CSnake: Detecting Self-Sustaining Cascading Failure via Causal Stitching of Fault Propagations
Shangshu Qian, Lin Tan, Yongle Zhang
Abstract
Recent studies have revealed that self-sustaining cascading failures in distributed systems frequently lead to widespread outages, which are challenging to contain and recover from. Existing failure detection techniques struggle to expose such failures prior to deployment, as they typically require a complex combination of specific conditions to be triggered. This challenge stems from the inherent nature of cascading failures, as they typically involve a sequence of fault propagations, each activated by distinct conditions.
This paper presents CSnake, a fault injection framework to expose self-sustaining cascading failures in distributed systems. CSnake uses the novel idea of causal stitching, which causally links multiple single-fault injections in different test workloads to simulate complex fault propagation chains. To identify propagation chains between faults, CSnake designs a counterfactual causality analysis of fault propagationsfault causality analysis (FCA): FCA compares the execution trace of a fault injection run with its corresponding profile run (i.e., running the same test without the injection) and identifies any additional faults triggered, which are considered to have a causal relationship with the injected fault.
To address the large search space of fault and workload combinations, CSnake employs a three-phase allocation (3PA) protocol of test budget that prioritizes faults with unique and diverse causal consequences, thereby increasing the likelihood of uncovering conditional fault propagations. Furthermore, to avoid incorrectly connecting fault propagations from workloads with incompatible conditions, CSnake performs a local compatibility check that approximately checks the compatibility of the path constraints associated with connected fault propagations with low overhead.
CSnake has detected 15 bugs that resulted in self-sustaining cascading failures in five widely deployed distributed systems, five of which have been confirmed with two fixed.
Ask about this paper
Your agent reads all of it.
Lune indexed this paper to the last equation, along with the top-tier papers that cite it. Ask a question and the answer quotes them.
Your agent calls
Luneget_paper_fulltext
Free to start. No credit card required.
Terminal
Install the CLIlune papers fulltext b0e5ca03-d352-4375-adc5-6ecc74ecc541Builds on26
- Directed Greybox FuzzingMarcel Böhme, Van-Thuan Pham, Manh-Dung Nguyen, Abhik RoychoudhuryCCS 2017 · 836 citations
- RetroWrite: Statically Instrumenting COTS Binaries for Fuzzing and SanitizationSushant Dinesh, Nathan Burow, Dongyan Xu, Mathias PayerS&P 2020 · 187 citations
- HEALER: Relation Learning Guided Kernel FuzzingHao Sun, Yuheng Shen, Cong Wang, Jianzhong Liu et al.SOSP 2021 · 59 citations
- TCP-Fuzz: Detecting Memory and Semantic Bugs in TCP Stacks with FuzzingYonghao Zou, Jia-Ju Bai, Jielong Zhou, Jianfeng Tan et al.USENIX ATC 2021 · 53 citations
- Automatic Reliability Testing For Cluster Management ControllersXudong Sun, Wenqing Luo, Jiawei Tyler Gu, Aishwarya Ganesan et al.OSDI 2022 · 44 citations
Related papers
- CAFault: Enhance Fault Injection Technique in Practical Distributed Systems via Abundant Fault-Dependent ConfigurationsYuanliang Chen, Fuchen Ma, Yuanhang Zhou, Zhen Yan et al.USENIX ATC 2025 · 7 citations
- Efficient Exposure of Partial Failure Bugs in Distributed Systems with Inferred Abstract StatesHaoze Wu, Jia Pan, Peng HuangNSDI 2024 · 15 citations
- Efficient Reproduction of Fault-Induced Failures in Distributed Systems with Feedback-Driven Fault InjectionJia Pan, Haoze Wu, Tanakorn Leesatapornwongsa, Suman Nath et al.SOSP 2024 · 4 citations
- KernelRCA: Facilitating Root Cause Analysis of Memory Corruptions in Linux Kernel with Contextual Causality ChainKangzheng Gu, Yifan Zhang, Yuan Zhang, Min YangUSENIX Security 2026
- Rose: Reproducing External-Fault-Induced Failures in Distributed Systems with Lightweight InstrumentationSebastião Amaro, Pedro Fonseca, Miguel MatosEuroSys 2026
