Extended Finite State Machine based Test Generation for an OpenFlow Switch

Asma Berriri 1, 2 Natalia Kushik 3 Zeghlache Djamal 4
1 R3S-SAMOVAR - Réseaux, Systèmes, Services, Sécurité
SAMOVAR - Services répartis, Architectures, MOdélisation, Validation, Administration des Réseaux
3 METHODES-SAMOVAR - Méthodes et modèles pour les réseaux
SAMOVAR - Services répartis, Architectures, MOdélisation, Validation, Administration des Réseaux
Abstract : Implementations of an OpenFlow (OF) switch, a crucial Software Defined Networking (SDN) component, are prone to errors caused by developer mistakes or/and ambiguous requirements stated in the OF documents. The paper is devoted to test derivation for related OF switch implementations. A model based test generation strategy is proposed. It relies on an Extended Finite State Machine (EFSM) specification that describes the functional behaviour of the switch-to-controller communication while potential faults/misconfigurations are expressed via corresponding mutation operators. We propose a method for deriving a test suite that contains distinguishing sequences for the specification EFSM and corresponding mutants. The proposed approach is implemented as a testbed to automatically derive and execute the test suites against different versions of an OF implementation. Preliminary experimental evaluation has shown the effectiveness of the proposed approach. Further on, the derived test suites have been able to detect a number of functional inconsistencies such as erroneous responses to the $Flow\_Mod$ adding rules with specific `action' values in an available Open vSwitch implementation.
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-02262841
Contributor : Asma Berriri <>
Submitted on : Friday, August 2, 2019 - 6:08:07 PM
Last modification on : Sunday, August 4, 2019 - 1:10:42 AM

Identifiers

  • HAL Id : hal-02262841, version 1

Citation

Asma Berriri, Natalia Kushik, Zeghlache Djamal. Extended Finite State Machine based Test Generation for an OpenFlow Switch. 2019. ⟨hal-02262841⟩

Share

Metrics

Record views

11