Citation:

Rajesh Balaji, 2025 "Autonomous SRE: SLO-Aware Multi-Agent Root-Cause Analysis and Verified Self-Healing for Cloud-Native Systems", ESP Journal of Engineering & Technology Advancements  5(4): 211-221.

Abstract:

Automated incident response in microservice systems fails in a predictable way. It blames the component that looks worst, which is often a victim, and then acts on it with confidence it has not earned. We present Autonomous SRE, a design that makes the service-level objective the judge at every step. Multi-window error-budget burn rates decide when to act. Four deterministic evidence agents (metrics, capacity, change and topology) and a coordinator separate culprits from victims and rank candidate remedies by likelihood and blast radius. Each remedy is then verified against the burn rate and reverted if it does not help, and a person is paged only when the hypotheses run out. We evaluate the design on a fault-injection simulator of a nine-component application, with eight fault types used during development, three held out until the design was frozen, and benign deployments and blips. On 240 test incidents of known types it spent 2.4% of a 28-day error budget per incident, against 12.4% for threshold runbooks, 12.6% for graph-based root-cause analysis with automatic remediation and 15.2% for manual response. It caused 0.05 harmful actions per incident, against 0.53 and 0.17. On never-seen faults, verification raised the share fixed without a person from 3% to 37%. Burn-rate triggering reacted more slowly than thresholds to faults that burn slowly. The simulator, source code, generated episodes, and archived results are supplied as a reproducibility package

References:

[1] B. Beyer, C. Jones, J. Petoff and N. R. Murphy, Eds., Site Reliability Engineering: How Google Runs Production Systems. Sebastopol, CA, USA: O'Reilly, 2016.

[2] B. Beyer, N. R. Murphy, D. K. Rensin, K. Kawahara and S. Thorne, Eds., The Site Reliability Workbook: Practical Ways to Implement SRE. Sebastopol, CA, USA: O'Reilly, 2018.

[3] J. O. Kephart and D. M. Chess, "The vision of autonomic computing," Computer, vol. 36, no. 1, pp. 41–50, 2003.

[4] L. Wu, J. Tordsson, E. Elmroth and O. Kao, "MicroRCA: Root cause localization of performance issues in microservices," in Proc. IEEE/IFIP NOMS, 2020.

[5] M. Ma, J. Xu, Y. Wang, P. Chen, Z. Zhang and P. Wang, "AutoMAP: Diagnose your microservice-based web applications automatically," in Proc. The Web Conference (WWW), 2020.

[6] P. Chen, Y. Qi, P. Zheng and D. Hou, "CauseInfer: Automatic and distributed performance diagnosis with hierarchical causality graph in large distributed systems," in Proc. IEEE INFOCOM, 2014.

[7] Y. Gan, M. Liang, S. Dev, D. Lo and C. Delimitrou, "Sage: Practical and scalable ML-driven performance debugging in microservices," in Proc. ASPLOS, 2021.

[8] A. Ikram, S. Chakraborty, S. Mitra, S. Saini, S. Bagchi and M. Kocaoglu, "Root cause analysis of failures in microservices through causal discovery," in Proc. NeurIPS, 2022.

[9] T. Ahmed, S. Ghosh, C. Bansal, T. Zimmermann, X. Zhang and S. Rajmohan, "Recommending root-cause and mitigation steps for cloud incidents using large language models," in Proc. ICSE, 2023.

[10] Y. Chen et al., "Automatic root cause analysis via large language models for cloud incidents," in Proc. EuroSys, 2024.

[11] A. Basiri et al., "Chaos engineering," IEEE Software, vol. 33, no. 3, pp. 35–41, 2016.

[12] Z. Li et al., "Gandalf: An intelligent, end-to-end analytics service for safe deployment in cloud-scale infrastructure," in Proc. USENIX NSDI, 2020.

[13] H. Liu, S. Lu, M. Musuvathi and S. Nath, "What bugs cause production cloud incidents?" in Proc. HotOS, 2019.

[14] P. Huang, C. Guo, L. Zhou, J. R. Lorch, Y. Dang, M. Chintalapati and R. Yao, "Gray failure: The Achilles' heel of cloud-scale systems," in Proc. HotOS, 2017.

[15] The Kubernetes Authors, "Configure liveness, readiness and startup probes" and "Horizontal Pod Autoscaling," Kubernetes Documentation. [Online]. Available: https://kubernetes.io/docs/

[16] Argo Project, "Argo Rollouts – Kubernetes progressive delivery controller," and Flux Project, "Flagger." [Online]. Available: https://argoproj.github.io/rollouts/ and https://flagger.app/

[17] L. Page, S. Brin, R. Motwani and T. Winograd, "The PageRank citation ranking: Bringing order to the web," Stanford InfoLab, Tech. Rep., 1999.

[18] J. Soldani and A. Brogi, "Anomaly detection and failure root cause analysis in (micro)service-based cloud applications: A survey," ACM Computing Surveys, vol. 55, no. 3, 2022.

[19] D. Huye, Y. Shkuro and R. R. Sambasivan, "Lifting the veil on Meta's microservice architecture: Analyses of topology and request workflows," in Proc. USENIX ATC, 2023.

[20] N. Zhao et al., "Identifying bad software changes via multimodal anomaly detection for online service systems," in Proc. ESEC/FSE, 2021.

Keywords:

Site Reliability Engineering, Service-Level Objectives, Error Budgets, Root-Cause Analysis, Self-Healing Systems, Microservices, Aiops