Here’s the Deal
Slant is powered by a community that helps you make informed decisions. Tell us what you’re passionate about to get your personalized feed and help others.
You'll often end up having PhantomJS binaries connected via WebDriver to your testing framework, possibly using client/server especially if you want your test running with something else than Java. This means an overhead in terms of maintenance and performance, but still usually lighter than running a full browser (like Chrome, Firefox, IE). See More
It often happens when running on more then 5 (my measurement) JVM instances that the browser gets stuck and quits unexpectedly. This can be partially solved by running the instances one by one instead of parallel (this is a problem when testing Jenkins and Bamboo agents) but I don't believe this qualifies as a solution. The error is called UnreachableBrowserException. See More
Built by Slant
Find the best product
Stop wasting time searching endlessly. Lustre recommends the best products at their lowest prices.
4.7 star rating
Add to Chrome - It's FreeAdd to Edge - It's FreeAdd to Firefox - It's FreeAdd to Opera - It's FreeAdd to Brave - It's FreeAdd to Safari - It's FreeTry now - it's free
The open source community behind Selenium has a very large & engaged developer base. This ensures a stable support channel for the tool. In addition, various companies also provide support for Selenium. This active and multi-channel support provides a much more lucrative option for developers looking to implement Selenium in their workflow. See More
The WebDriver protocol is standardized by the World Wide Web Consortium which makes it easy for third-parties to get involved and contribute to Selenium. WebDriver has been the foundation of many additional automation projects besides core selenium and has become the de-facto standard for UI automation. See More
If you are running multiple instances on some browser and you use xvfb to run instances on agents on Bamboo or Jenkins you will potentially have a problem with using ports. The first instance you are starting takes on a port on the agent and then the second instance that would be run doesn't have a port to go to. You can partially solve the problem by dynamically assigning the next available port to the new instance that comes to be executed. This only works if we assume that all the instances start in different time intervals, but what you cannot know is which instance is going to get a port first if they start at the same time, then the instances will crash, those that started at the same time. See More