|
|
|
|
You can use the TestDriver to run the test packages remotely over a
TCP/IP network, this is to minimise the use of development boards (Lubbock or
H2/H4). You need to install the TestDriver on the local host, and configure
using 'testdriver config' command. The client and the server
reside inside the TestDriver and are installed along with the TestDriver.
Wintap can interfere with remote connections, therefore you need set the following to ensure that the primary TCP connection is listed first:
Go to Start>Settings>Network and Dial-up connection
Select Advanced>Advanced Setting...
The following diagram shows the complete setup required to run tests remotely. The client and server are shown seperately in the figure which are integral part of the TestDriver. The Server program (TestMaster) accepts the TestDriver jobs and executes the test jobs on a first in first out basis.
Some of the terms used in the test setup are explained below:
Please note that TestClient and TestMaster are now integrated in TestDriver. So all what the user needs is to install TestDriver on the local and remote host and configure both sides using testdriver config. please replace lubbock by H2, H4 and whatevere boards we are supporting.
TestMaster - The TestMaster is a software program that needs to run on the Server in order for the remote test execution to work. The TestMaster receives test jobs sent by the TestClient, and then passes them on to the TestDriver, which executes the tests on the H2/H4 development board.
H2/H4 development board - This is a development board on which tests are executed using TestDriver. The board needs to be connected to the Server via a null modem cable and a power supply.
TestDriver - The TestDriver runs both the Client and the Server in order to create test packages which form part of the test jobs to the TestMaster, and then executes the tests on the H2/H4 development board. For more information on using TestDriver and creating test packages, see Using TestDriver and Building test packages.
When several test jobs are to be processed, they are placed in a queue by the TestMaster, and progress information is sent either directly (synchronous mode) or can be retrieved when needed (asynchronous mode).
Test job - The test job is created on the Client part of the TestDriver and contains the test package (created by TestDriver) and the ROM image file to be used when executing tests on the development board.
Test package - The test package contains tests to be executed by the TestDriver on the development board.
Test results - The test results are stored on the Server as
testresult.zip file, and can be retrieved by the Client.
When using TestDriver to run tests remotely over a network (WAN or LAN) in asynchronous mode, the workflow processes on the Server and Client sides are as follows:
Install TestMaster which is installed along with the TestDriver on the PC. For details, see Installing the TestDriver.
Connect a Lubbock or H2/H4 development board to the Server. The tests will be run on this development board.
Check for the status 'TestMaster running' of the TestMaster on the console.
The following is the general process to follow to execute tests remotely in asynchronous mode.
Install the TestDriver on the PC that acts as the Client as outlined in Installing the TestDriver.
Create the test package containing tests to be executed on the development board, using TestDriver. For more information on creating test packages, see Building test packages.
Create the test job, by placing the test package and the ROM
image to be used for tests in a directory, for example testjob1.
Submit the test job using the runremote command with
the relevant options (see How to run test packages remotely).
The test job is then sent to the TestMaster for execution. If several test jobs are to be executed on the server, they will be held in a queue.
When the test job is ready to run, it will be sent to the TestDriver, which executes the tests (defined in the test package) on the Lubbock development board, using the ROM image supplied with the test job.
If you are running a test job in synchronous mode, progress information on the status of the test job will be displayed automatically.
If you are running a test job in asynchronous mode, you can check
the status of the test job any time, using the command testdriver
jobstatus, as described in How to run the test packages remotely.
Define the directory location for collecting the test results on the client side.
Retrieve the test results from the Server to the Client using
testdriver results command. The results will be stored in a
testresult.zip file, within the defined location.
View the results by opening the HTML report contained in the
testresult.zip file.
For more information on running the test jobs, see How to run the test packages remotely.