My test case is: Opening an windows application and closing the same.
I am able to trigger automation in azure pipeline successfully. The test execution is happening successfully. But I am not able to see the UI of execution.
Right now its kind of headless execution.
I want to see where the actual execution happens.
Where can I see the execution(want to see the UI)?
Not sure which kind of UI test you were running. Azure DevOps support both Headless mode or visible UI mode.
Headless mode. In this mode, the browser runs as normal but without any UI components being visible. While this mode is obviously
not useful for browsing the web, it is useful for running automated
tests in an unattended manner in a CI/CD pipeline. Chrome and Firefox
browsers can be run in headless mode.
Visible UI mode. In this mode, the browser runs normally and the UI components are visible. When running tests in this mode on
Windows, special configuration of the agents is required.
More details please take a look at official tutorial here-- UI testing considerations
Related
Debugging an ASP.NET MVC application on the Azure Service Fabric platform takes a long time. How do I speed it up?
If you are talking about cloud cluster debugging then that is going to be a bit slow process if you are attaching debugger to that cluster, in which case you should be relying more on your application logs to find a problem because you should try to replicate the problem in your local cluster first and debugging that locally will be much easier.
Now if you are trying to debug on local cluster then couple of things are important to speed up things.
Set your cluster to run in 1 node configuration rather than 5 node
configuration.
Set your "Application Debug Mode" to "Refresh Application". To do
this, right click on the fabric project in your solution and click
on properties. you will see a window where you can set project
properties, here you will see this option of setting "Application
Debug Mode".
once you have done this then you can hit F5 and the debugging will be a lot faster. Also remember that you don't need to publish your application every time you make code changes, all you need is build and cluster will pick up changes and you can debug any time. one last thing to remember is that if you make any changes to your settings.xml or manifest.xml files you will need to publish in order for these settings to be picked up by the application.
Hope that helps.
I am using BluePrism v6.1.0 and I am trying to identify elements from a launched application in Application Modeller. (I used the Windows application type in the modeler configuration)
From some tutorials I saw, the launch button changes to Identify button after the app is launched, however, mine is still showing "Launch". Please see screenshot below
screenshot
You need to launch the application from the application modeller, not on your own.
Since you are using Windows 10, the built in applications are use differently than normal in windows 7. Here is how you can add the code to make it work with your system Screenshot. Also for future applications use this code in Windows Power Shell to get information about the application you wanted to work with in Window 10 : "Get-AppxPackage"
Not fully explained here but this can also mean not only is the app not launched but blueprism can't "see" the app is launched. Two ways to solve this so close the application and relaunch it then it will change to identify, OR you can attach the current instance of the object to the running application.
Either way the outcome is you're making blueprism see the active application so it will change the launch to an identify option.
Edit: watch out as well, if you connect your application to blue prism and then detach it blue prism will no longer see the running application hence will revert the identify option into a launch option.
I am an automation engineer where I need to run 450 CodedUI scripts on multiple machines. I have 15 machines on which I run these scripts.
To resolve my trouble I am using Microsoft's tool Remote Desktop Connection Manager to login to these machines. But I am getting the error on failed scripts that "Either the window is locked or minimized", but when I used to directly login to these machines and run the scripts there were no such issues.
I am unable to find any resolution. I tried one more tool to connect to 15 machines , i.e. AppVision tool as well. Even with that tool I am facing the errors on all my scripts that Some control is blocking the control to be clicked in.
I need to know if I can have any other tool or way where I would be able to login to the machines in one go and run automation scripts without any errors.
Any help is appreciated.
Thanks in adavance.
Coded UI requires that the screen saver is disabled on the remote machines.
Coded UI interacts with the desktop of the machine running the tests. When the screen saver is active it controls the desktop and, effectively, prevents Coded UI from interacting with the application under test.
The question refers to "Microsoft's tool Remote Desktop Connection Manager" so perhaps you are not using test agent software to run the test. Check this Microsoft web page and this Microsoft forum question for more details on how to set up remote computers to run Coded UI.
I am using Hudson CI to run an ant script which compiles and executes Selenium2/Webdriver tests (written as TestNG tests). These tests are set to run in a Firefox browser, not with headless HTMLUnit. During the Hudson build, I don't see any errors in the console output or the TestNG report, and my custom report seems to indicate that my tests do run normally, but I noticed that no actual Firefox browser is ever launched during the build.
In contrast, whenever I run my Selenium2/WebDriver tests via Eclipse or the command line either as a TestNG test or an Ant build, the tests always launch a browser on which I can see the tests being run. I just can't figure out why the browser does not appear. All of the aforementioned details suggest that the tests are in fact run properly, but the Hudson server is run on my local machine and no additional resources are apparently used to run the tests in their own Firefox instance.
I am not using Selenium Grid at the moment. Also, regardless if I run the tests sequentially or in parallel across multiple threads through Hudson, I am still not able to see any active browsers. I do vaguely remember at one point seeing Hudson launch an actual browser yesterday when I began this work, but I have yet to see once since. This makes me wonder if I possibly configured something incorrectly in my Hudson job.
Any help is greatly appreciated!
Login to test node
Open the window services and locate Jenkins Slave services
Right click and select properties
Go to “Log On” tab
Make sure option “Local System Account” is selected and tick the option “Allow service to interact with desktop”
Click OK to save the setting
Restart the services.
I want to automate the process of running coded UI tests.
Do i need to have user logged in to the system?
Or is it possible to run it using XYNT service?
We are able to run our coded ui tests in an automated fashion using windows scheduler and the command line MSTest.exe. Take a look at the following MSDN article on how to accomplish this:
http://msdn.microsoft.com/en-us/library/ms182486.aspx
I believe you can use Microsoft's Test Manager to accomplish this as well but we don't have TFS (yet) so I can't vouch for it.
As far as I know, CodedUI Tests require an active desktop session, so running them under a service account without an open desktop session may work, but not without a session.
I used the mstest to run the coded ui test and use Windows TaskScheduler to schedule the test.
You can find detail in MSTest.exe command-line documentation.
When you want to run it automatically, there are two things you should pay attention:
1. Set auto log on
2. Close the screen saver
You need to take the snapshot of your enviroment with user logged in and then you can run your test through MS Test Manager or TFS (nightly build process)
Write the test's in Visual Studio 2010/2012, this means you can manually edit the test's the code, then you should run them in Test Manager,
BTW you need TFS and Test Manager to be the same year otherwise it won't work.
I've being doing a lot of these recently if you have any questions post it online and send me a message, I'll be happy to help.