There are many tools that create a metric for Website performance, for example https://tools.pingdom.com/. Regarding the question if "Too long" is based on opinions, I'd say too long is every time that can be reduced.

You could check if packages are aggregated as far as possible (Sure, aggregating all resources doesn't really change the amount of data transferred but every single call creates avoidable overhead), if every resource is minimized and gets delivered gziped and of course if expire dates make sense so the browser can effectively cache them. There are incredible amounts of checks that you could run to make sure that your pages become as fast as possible, which of those are feasible for your application is another question though.

Is there any way to display more info on the JMeter console during the test execution in non-GUI mode.

I would like to see the calls names next to the summary so I can know exactly what call has failed while running the test in non-GUI mode. The idea behind is to stop the test the moment I see the error and easy way to debug and identify the call.

For ex: one of the call name of script from the below screenshot is "SupportUnemploymentRequest_PJLReport_28_SOAPAction_RptReportProvider_GetReportWsDtoDataSetBytes"

At my company we are developing a Desktop application and regression after regression we finally decided to implement the acceptance tests in addition to the unit tests.

I am relying on this project to use selenium in our application (I tested it and works flawlessly, we use QML to build the front-end and it's translated in XML).

While I was designing the tests architecture, anyway, I had a doubt: how can I guarantee that every single test scenario is self-contained and totally independent from the others?

In web application testing this is not a big deal since the front-end is usually divided among multiple html pages and urls. The "given" preconditions can be easily guaranteed by requesting the right url and, usually, in the worsts scenarios, it's just needed to execute a few instructions.

With desktop applications, however, it's a totally different story. You can't just ask for the page you want to comply with the preconditions. To get to a certain state you may need to execute multiple commands. In our case many of these commands require 30+ seconds to be executed. It seems like that the best way to test every feature, with their scenarios, is to execute them one after another, in a precise order. But that's totally against the testing principles.

What I am missing? Which are the best practices in this case? Is just selenium too restrictive for our needs?

I need to locate a link on the page and copy it or store it as a string without clicking it in a protractor test. I can locate the element but am not sure how to store it as a string. Suggestions?

When interacting with different elements and applying multiple actions it may be quite convenient in some situations if we could do:

WebElement myInput = driver.findElement(By.id("my-input-id"));
myInput.clear().sendKeys("test");

Instead of:

WebElement myInput = driver.findElement(By.id("my-input-id"));
myInput.clear()
myInput.sendKeys("test");

But, the problem is clear(), sendKeys() and other "action" methods cannot be chained as is - they return void and not the current WebElement.

What would be the easiest way to achieve chaining multiple actions?

30 Good Practices to improve Exploratory Testing

• Use a bug tracking system
• Use boundary testing of values
• Consider using testing personas
• Use happy, sad and optional paths
• Become skilled at reading server logs
• Learn about usability and accessibility
• Learn to use emulators and simulators
• Learn more about the customers needs
• Be present in business process meetings
• Maintain an easy to use library for devices
• Ask immediate product questions in person
• Understand the company metrics for quality
• Develop good relations w/ product managers
• Learn the business domain and the key factors
• Learn about visual, auditory and tactile disabilities
• Understand the database design and implementation
• Learn about both internal and external business processes
• Use the value proposition to determine what bugs to report
• Get documentation of application workflows and functionality
• Learn to use local virtualization, e.g. Parallels, VirtualBox, etc.
• Categorize bug severity in a way that is relevant to the business
• Learn relevant knowledge for localization and internationalization
• Learn and use remote services such as browserstack and saucelabs
• Become skilled in using browser tools such as console and inspectors
• Use automation for repetitive tasks and humans to spot other changes
• Replicate real user environments for network speed, sunlight, heat, etc.
• Ensure the system used for exploratory testing closely mimics production
• Use company data to determine what devices and browsers customers use
• Use a well integrated 'drop to cloud' system* for screenshots and links to them
• Gain a wide variety of experiences in life to help think about different approaches

For Agile Folks:

* CloudApp for Mac is a good example

1. First of all it is highly unlikely you will be able to spawn 100 000 virtual users using single JMeter instance so most probably you will have to go for Distributed Testing, it means that you will need to get several JMeter Slaves which will be orchestrated by JMeter Master
2. The number of virtual users which can be simulated from one host varies and depends on many factors, i.e. test nature (hardware specifications, request/response sizes, number of samplers, number of preprocessors/postprocessors/assertions, etc.) so you need to identify how many users you will be able to simulate using JMeter from a single machine. Once you figure that out you will be able to calculate how many JMeter Slaves you need.

3. Always stick to JMeter Best Practices, i.e.

   • Run your test in non-GUI mode
   • Use only those pre/post processors and assertions which are absolutely required
   • Disable or remove all the Listeners during the test run as they don't add any value and consume a lot of resources.

Assuming all above you should be able to simulate 100 000 users. Just remember one thing: all JMeter Slaves are basically executing the same test plan so you will need to divide the total number of users by the number of JMeter Slaves, examples:

• 10 JMeter Slaves, each has 10 000 users - you will get 100 000 users
• 20 JMeter Slaves, each has 5 000 users - you will get 100 000 users
• etc.

I downloaded HAXM installer from intel and used the Google APIs Intel Atom (x86). This seemed to resolve my issue.

I am running my selenium test thru Selenium Grid and I couldn't take screenshot on the node from my hub. I have followed the code from the Selenium HQ:

public class ScreenShotRemoteWebDriver : RemoteWebDriver, ITakesScreenshot
{
    public ScreenShotRemoteWebDriver(Uri RemoteAdress, ICapabilities capabilities)
    : base(RemoteAdress, capabilities)
    {
    }

    public new Screenshot GetScreenshot()
    {
        // Get the screenshot as base64.
        Response screenshotResponse = this.Execute(DriverCommand.Screenshot, null);
        string base64 = screenshotResponse.Value.ToString();

        // ... and convert it.
        return new Screenshot(base64);
    }
}


public static void takeRemoteScreenShot(string fileName)
{
    string currentDate = DateTime.Now.ToString("ddd, ddMMMyyyy HHmm");
    ScreenShotRemoteWebDriver webDriver = new ScreenShotRemoteWebDriver(new Uri("http://localhost:4444/wd/hub"), DesiredCapabilities.Chrome());
    Screenshot ss = ((ITakesScreenshot)webDriver).GetScreenshot();
    string screenshot = ss.AsBase64EncodedString;
    //byte[] screenshotAsByteArray = ss.AsByteArray;
    string dir = Settings.DirScreenShots;
    ss.SaveAsFile(dir + currentDate + fileName + ".png", ScreenshotImageFormat.Png);
}

My test is failing due to this error:

 Result StackTrace: 
at SeleniumE2E.Utilities.ScreenshotCapture.takeRemoteScreenShot(String fileName)
   at SEL_ADAccountReactivate2.SmokeTest.RunTest() in C:\Users\Documents\Visual Studio 2015\Projects\App\App\Tests\SmokeTest.cs:line 16
Result Message: 
Test method App.SmokeTest.RunTest threw exception: 
System.MissingMethodException: Method not found: 'Void OpenQA.Selenium.Screenshot.SaveAsFile(System.String, OpenQA.Selenium.ScreenshotImageFormat)'.

I worked for Testbirds on some tests, but they invite me to test really rarely. I have more free time, so would like to work more. Are there another crowdtesting platforms? Which ones have you tried?