I'm drowning in the Selenium documentation. What I've got so far is the ability to record a script in FireFox, export it to the new WebDriver format (JUnit4), open and tweak the test in Eclipse, then run it as a JUnit test in Eclipse.
What I'm wanting next:
to run the same test on multiple browsers
to have a suite of tests run on a remote server at a set interval
receive notification if a test fails
Is this possible using the path I've started down above? It's hard to tell, but I'm not sure that WebDriver is compatible with the RC server. Or, should I just take my suite of JUnit tests and integrate them with something like Cruise Control?
Thank you...
to run the same test on multiple browsers
See: How to run Google Chrome with Selenium RC?
to have a suite of tests run on a remote server at a set interval
This is hard, especially when you are running on headless (terminal only) machine. See this article. But not impossible, read further.
receive notification if a test fails
The typical configuration is to use CI server like Bamboo, take advantage of cargo-maven-plugin to deploy your application and run Selenium tests during integration-test phase. Easier said than done. Since CI server is typically headless, you will need freaks of nature like xvfb, taking a lot of screenshots since you have no idea what is happening during the build when it fails, and so on... But it is possible, we managed to setup such a process in our company and are very happy for it.
Bromine helps you to run on several browsers. Regarding "Scheduling" as already said, maven would help. Try to gather infos about Hudson. It helps a lot in scheduling. It supports selenium too. It helps to send notification also.
Related
How does Puppeteer handle PWA installation? Can I just assume that no installation will take place?
I searched for "puppeteer pwa" but could not find any info.
Puppeteer is really a tool for testing an application. It is a headless browser and therefor would not offer any real way to test or simulate the installation flow.
The service worker should register and the initial life cycle should trigger. However it is headless, which means state is not preserved. So you are more or less doing an InCognito session each test run.
What is the best way to run an API test that was created via the IBM API Connect Test and Monitor ?
I published a test and I would like my CI sever (jenkins or azure devops) to run it?
Many thanks,
Assaf
Great question Assaf. It is currently not possible in the product, however, is on the roadmap and is coming soon.
We will have APIs and Webhooks to help execute tests as part of your CI/CD processes. Meaning, the same tests you generated using the test composer and run in production can be recycled to ensure your deployments are error free.
As well, Jenkins is an industry standard so we will be providing a plugin to help facilitate the API testing processes via GUI. More details to come, will update this space when it does.
Alternatively keep your eye out here: http://ibm.biz/apitest
I have integrated my selenium Webdriver scripts (using TestNG) with Hudson. i invoked my job through ant . My problem is my scripts are not running successfully also the IE browser is not getting opened However Build is creating successfully.
Note -
1) I am triggering build on Hudson from different machine on the same network with administrative access.
2) I have used excel 2007 for developing the script ( data driven framework) however on Hudson server its open office.could that be the issue?
At this moment i am not using selenium grid .please provide any suggestion
The reason for this is because Hudson/Jenkins is running as a Windows service. Recently, Microsoft changed services so that they run on their own invisible desktop. This didn't used to be the case until a few years ago. For this reason, even if you check "interact with desktop" , the desktop that is being referred to is invisble .
So, what you have to do is run Jenkins (not Hudson) as a service only for the master server. For the Selenium tests, you need to run another slave Jenkins server as a foreground shell process and the selenium tests need to be launched from that Jenkins instance instead.
If you think the instructions for setting up a slave are too hard, then you should know that it isn't required that you run as an official configured slave. You can run as a separate slave (make sure port number is different).
Also, if you are running Grids and Nodes on Windows, you might like these scripts I made.
I'm using the JUnit runner in IntelliJ to run all tests in a package. When I run my tests in isolation they pass, when I run the entire package some fail, suggesting interplay between the tests. I'd like to force the tests to be run in serial rather than parallel to prove this - can anyone tell me how I can configure this to happen?
Thanks
Tests don't run in parallel, at least not in IDEA 9.x. Unless you are using IDEA 10 EAP, your problem is different and you need to provide more details to get a good answer.
I'm using selenium-client to run a few tests, but Selenium seems to be using my development database. How can I point it to use my test DB?
You need to point Selenium to a site that uses the test databases. Selenium talks to your rails, not the database. So, you need to point Selenium to a web app whose config points to your test db, not the dev db.
Hope that helps.
You need to point the Selenium test at an instance of your app that uses the test database.
I wrote a separate deploy task in my capistrano deploy.rb that deploys the application to a testing server, points it at the test db, and loads the selenium-specific fixtures. Selenium treats a deployed app as a black box, so it's up to you to fit all that in to your existing deployment setup.