I'm trying to automate a third party windows application. I'm using UIAutomation which is part of .net 3.0 framework.
It works fine when run locally but if I deploy it in a windows server 2003 and make it run scheduled without logging in it does not work. I've done a fair amount of googling and I understand that for security reasons windows may not permit this. Is there a way to achieve this ? Its very similar to doing automated UI testing. How do people do large number of automated UI testing (without opening the server in remote and watching the test cases execute)
My solution is a windows application. (Ideally I would want it to be a web application which automates a manual task on a windows software installed on the server but that seems to be even more of a challenge)
Desperate ! Will provide more information on request.
Related
So that we may perform front-to-back web UI testing, we are using Selenium and ChromeDriver to automate page loads/interaction as part of our testing pack.
This is behaving as expected during developer testing (on a developer's local machine), but we are struggling to perform these checks as part of our continuous integration build.
Our server plant is *NIX based, and all of our CI infrastructure runs on these machines. So that we may test Chrome under Windows (our delivery mechanism), we have configured a Selenium Grid. When the CI tests run, they access the grid, in order to locate a Windows node to run the tests on.
We have had a Windows desktop provisioned solely for the purpose of running these tests. This contains our standard enterprise build of Windows 7. This machine will be periodically rebooted in-line with the IT department's update policy.
In an effort to ensure the Selenium server is always running, we have added the Selenium Server (running in "node" mode) as a Windows service. The selenium Server is configured to start-up ChromeDriver to invoke the simulated user-interaction.
However, when running the tests from CI they fail due to timeout. Our working theory is, the system user that is running the service cannot create interactive windows. A web search has raised reference to the "Session 0" problem, but with little to no constructive advice on how to move forward.
Starting the Selenium Server process manually from an interactive session is not a viable solution, as this is leading to brittle tests - which are failing due to an infrastructure problem, rather than a genuine test regression.
How can we have an instance of Selenium Server started via a Windows Service whenever the system reboots, that is capable of launching Chrome instances?
It could be easily done with NSSM.
Installation of services looks like these:
nssm install seleniumhub java -jar C:\selenium\selenium-server-standalone-2.45.0.jar -role hub -hubConfig C:\selenium\hub.json
nssm install seleniumnode java -jar C:\selenium\selenium-server-standalone-2.45.0.jar -role node -nodeConfig C:\selenium\node.json
It provides easily way to remove service if needed:
nssm remove seleniumnode confirm
Add destination to nssm to your PATH variable and run from console as admin
UPDATE April 2021
NSSM is not supported for more than 3 years. So please consider other options like winsw or any other. WinSW does the same job as NSSM and allows to keep run configuration in xml.
You cannot run Selenium Grid as a windows service ever since Windows Vista. Microsoft calls it "Session 0 Isolation". You could do it in Windows 2000 or XP but since the time that Vista came out, Microsoft no longer will let Grid interact with the desktop (or any other UI programs for that matter). Regardless of the fact that you still see that "interact with desktop" checkbox, it is a red herring. So, you MUST run Selenium Grid in the foreground on that server in order for it to get access to the session. If it is running Windows Server, you could in theory have multiple sessions and leave Grid running in the foreground on one of the non-zero user sessions.
Right now you can't help it - it used to work fine in session 0 but for the past few days after chrome update only works for interactive sessions.
Related bugs:
https://code.google.com/p/selenium/issues/detail?id=8029
https://code.google.com/p/chromium/issues/detail?id=422218
My preferred solution to this problem (and my default choice for running Selenium Grid as a service) is to use a simple tool called AlwaysUp. It has a free 30 day trial to try it out.
What to do:
Download AlwaysUp
Configure AlwaysUp to start the Selenium Grid node on startup
Configure AlwaysUp to run the Selenium node as a specific user (not the default System user)
This way the the node will run as a service, survive machine restarts and work with the latest version Chrome.
If the user account you use to login to the machine is different from the user account you specify to run the node as a service then you will not see the browsers pop up on the desktop as they are running in a different user session. The end result is that it is almost identical to running as a normal service but gets round the Session 0 issue.
Yeah, you should use NSSM. Important is, that you add your windows account in the "Log on" tab, or any other valid account. If you run your node with the "Local System account" option, you will get the session 0 problem. With a normal user session, the nodes run smoothly invisible in the background :)
we don't use selenium GRID, we were disappointed with its stability. We use a "Jenkins Grid", that is jenkins slaves nodes on various servers.
The slaves are services with the interact with desktop flag. They run as services with NSSM, and the SERVICE_INTERACTIVE_PROCESS flag. Making sure that NoInteractiveProcess is set to 1 (cf https://learn.microsoft.com/en-us/windows/desktop/services/interactive-services).
We don't have the fancy features of the grid (that is balancing according the browser types slots). Instead, we have Jenkins balancing the test jobs using a slave node or another.
Initially we did not use the interact with desktop flag, having browsers to run without "real" display, but the behavior was not very stable (especially with resize commands).
Hope this helps.
As I explained on this thread, I found that using a small paid tool FireDaemon Pro saved me a lot of time from trying to configure NSSM and other free tools.
It works well in the background, and restarts Selenium along with the server, which was my main requirement for running Selenium Standalone Server as a Windows Service.
This free tool would probably do it:
http://yajsw.sourceforge.net/
For that to work, you need a wrapper.conf file and a script to run the YAJSW wrapper. I takes time to read the documentation, but it is a free solution.
I wrote an example shared here, that installs JBoss7 as a Windows service.
Of course, you can simplify my example by a lot.
We are setting up our automation to run remotely so we can start incorporating them into the builds (you know, the whole CI/CD thing). These are a handful of important automated GUI tests that for obvious reasons, need an active VM to run. These are not browser tests, they are actually automated tests for a windows application so any support that Selenium brings to the table is off for us.
So now on to the challenge - how can I keep the VMs up and running without having to log into them using the Remote Desktop Connection to allow them to run the tests properly. Currently, I have to connect to them from my local machine and then minimize it and then I can kick off the builds. As soon as I exit however, the virtual machine is locked again.
I want the VMs to work completely independently from my machine, so I was skeptical about this approach because it seemed like it would still be tied to my machine only. Pretty much anyone in the company can log into the VMs from their machine using their credentials. What I would like to do is to programatically connect to the VM during my global TestStartup and then disconnect at TearDown. Is this possible to do? Has anyone had success or ran into similar situations with their automation integration process? We use a tool called LeanFT and NUnit as our test runner. .
Your idea to log in as part of the test is a bit fragile and prone to instabilities.
Here is the setup that works for every UI automation tool I've used for Windows
set up your VM to not lock / sleep /hibernate, etc.
Avoid using RDC (turn that feature off, even for admins if you can)
Only use the console viewer for your vm server
Limit access to those systems using the permissions in the VM server so that only you and your team can interact with them.
Here is why this works. You have already discovered that when you disconnect the RDP connection, the session locks and your automation fails. By using the vm console viewer, it's essentially like turning on/off the monitor connected to the system. By keeping them on all the time and not sleeping, they are always available for running tests.
We are using LeanFT and to encourage the stability of our tests, we have setup tasks to check the running processes to kill any stray leanft runtimes that didn't get closed cleanly from a prior run, as well as any stray applications that were not closed properly after a testing run.
These kind of issues are really annoying for UI automation.
In the end I found a solution. Not quite well but it works. All I did is created a Docker container and used it in UI automation job.
The container is composed by SSHD, Xvfb and xfreerdp, which let you connect to massive remote RDP, and because it use xvfb, a virtual display tool, it costs low resource.
Here's the image I created for your reference.
https://hub.docker.com/repository/docker/ariyuan/ubuntu1604_ssh_rdp
Before your UI automation start you just need to tell the container to open remote RDP connection to the machine where your UI automation hosted. In this case your display for UI automation will be kept all the time during the execution.(You can do it all by Jenkins with parameters to connect to different remote machine)
I have desktop application that can be installed on the users computer with "setup" msi file like any other application.
I'm developing web application and I need somehow to integrate the desktop application with the web application. Web application is developed using PHP (desktop application is developed using C#), and when clicking on one button on the web application, the desktop application needs to be launched.
Is there a way of doing this? I was thinking about a few scenarios:
-Maybe if possible to install the app directly on the server and to launch it from there?
-Maybe to be required the user to have the app installed on his computer and to call the app from there?
If possible I would prefer to not use the second approach because it's better if the third party to not be involved - it will be more user friendly if the application is launched directly. But any help will be appreciated because at this point I'm not sure if that is possible to be done at all.
You can install app at server an run it with exec() or "`" operator - but to control gui of app you need to use some like autoit ant etc.
If you ultimately decide to require your users to have the client application installed and expect majority of the users to be on Windows, you could register your application to handle specific scheme and parse its command line when started. More on this here:
http://msdn.microsoft.com/en-us/library/aa767914(v=vs.85).aspx
You'd register a custom scheme and then intercept in in the application's command line arguments. The whole URL will be passed as an argument, e.g.:
myapp://parameter1,parameter2
The browser will mostly ask whether or not the user trusts the application to handle this scheme with an option to remember this setting.
If you don't have access to the source code of the client application, you can develop a middleware, some sort of a launcher that handles the URL and then runs the client application after maybe modifying some configuration files based on the URL or otherwise controlling the third-party application to do as you with.
As for solution #1, I don't think C# matters if your application can run on Mono, so you should be able to just run it from PHP. However, this probably won't work on web hosting and you will have to get a VPS for that.
There are very good (security) reasons why it is hard to launch client-side processes from web browsers.
Do you have access to the source code of the C# app? If so, you could consider modifying it to take advantage of Microsoft ClickOnce deployment.
Some references:
http://msdn.microsoft.com/en-us/library/t71a733d.aspx
http://msdn.microsoft.com/en-us/library/t71a733d(v=vs.80).aspx
http://msdn.microsoft.com/en-us/library/6ae39a7c.aspx
http://www.codemag.com/Article/0902031
Here's an old article on deploying WinForms applications via ClickOnce with Visual Studio 2005: http://msdn.microsoft.com/en-us/library/ms953320.aspx
It can possibly be done with IIS but it can be cumbersome setting it up to run as the current user, especially since it would probably also need to run elevated for an app that needs to attach to the current user's desktop.
Easiest is to install a Windows LAMP distro (like WAMP: http://www.wampserver.com/en/) and then run the httpd.exe directly from an elevated (Admin) command prompt; do not run apache as a service!
Once you do this performing and exec("command"); call in php will bring up a desktop app as if it was invoked from an admin command prompt. Obviously you need to set up the apache server to be accessible from outside the local system, etc.
I did a windows application project with C#. Now, I wanna to run it as web application.(I mean, i want to install the application on server and the user from other places can connect to the server with a web link and test it).Since all the user do not have the visua studio software, I forced to do that. May you help me to figure this problem out?
Thanks in advance
You can't do that. You can have people be able to Remote Desktop to the server and then run the application. If you wanted it to be accessible via URL, you have to build a webpage.
Short answer: you can't run a Windows application directly in a web environment. And really there is NO WAY you should want that, either.
If you need remote access to the computer running the instance of your application, that can be done with the right software. I can think of TeamViewer, Radmin and Remote Desktop on top of my head.
If support for multiple concurrent users is required then you have two obvious choices:
Deploy your application on every clients, or
Make a Web application.
Now for option 1, there are different ways to do that. Depending of the complexity of your app you may require a Windows Installer setup package. For this there is multiple tools on the net. My personal favorite is WiX. Free, powerful and easy once you get the hang of it. Most installers will also ensure that requirements for the app to run are met, and install the missing components if necessary.
If all you have is a simple EXE file then you could be tempted to just package (ZIP) it and share it any way you want, but be warned: if the correct .NET Framework isn't installed on the client machine, it won't work. IMHO given the tools at your disposition to ensure a proper installation experience for the end user, you have no reason to do that. You can, however, consider using ClickOnce for this scenario.
If that doesn't help you, then please add more details on what your requirements are, maybe you'll get better advice.
PS: About "not every users having the Visual Studio software"... having VS installed should never, EVER be a requirement for the regular user.
You can't just run a Windows application as a Web Application. You need to re-write it as a web application. The users do not need Visual Studio to run your application as you have mentioned in your question. If you give some details of what your application does then maybe you will get some better help.
If you wrote your application using WPF it is possible to deploy it as an XBAP.
There are restrictions associated with deploying an application this way.
The minimum you would need to do is create a Web Application to host said XBAP.
The title is all-explaining I think. I want to develop .Net GUI application for Windows that will also run on Mac with no much extra effort. I guess WPF is not right choice here. Anyway, what do I do to make my Silverlight out-of-browser application to run on Mac ? Just install Silverlight on that machine ?
Silverlight's Out of Browser model works great on Mac, and has since its inception. Just develop the app and test it on Windows, it's basically guaranteed to act identically on Mac, unless you rely on a Windows-only feature (such as COM interop in Silverlight 4 Trusted Applications).
Edit: To further clarify my answer, nothing beyond the Silverlight Runtime install is needed to run OOB apps on a Mac. So if the user is able to run you app in-browser to do the install sequence, then they will be able to install the app OOB and use it.
This works with Silverlight 3+ only, of course, as that's when the feature was released. SO if the end-user has Silverlight 1 or 2 installed, they won't be able to install your app, but they could upgrade to the latest runtime to be able to do so (unless they're on a PPC Mac, as that only supports Silverlight 1. But most Macs these days are Intel anyway, so that's not really worth mentioning).
According to Getting Started with Silverlight 3, OOB is available on both Windows and Mac without an additional runtime. Here's a demo of the install process from channel9.
As far as specific hangups, that's beyond my experience. But it sounds like, in general, there are no special requirements to run SL apps out-of-browser on OS X.
I have built a state wide Silverlight 4.0 application that runs in browser and out of browser. In my experience it runs very well on a Mac, however I would say that it's not exactly the same. It's very close, but there are small differences in the install experience, and I've received some errors in Isolated Storage that I don't get on a Windows machine.
Overall I would definitely say Silverlight is the best cross platform development platform that I've ever used.
Having built a SL app for OOB on mac, I discovered one bug that gave me hard time. If you are using domain services, and have a method that try's to get the current authentication context to pull the userID on the server side, this will fail. It will work on Windows OOB, but not on Mac OOB.
The work around is pass in the userID from the client side to the domain service queries that need it.
I've had this issue using the standard SL Business app template and confirmed with various users on SL forums.