Daniel Sundmark

Per Erik Strandberg, Thomas J Ostrand, Elaine J Weyuker et al.

Software testing is sometimes plagued with intermittently failing tests and finding the root causes of such failing tests is often difficult. This problem has been widely studied at the unit testing level for open source software, but there has been far less investigation at the system test level, particularly the testing of industrial embedded systems. This paper describes our investigation of the root causes of intermittently failing tests in the embedded systems domain, with the goal of better understanding, explaining and categorizing the underlying faults. The subject of our investigation is a currently-running industrial embedded system, along with the system level testing that was performed. We devised and used a novel metric for classifying test cases as intermittent. From more than a half million test verdicts, we identified intermittently and consistently failing tests, and identified their root causes using multiple sources. We found that about 1-3% of all test cases were intermittently failing. From analysis of the case study results and related work, we identified nine factors associated with test case intermittence. We found that a fix for a consistently failing test typically removed a larger number of failures detected by other tests than a fix for an intermittent test. We also found that more effort was usually needed to identify fixes for intermittent tests than for consistent tests. An overlap between root causes leading to intermittent and consistent tests was identified. Many root causes of intermittence are the same in industrial embedded systems and open source software. However, when comparing unit testing to system level testing, especially for embedded systems, we observed that the test environment itself is often the cause of intermittence.

Stefan Karlsson, Adnan Causevic, Daniel Sundmark

RESTful APIs are an increasingly common way to expose software systems functionality and it is therefore of high interest to find methods to automatically test and verify such APIs. To lower the barrier for industry adoption, such methods needs to be straightforward to use with a low effort. This paper introduces a method to explore the behaviour of a RESTful API. This is done by using automatic property-based tests produced from OpenAPI documents that describe the REST API under test. We describe how this method creates artifacts that can be leveraged both as property-based test generators and as a source of validation for results (i.e., as test oracles). Experimental results, on both industrial and open source services, indicate how this approach is a low effort way of finding real faults. Furthermore, it supports building additional knowledge about the system under test by automatically exposing misalignment of specification and implementation. Since the tests are generated from the OpenAPI document this method automatically evolves test cases as the REST API evolves.

Stefan Karlsson, Adnan Čaušević, Daniel Sundmark

In recent years, GraphQL has become a popular way to expose web APIs. With its raise of adoption in industry, the quality of GraphQL APIs must be also assessed, as with any part of a software system, and preferably in an automated manner. However, there is currently a lack of methods to automatically generate tests to exercise GraphQL APIs. In this paper, we propose a method for automatically producing GraphQL queries to test GraphQL APIs. This is achieved using a property-based approach to create a generator for queries based on the GraphQL schema of the system under test. Our evaluation on a real world software system shows that this approach is both effective, in terms of finding real bugs, and efficient, as a complete schema can be covered in seconds. In addition, we evaluate the fault finding capability of the method when seeding known faults. 73% of the seeded faults where found, with room for improvements with regards to domain specific behavior, a common oracle challenge in automatic test generation.

Stefan Karlsson, Adnan Čaušević, Daniel Sundmark et al.

Automatic testing of mobile applications has been a well-researched area in recent years. However, testing in industry is still a very manual practice, as research results have not been fully transferred and adopted. Considering mobile applications, manual testing has the additional burden of adequate testing posed by a large number of available devices and different configurations, as well as the maintenance and setup of such devices. In this paper, we propose and evaluate the use of a model-based test generation approach, where generated tests are executed on a set of cloud-hosted real mobile devices. By using a model-based approach we generate dynamic, less brittle, and implementation simple test cases. The test execution on multiple real devices with different configurations increase the confidence in the implementation of the system under test. Our evaluation shows that the used approach produces a high coverage of the parts of the application related to user interactions. Nevertheless, the inclusion of external services in test generation is required in order to additionally increase the coverage of the complete application. Furthermore, we present the lessons learned while transferring and implementing this approach in an industrial context and applying it to the real product.