SysResolve: Study on In-Context LLM Generation of Resolution Scripts

Harsh Borse, Utkalika Satapathy, Mainack Mandal, Bivas Mitra

Abstract

Microservices pose challenges for automated fault resolution due to their distributed and complex nature. We present SysResolve, a framework that automates the entire resolution pipeline by combining multi-modal Root Cause Analysis (RCA) with Large Language Models (LLMs). RCA outputs are converted to natural language and passed through a Retrieval-Augmented Generation (RAG) pipeline to produce executable scripts. We evaluated and experimented on two microservices applications with three LLM (LlaMa3-70B, GPT-4, Claude 3.7). Our analysis highlights significant gains of current LLMs generation power from few-shot learning, with SysResolve achieving expert-level remediation while reducing recovery time.

Bibtex

@inproceedings{borse2025sysresolve,
  title={SysResolve: Study on In-Context LLM Generation of Resolution Scripts},
  author={Borse, Harsh and Satpathy, Utkalika and Mondal, Mainack and Mitra, Bivas},
  booktitle={International Conference on Database and Expert Systems Applications},
  pages={130--136},
  year={2025},
  organization={Springer}
}

Cite

Borse, H., Satpathy, U., Mondal, M., & Mitra, B. (2025, August). SysResolve: Study on In-Context LLM Generation of Resolution Scripts. In International Conference on Database and Expert Systems Applications (pp. 130-136). Cham: Springer Nature Switzerland.

← Back to all publications