一句话：提出 TSGen，用 LLM 从历史事故报告自动生成结构化、可迭代更新的故障排查指南。
云系统可靠性依赖及时更新的故障排查指南（TSG），但人工维护效率低且易过时。本文提出 TSGen，一个基于大语言模型的自动化流水线：先过滤并分类历史事故数据，保留诊断相关类别；再蒸馏核心事故以保证多样性与泛化性；最后将根因与解决方案组织成有向无环图（DAG），生成结构化、可复用的动态指南。在 Microsoft 真实事故数据上，TSGen 的故障覆盖率约 54.8%，检索准确率较基线提升约 3 倍；系统支持随云环境演化进行迭代更新。人工评估显示，值班工程师对生成 TSG 的评分显著高于人工编写版本。
Maintaining up-to-date troubleshooting guides (TSGs) is critical for the reliability of cloud systems, yet manual maintenance often leads to inefficiencies and outdated documentation. This paper proposes TSGen, an automated pipeline for generating high-quality, structured TSGs from historical incident reports using large language models (LLMs). Our approach consists of three stages: (1) filtering and classifying incident data into diagnostically relevant categories, (2) distilling core incidents to ensure diversity and generalizability, and (3) organizing the distilled knowledge into a directed acyclic graph (DAG) that captures root causes and resolutions in a structured manner. By leveraging real-world incident discussions, TSGen produces dynamic and reusable guides tailored for live troubleshooting. Experiments on real-world incidents from Microsoft demonstrate that TSGen achieves 54.8% incident coverage and approximately 3× higher retrieval accuracy compared to baselines. Furthermore, the system supports iterative updates, allowing guides to evolve alongside dynamic cloud environments. Human evaluation shows that on-call engineers rate these generated TSGs significantly higher than human-crafted ones.
@inproceedings{XiaoZGZWBRRM26,
title = {TSGen: Automated Troubleshooting Guide Generation},
author = {Yi Xiao and Hongyu Zhang and Daniel Genkin and Chaoyun Zhang and Rujia Wang and Chetan Bansal and Bhala Ranganathan and Saravan Rajmohan and Minghua Ma},
booktitle = {FSE},
year = {2026},
doi = {10.1145/3803437.3805239}
}