and sends cumulative test results to the sServer.exe application on the development workstation. For information about viewing test results, see Interpretation of Windows CE Stress Tool Results.
The harness has following capabilities:
The harness manages test modules that perform the actual testing.
The harness runs a specified number of test modules at all times.
The harness tracks memory and object store usage over time, records whether each test module passes or fails, and detects when a test hangs.
The harness records the state of the system in the results for the test.
The harness provides the ability to set the level of logging verbosity for the test modules.
With verbosity levels, you can filter the information displayed in the Output window for Microsoft Platform Builder.
The harness can be configured from the Windows CE Stress dialog box.
For information about the command-line options for the harness, see Options for the Windows CE Stress Tool.
The harness governs the nature and intensity of the stress environment by controlling the following parameters:
The mix of test modules.
By default, the Windows CE Stress tool runs the largest subset of test modules corresponding to the functionality that the run-time image supports. You can modify this module mix to test individual functional units of the OS. You can also create and add your own test modules. For information about creating a test module, see Custom Module Creation for the Windows CE Stress Tool.
The number of concurrently running test modules.
The harness keeps a specified number of test modules running concurrently at all times. When an instance of a test module exits, the instance is replaced by a new instance of a test module that is selected at random from the mix of test modules. When the harness runs a number of test modules concurrently, it can provide information about the interoperability of the functional units of the OS being tested.
The duration over which each test module runs.
The harness can request that test modules run for a specified duration. A short run time tends to provide creation, destruction, connection, disconnection, initialization, and cleanup scenarios. A long run time tends to provide information about the accumulation of resources. By default, the harness runs each test module for 10 minutes, which provides a reasonable mix of both types of stress testing.
Overall duration of the test.
By design, the Windows CE Stress tool introduces varied testing combinations over the course of a testing run. You are more likely to uncover problems with your run-time image or confirm that it is robust when you increase the overall duration of the test. The default duration is 900 minutes.
See Also
Windows CE Stress Tool | Running the Windows CE Stress Tool
Send Feedback on this topic to the authors