Fault Detection for Message Passing Systems

Abstract

Many crucial dependable and secure services including atomic commitment, consensus and group membership, and middleware services (such as replica, communication and transaction services) use fault detectors. Through the use of fault detectors, the overlying service can be exempted from failure treatment and synchronization requirements. Fault detection is essential for proving that the services carried out are correct. In this paper, we first identify the necessary conditions to detect faults in a message passing system where multiple disjoint paths exist between each pair of endpoints. We then present the first fault detection protocol capable of detecting message meta-data modification in the presence of various message interferences in addition to other faults including omission faults, message replay and spurious messages using disjoint paths, where paths with faults are not known a priori. In addition, it authenticates message origins allowing Sybil attacks to be detected, identifies faulty paths, and classifies faults in the presence of multiple messages sent by various system processes. We establish the completeness and soundness properties of the proposed algorithm, i.e., it detects each considered fault and each detected fault is an actual fault, respectively. We also show that our algorithm does not yield a significant packet size and delay overheads. The algorithm shows the viability of the use of disjoint paths in fault detection.

An Application-Level Method of Arbitrary Synchronization Failure Detection in TTEthernet Networks

“…Another adaptive fault diagnosis algorithm 22 uses a de¯nite number of test rounds and sends a de¯nite number of messages to¯nd the fault conditions in the CAN-based distributed embedded system. Karaata et al 11 present the protocol to detect Byzantine faults using disjoint paths. However, these methods may not be available for TTEthernet network because the diagnosis of clock failure has to be based on accurate time information which is not included in common tra±c classes.…”

“…However, any two CMs do not always make the same decision due to the di®erences of Eqs. (11) and (12). For a correct CM, Eqs.…”

“…In this case, the fail-arbitrary failures of CMs may do harm to the quality of clock synchronization and even destroy the global time base of the whole network. Furthermore, there are malicious attacks 10 which can be considered as fail-arbitrary failures 11 in some cases and then there may exist multiple fail-arbitrary failure devices. Experiments 12 indicated that the clock synchronization mechanism can be driven by dedicated tester nodes for FlexRay 13 with the same clock synchronization algorithm as TTEthernet.…”