Skip to content

lightblue-platform/lightblue-tests

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

34 Commits
 
 
 
 
 
 
 
 

Repository files navigation

lightblue-tests

Functional and load tests for lightblue

Functional Tests

Black box (using only external interface) testing to verify the evaluated software is following the specifications. For lightblue that means testing the crud and metadata services

Setup

Assumes python 2.6 is installed on the host that will run the tests. To install dependencies:

sudo su -
yum install -y python-setuptools
easy_install argparse pyyaml pycurl
exit

Install pyresttest:

git clone https://github.com/lightblue-platform/pyresttest.git
cd pyresttest
sudo python setup.py install

Run All Tests

To run tests against metadata and crud services deployed on the same host:

cd functional
./all.sh <base URL> warning

Note the last argument suppresses logging which is set to debug by default at the time of writing this.

Load Tests

Testing the software to understand it limits, making sure that an anticipated peak will be supported and the software will behave correctly. It also gives support to determine the capability testing and then stress testing.

Web App Tests

Testing the management application interfaces. Some of the testing might overlap with functional tests but the core intent is to verify the applications function correctly.

How are functional tests different from unit tests?

With Arquillian and using in-memory databases there is a lot that can be tested about how the applications behave before being deployed. But there are some things you want to check against a deployed application:

  • Functional tests provide verification that a deployment of the application was successfull and that authentication and authorization is working.
  • Load tests stress the application ideally in conditions that match (or are) production. This cannot be done with unit tests.
  • Web app tests verify the deployed application behaves correctly and that authentication and authorization works

How do you define success?

For a functional test to be successful the API called must:

  • not change data in any way to impact production systes
  • return an expected HTTP code
  • return well formed JSON
  • pass all field level verifications for the response

For a load test to be successful:

  • meet all the requirements of a functional test
  • performance as measured over a pervious test execution set degrade more than X stdev [1]

[1] previous test set to verify against and acceptable deviation are not defined yet

For a web app to be successful:

  • TBD

License

The license of lightblue is GPLv3. See LICENSE in root of project for the full text.

About

Functional and load tests for lightblue

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 3

  •  
  •  
  •  

Languages