thomas_brossama_1
Contributor
Joined 11 years ago
User Profile
User Widgets
Contributions
Run a command prompt line disconnecting a RDP session without stopping test execution (as admin)
Hi, I am struggling finding a way to disconnect automatically an RDP session so the remote test computer does not lock its session and makes failing the test execution (white screens due lock session). I looked closely at the article - Disconnecting From Remote Desktop While Running Automated Tests https://support.smartbear.com/viewarticle/72794/ Through a batch file, it is possible to disconnect so with the following batch file for /f "skip=1 tokens=3" %%s in ('query user %USERNAME%') do ( %windir%\System32\tscon.exe %%s /dest:console ) Manually in a command prompt, as admin, the command works fine. However, When it comes to run it through TC11 I do not manage to use this batch file whether it is a Keyword (cannot get started the command line prompt as administrator, whereas I set the TestedApp to Run as and TestComplete is run as admin) or or a vbscript. Here is my code Sub SafeRDPDisconnection() Dim WScriptObj, ExecOperation, ExCode, StdOut Set WScriptObj = Sys.OleObject("WScript.Shell") ' disconnecting Set ExecOperation = WScriptObj.Exec("cmd /c ""cd C:\my_project & disconnectRDPforTC.bat""") ' Get the operation's exit code ExCode = ExecOperation.ExitCode ' Get the application's StdOut stream Set StdOut = ExecOperation.StdOut If ExCode = 0 Then Call Log.Message("Connection succeeded", StdOut.ReadAll) Else Call Log.Error("Server is unavailable", StdOut.ReadAll) End If end Sub It seems that the batch file is not run as admin. Could you please shed some light? I believe that I am close to the solution. In advance, thank you Regards Thomas2KViews1like2CommentsHow do I access to the Import Test wizard in TC10.6 (disappeared)?
Hi I would like import test instructions into TestComplete as it is explained in http://support.smartbear.com/viewarticle/55555/ However, I cannot perform "To call the wizard, select File | Import | Test from TestComplete’s main menu." The import button is not present? Is it a bug, a wrong configuration from me, or the import feature has been removed? Thank you for your help best regards Thomas Brossamain728Views1like0CommentsHow to prevent Testcomplete automated tests from terminating during execution from QAComplete?
Hi, I am running automated tests from QAComplete towards test machines with TestExecute and TestComplete installed. When running tests from the Test Library or from a Test Set, the execution starts on the test machines, but is interrupted at a certain point in a test, mainly after the second test of a project. The logs indicates that "Possible reasons: - An error or warning occurred during the test run. See the preceding log entries to locate and identify the error. - The test run was stopped manually. - The test run was terminated as a result of the Runner.Halt method call."" Those tests work fine whether they are executed with TestComplete or towards teest machines with TestExecute (both Manual and RDP connection mode) I have checked the content of the project suite from where the tests have been copied on the test machines, and all test items under the project suite do not have the "Stop on error" ticked on. And under each project of the project suite, I set None for Stop on Error and Exception for each Test Item and ticked off under Properties the Stop on Error as well. So could you please explain me the reason it was the execution was terminated? Is it from QAcomplete, from the Test Agent, or from the TestExecute/TestComplete instance? NB: I saw that under Test Set, you can set when creating a Test whether it shall "stop on failure" or not. I did not set it, but it still fails. So I believe that this does not come from QAComplete, but either from the TestAgent or my settings in the project suite. I also find out that in the Help, http://support.smartbear.com/viewarticle/63774/#DistributedTesting How come the project stops executing? Sorry but I don t get it. Thank you for your help Bes tregards Thomas Brossamain1.2KViews0likes0CommentsHow do you set an appropriate delay to each host when doing parallel testing?
Hi, I have several hosts running the same keyword tests for an application (parallel testing) without interacting between each other. Some of the tests write the exact same data at the same time to a database and therefore, the tested application sees a conflict where the same data are in used/edited by another machine. To avoid such conflict, I would like to add a different delay at the beginning of the execution of the tests on each host. Is that possible to configure the Network Suite, or a Job and indicate how much delay is needed to ensure that the same tests for writing are not executed at the same time? I tried to add a random delay in the keyword tests directly, to minimize the probability that the conflict happens. Sometimes the random delays are too close to each other, which triggers the conflict. This is the solution I came up with. Another solution is very welcome as well. Thank you in advance /Thomas759Views0likes1CommentRe: Automated running of TestComplete scripts by QAComplete
Hi Masha, Thank you for your concern, I appreciate. Not sure if you meant about my last reply, but I can still reply to you (althought I do not work with Mathijs) and share my thoughts. In general I am very satisfy of TestComplete/TestExecute. With QAComplete, I am not entirely satisfy since I have not been able to use the automation features fully. Probably due to their limits and time. I feel QAComplete is still more dedicated to manual testing than automated testing. Regarding my project, it is now delivered with a full setup TestComplete, TestExecute, Automated Testing Bridge complete setup and QAComplete configured to retrieve the test results from the test runs (with the old legacy feature) I d like to focus on QAComplete now. The new functionalities regarding test automation (Under Test Library, Automations tab with Test Schedules, Hosts) in QAComplete do not satisfy me. I lke the intention to separate schedules and host, but this is cumbersome to use compare to the old legacy features. Correct me if I am wrong, you need to import "manually" ZIP file of the project suite (pjs/mds files) for each test you want to run according to the Test Schedules you setup. Therefore when any test change in your test suite you need to upload a new ZIP file. How do you manage the changes of your project suite without needing packaging every time your project suite. In the help document (page 163 in PDF file of QAComplete) stands "Pack your TestComplete test projects or project suites to a .zip archive. You can store the TestComplete projects on a shared network folder or website. You can also upload the files directly to your Tests in the Test Library." Does this mean I don t need to zip my project suite? This might be the solution. Also the old legacy features (Under Test management --> Automation) where it was possible to "Start Now" the test suites was useful, because you were able to verify at any time whether the bridge btw QA and TC was working (without waiting for the scheduled time you set up). The new features do not allow this anymore. My wish would be that the Test agent/Automated Bridge Testing could synch between TestComplete project suites located on a shared folder and populate automatically the Test library in QAComplete. Then we could choose the test suite/project suite to run from QAComplete ( and not prepare your test suite in advance in TC) and gets the related test results under Test Management -> automation. The integration btw TC and QAComplete is not finished and new features/updates might come. You probably know. Regarding the displayed test results, the defined numbers of nbr tests/passed/failed are unclear (still under Test Management -> automation). I am not able to understand what they represent after executing the test runs. In TC, the test logs display 1 pass for 1 test case. While in QAComplete, it says 10 passed for 10 tests whereas I ran only 1 test case. I could not find today an answer to a post I sent in another topic (http://smartbear.com/forums/f68/t91862/differences-between-number-of-tests-executed-by/) I wish to be able to use QAComplete more, but I am not working anymore in the project. And my guess is that with time, this project will increase with new test cases. And the management of the test automation with QAComplete might not be easy. The new features introduce in some ways new ways of working with QAComplete that needs to be learned. I am not enough experienced with the QAComplete and the Automation features to conclude whether this is good or not. But I am confident it is getting there. Another discussion would be how to execute a single test case which has parameters from QAComplete instead of specify a whole pre-configured project suite with only one test case included. Voila! This was maybe a long feedback and lessons learned from this project. but I think working with the Smartbear products was very interesting and enriching. I d like to continue in the future. Dont hesitate to contact me if needed ;-)2.5KViews0likes0CommentsRe: Run/Schedule automated tests with QAComplete towards remote machines
Hi again, After discussing with the customer support, simply, for each remote machine install the Automated Testing Bridge executable whether you have TestExecute or TestComplete installed. Therefore when configuring the bridge, the drop down list at the bottom "Run with" display the automation tool you have installed on the machine (TestExecute or TestComplete) Leave some time for QACompelte to register the machine, and then go to Software Planner website, and find the remote host recently registered when create automated scheduled test. Enjoy ;-)771Views0likes0CommentsRe: Differences between number of tests executed by TestExecute and in report qaComplete
Hi, I have a similar issue, where the number of reported run, passed, failed tests do not match with the test logs created by TestComplete. What means a "test" in the logs? A whole test, a step generated automatically by TestComplete in the created test? I have not found any info about this at the moment. Looking forward from hearing from anyone. Thank you for raising the question616Views0likes0CommentsRe: Blank screen on the remote desktop window in Run State monitor Tab (NetworkSuite)
Hi again, I have registered a case to the customer support (00052210) After investigation, I found out that on the Virtual machine (where TestExecute is installed), by uninstalling the Automated Testing Bridge SW, the screen monitor in TestComplete comes back to normal, no more white screen. However, if I want to use QAComplete, I need to reinstall the SW. Any alternative? My first conclusion is that the presence of Automated Testing Bridge SW on a VM host disturbed the visualization of this host via Testcomplete while executing distributing tests. I will report it to the customer support. Do you know if a newer version of the Automated Testing Bridge SW has been released, which could solve this issue? I am using the version 9.8.6. If you have any information please let me know, thank you1.1KViews0likes0CommentsHow to Import automated test cases from TestComplete to Test Libraries in QAComplete
Hi, Today I am running only automated tests via TestComplete and have managed to execute them over time via QAComplete. However, the Test management needs to be done via TestComplete first before launching them by QAComplete. This is a bit cumbersome to be honest. QAcomplete only detects the slave projects found through the Automated Testing bridge SW installed between QAComplete and TestComplete. So before running any planned tests, you need to select the tests of each slave project to execute in TestComplete. A slave project in Testcomplete contains a set of tests enabled by the user. QAComplete cannot those single tests, or it is not possible to select the needed tests among a slave project. Am I wrong? Is that actually possible today? Maybe I am not aware of this. As I understood, the Test Management in QAComplete allows building a test library of manual tests. But can it be possible to do so for automated tests? Meaning, is that possible to import the tests detected via the Automated Testing bridge SW from TestComplete into the Test Library in QAComplete? Therefore, It would be very nice to control Testcomplete from QAComplete (select the slave project to run, test items per project and host on which to run). Is that a feature to be released later? Thank you in advance for the info1.6KViews0likes0CommentsRe: Blank screen on the remote desktop window in Run State monitor Tab (NetworkSuite)
I have tried the command given by http://homeservershow.com/remote-desktop-session-locks-workstation-after-exit.html %windir%\System32\tscon.exe RDP-Tcp#0 /dest:console But it did not work. What is frustrating, is that the verification of the host and the jobs work fine. When it comes to run, the test steps are executed but some fail. The same tests are run on another machine without issues Other comment: I noticed that when I run distributed tests via an automatic RDP session, and I connect during the execution to the remote host (VM) under test, the window in Run state tab becomes white. I did this operation because I wanted to unlock a test that was blocked to due the appearance of an unexpected window. So probably, the white screen might be caused due to the fact I opened a manual RDP session while an automatic RDP session is already opened. And both RDP session are opened with the same user.1.1KViews0likes0Comments