Microsoft now recommends using the Online TFS Test Hub over Microsoft Test Manager Stand Alone Client
for all test management requirements as they’ve been slowly moving all the features from the stand alone client to web
- Run with option Data Collectors via the Online TFS Test hub will launch MTM if selected and provides: Event Logs, Action Logs, Screen & Voice Recorder, System Info (additional logging)
- Online TFS Test hub via Web Browser based runner does not give these options
- TFS Online Test Hub now has Charts & Results Tabs
The biggest change is that pretty much everything’s moved to the web and you no longer have to fight your way through MTM’s terrible user interface design. The web interface makes sense. There’s no longer an artificial separation between planning your tests, running your tests, and tracking the results – everything is in one, unified interface.
The second biggest change is that Lab Management is done. The build system that was at the core of Lab Management has changed. The deployment features of Lab Management have been moved in to TFS Release Management. TFS Build vNext and TFS Release Management work in concert to get stuff deployed on to servers and the integration with the QA tools are now entirely optional.
You’ve now got two options for running tests:
1) run test cases using the web interface
2) run test cases using the MTM client.
You still have the option to run tests using the MTM rich test client. This makes sense because the really handy stuff from the MTM Test Runner was
the ability to create really descriptive bugs that contained screenshots and video recordings. Microsoft split that test execution functionality into its own EXE so you can still do all that great stuff.
Using the MTM Test Runner also lets you continue to do lightweight test automations using MTM Action Recordings. The idea with Action Recordings is
that a non-technical QA person would be able to create automated tests that help speed up the QA testing process and eliminate QA testing tedium. The Action Recordings feature makes for some awesome demos. In real life, it’s still pretty limiting and brittle. It only works with certain types of Windows application technologies and for web apps it only works with Internet Explorer.
If you want to do more free-form exploratory testing, there’s a new Exploratory Testing client for Chrome. It’s pretty good and it gives you the ability
to do screenshots and video recording, too. Last time that I checked, it was still in beta and as such it still had some sharp edges. Microsoft is definitely going in a good direction with that plugin though.
Another thing that’s missing (more like deprecated) is the Associated Automation feature. That used to let you hook a Visual Studio Coded UI test directly
to a Test Case and then have it run as part of a Lab Management build. Since Lab Management is pretty much done, this means that Associated Automations
are just about done. I wouldn’t sweat this too badly though because running just one Coded UI test is kind of a weird choice in the first place and it makes a lot more sense to just hook your entire Coded UI Test suite in to your vNext Build or Release Management flow. If you do that, the results of those tests aren’t tied to a specific Test Case but you still get to see the results in the Test Hub’s “Runs” tab.
Testing activities processes as of 2017 are:
- Create backlog items (in TFS)
- Create tasks (in TFS)
- Design test cases (in TFS, VS, or MTM)
- Link cases with tasks (in TFS, VS, or MTM)
- Associate automation script (in VS)
- Create test plan (in MTM or TFS)
- Execute test cases (in MTM or TFS)
- Check result (in MTM or TFS)