Survey of Attacks and Defenses on Consensus Algorithms for Data Replication in Distributed Systems

Distributed systems are fundamental to modern computing infrastructures, enabling scalability, fault tolerance, and efficient resource sharing. To maintain a consistent view of resources and support data replication across network nodes, systems such as distributed databases, cloud infrastructures, and decentralized storage rely on consensus algorithms. However, these algorithms are increasingly targeted by sophisticated attacks, threatening not only system integrity but also network-wide consistency. In this survey, we (i) systematically classify consensus algorithms employed for data replication in distributed systems, covering their operation, strengths, limitations, and deployment contexts, and (ii) analyze attacks that compromise their integrity, categorizing them according to strategy and targeted vulnerabilities. We review mitigation mechanisms and evaluate algorithmic resilience, drawing from both peer-reviewed and grey literature. Beyond literature aggregation, we identify cross-cutting weaknesses, such as overreliance on benign fault models, and highlight the lack of robust defenses against strategic adversaries. Our analysis reveals that many systems continue to treat targeted attacks as transient faults, underestimating their impact on consistency guarantees. By exposing these gaps and synthesizing known defenses, our study provides a foundation for future research aimed at securing consensus protocols in distributed data replication settings.