The assessment of operating systems robustness with respect to unexpected or anomalous events is a fundamental requirement for mission-critical systems. Robustness can be tested by deliberately exposing the system to erroneous events during its execution, and then analyzing the OS behavior to evaluate its ability to gracefully handle these events. Since OSs are complex and stateful systems, robustness testing needs to account for the timing of erroneous events, in order to evaluate the robust behavior of the OS under different states. This paper presents SABRINE (StAte-Based Robustness testIng of operatiNg systEms), an approach for state-aware robustness testing of OSs. SABRINE automatically extracts state models from execution traces, and generates a set of test cases that cover different OS states. We evaluate the approach on a Linux-based Real-Time Operating System adopted in the avionic domain. Experimental results show that SABRINE can automatically identify relevant OS states, and find robustness vulnerabilities while keeping low the number of test cases
This paper investigates the impact of state on robustness testing, by enhancing the traditional approach with the inclusion of the OS state in test cases definition. We evaluate the relevance of OS state and the effects of the proposed strategy through an experimental campaign on the file system of a Linux-based OS, to be adopted by Finmeccanica for safety-critical systems in the avionic domain. Results show that the OS state plays an important role in testing those corner cases not covered by traditional robustness testing
The reuse of Open Source Software (OSS) for safetycritical systems is seen with interest by industries, such as automotive, medical, and aerospace, as it enables shorter time-tomarket and lower development costs. However, safety certification demands to supply evidence about OSS quality, and a gap analysis is needed to assess if the cost to produce certification evidence is worthwhile. This paper presents an empirical study on an open-source RTOS (RTEMS). The study investigates the relationship between software complexity and the effort to achieve a high test coverage, which is one of the most impacting activity for certification. The objective is to figure out if, and to what extent, it is possible to predict such effort preventively, by looking at software complexity metrics. This would enable a preliminary screening and benchmarking of OSS items, supporting strategic decision making. The study shows that combining metrics with classifiers can achieve a good prediction accuracy
In present-day, software is taking over functionalities traditionally implemented in hardware, therefore the software architecture has been more complex and large. In such software architecture is common to be present an Operating System (OS). However, in safety domains (e.g., avionic, railway) it is mandatory to be compliant with a safety standard (e.g., D0178B), this means that evidence on the software life cycle of the software components, and therefore also of the OS, should be available. Those evidences that represent the certification package of the OS might not be available for commercial or Open Source OSs, hence their certification requires a complementary creation of evidence to serve as certification inputs. The certification process is costly, thus the system integrator must carefully select the candidate OS. Hence, it would be of great value to support the system integrator in selecting the more suitable OS to certify. In this position paper, we introduce our future research on the development of a Precertification kit (PK), that is, a framework that supports the evaluation of OS in what concerns certification requirements. Also, the PK is a valuable tool that can be integrated in the development tool-chain for the implementation of safer and higher quality OS and, provides additional evidences to use for the certification package
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.