paperbot · PL 论文追踪

全部论文 待读 收藏

P2 Declarative Test-Native Software Fault Injection

声明式、测试原生的软件故障注入
FSE Proceedings of the 34th ACM International Conference on the Foundations of Software Engineering2026匹配度 2/10 · 该论文关注软件故障注入与韧性测试,提出声明式测试框架,属于软件工程/测试与 DSL 方向,与用户核心的 LLM/Agent 结合静态分析、形式化验证、浮点或贝叶斯程序分析均无直接关联。仅在“领域特定语言/测试工具”层面与 PL 略有交集。引用 0 inbox
Matthew MacRae-Bovell
crossref: 2026-09-12dblp: 2026-09-12

AI 速览

一句话:提出 ChaosSpec 原型框架,将故障注入场景编排以声明式方式集成进常规自动化测试套件,降低韧性测试门槛。

软件故障注入(SFI)是一种用于验证系统在网络延迟、依赖服务中断和资源耗尽等不利条件下行为的有力韧性测试技术。然而,现有工具通常依赖专门的基础设施与工作流,操作复杂且与常规自动化测试差异显著,导致故障注入在企业开发团队之外并不常见,韧性测试常被视为专门活动而非日常开发的一部分。本文提出把故障注入场景编排直接集成到开发者熟悉的自动化测试套件中,并实现原型测试框架 ChaosSpec。开发者可以用声明式方式编排服务与故障场景,而无需掌握专门的基础设施知识。摘要未报告具体实验或评估结果。

问题
现有软件故障注入工具依赖专门基础设施与工作流,操作复杂,难以融入日常自动化测试,导致韧性测试难以在普通开发团队中普及。
方法
提出原型测试框架 ChaosSpec,将故障注入场景编排以声明式方式直接集成进常规自动化测试套件,使开发者无需专门基础设施知识即可编排服务与故障场景。
评估
摘要仅介绍原型框架的设计理念,未报告具体实验、案例研究或量化评估结果。
原文摘要(Abstract)

Software fault injection (SFI) is a powerful resilience testing technique for validating how systems behave under adverse conditions such as network latency, dependency outages, and resource exhaustion. However, fault injection remains uncommon outside enterprise software development teams due to the operational complexity of existing tools, which often rely on specialized infrastructure and workflows that differ substantially from regular automated tests. As a result, resilience testing is frequently treated as a specialized activity rather than an integrated part of everyday software development. In this work, we present an approach that integrates fault injection scenario orchestration directly into familiar automated test suites. We introduce ChaosSpec, a prototype test framework that enables developers to declaratively orchestrate services and failure scenarios without specialized infrastructure knowledge.

链接与引用

DOI 原文 · PDF(开放获取) · DBLP

BibTeX
@inproceedings{MacRae-Bovell26,
  title = {Declarative Test-Native Software Fault Injection},
  author = {Matthew MacRae-Bovell},
  booktitle = {FSE},
  year = {2026},
  doi = {10.1145/3803437.3804442}
}