Selenium test tryingnto use firefox instead of chrome - selenium

I have the following in my selenium-webdriver test (the test is written as coffee-script):
#driver = new selenium.Builder()
.withCapabilities(selenium.Capabilities.chrome())
.build()
Yet when I run this test the error I get is:
Error: Failed to install profile; firefox terminated with Result(code=-1073
741819, signal=null)
I am perplexed why its trying to start firefox (there is no reference to firefox in my code). Also, at one point this was working. I am not sure what changed in my laptop which has resulted in this behavior.
Thanks for any help.

Related

Chrome failed to start - Chrome version 75.0.3770.100

We have selenium tests in C# running on jenkins. With the latest version of Chrome 75 the tests are starting to fail during execution with the error "Chrome failed to start"
I went through some articles and have implemented the below code as part of my arguments for Chrome.
chromeOptions.AddArgument("--enable-automation");
chromeOptions.AddArgument("--no-sandbox");
chromeOptions.AddArgument("--disable-extensions");
chromeOptions.AddArgument("--disable-print-preview");
chromeOptions.AddArgument("--disable-gpu");
chromeOptions.AddArgument("--disable-software-rasterizer");
chromeOptions.AddArgument("--disable-gpu-sandbox");
chromeOptions.AddArgument("--disable-features=VizDisplayCompositor");
chromeOptions.AddArgument("--start-maximized");
chromeOptions.AddArgument("--disable-dev-shm-usage");
chromeOptions.PageLoadStrategy = PageLoadStrategy.Normal;
This happens during jenkins execution only.
I'm seeing this on Chrome 76. The solution seems to be to remove the --disable-software-rasterizer argument.

Selenium WebDriver - Missing 'text' parameter (IndexOutOfBounds) - SendKeys

I'm trying to run a test on google's website and when I send the text I'm saying the following message:
Missing 'text' parameter (IndexOutOfBounds)
Code:
browser.FindElement(By.Name("q")).SendKeys("a");
What I do not understand is that until yesterday the code was running normally.
Can someone help me solve this error?
tks!
Version:
geckodriver-v0.16.0-win64
Selenium Client & WebDriver - C# 3.4.0
Firefox 53.0v
Try updating your Selenium.WebDriver and Selenium.Support packages to latest (v3.4.0). If you already have v3.4.0 make sure you have rebuilt all.
I just hit the same issue and updated v3.2.0 -> v3.4.0 and that fixed it for me.
(Just for context, I ended up with that error after getting 'Expected [object Undefined] undefined to be a string (IndexOutOfBounds)' following an automatic Firefox update to v53. The fix for that was to upgrade to geckodriver v0.16.0)
#Kayh and #NickBaker
There is still an issue: how did that work?
https://github.com/mozilla/geckodriver/issues/511
I am using Firefox 53.0.3 (32-bit), geckodriver 0.16.1 to drive browser with Selenium 3.4.0 but still getting
System.InvalidOperationException : Missing 'text' parameter (IndexOutOfBounds)
EDIT:
https://github.com/mozilla/geckodriver/issues/511 fixed now on 02 July 2017. Combination of Selenium client binding 3.4.3, geckodriver 0.17, Firefox 54 works now.

Selenium Webdriver + Protractor + CucumberJS

Hello All
I've discovered an issues, which I'm not sure it really is an known issue with the framework or it's due to the versions of Protractor + CucumberJS that I am using, these are:
"cucumber": "~0.7.0",
"protractor": "~2.5.1",
"selenium-standalone": "3.0.0",
I'm using this versions as of today, I know that CucumberJS was separate from Protractor, but since our current project is using Node < 4, I cannot update it.
Having said this, the error that I'm having is when I run the test suite with protractor config.js and the seleniumAddress attribute commented (it starts up a webdriver automatically at random port)
The error that appears, when the suite finishes is:
throw new Error('This driver instance does not have a valid session ID ' +
^
BUT when I run the exact same suite, starting webdriver manually and setting the selenium address, the test passes!!!
The suite contains:
17 scenarios (17 passed)
55 steps (55 passed)
Does anyone knows what could be the reason? My first thought is that starting the server manually it would make webDriver slower, and therefore test actions would be too, so... but I had compared both time executions and there's only 1 second difference between both....
I had realized, that the problem with all is that since I had commented out the seleniumAddress line, Protractor was starting up webDriber-manager automatically, and when I added the line and manually started WebDriver, the error did not happen again. NOT sure what's the matter with it starting up automatically but, in case you ran into the same problem here's the solution!
Thanks

selenium phpunit click function not working

I am using friefox 27.0.1 and selenium stand alone 2.15 i am getting below error for click()
RuntimeException: Invalid response while accessing the Selenium Server at : ERROR: Command execution failure. Please search the user group at https://groups.google.com/forum/#!forum/selenium-users for error details from the log window. The error message is: Argument 1 of EventTarget.dispatchEvent does not implement interface Event.
You are extremely out of date. Upgrade to the selenium standalone of 2.40.0 and this will fix your issue.
Source:
Person that had experience with this exact issue.
I updates selenium to 2.39.0 and firefox version to 28 then it worked fine for me.

Selenium XHR ERROR: Response_Code = -1

I am using the newest version of Selenium RC and firefox 3.5 When I run my test from eclipse I get this error XHR ERROR: Response_Code = -1 Error_Message = Request Error.
Firefox and Selenium RC open up fine, it seems to try to connect to the remote site I want, but then firefox crashes, any ideas?
I'm pretty sure this is an issue fixed in trunk (and will come out in the pending 1.0.4 release). If you download 2.0a5, this includes a 1.0.x compatible Selenium server. You should be able to drop the -standalone JAR in place and have it work. If not, building from trunk is your next best bet.
Alternatively, you could try to modify the open command call. In dynamic languages, such as Ruby, this is fairly straightforward. open() takes a URL and a boolean as its values. You'll want to invert the logic for the second param (I think it's false by default, so you'll want true).