Safe Remediation as Risk-Constrained Intervention Decision in Microservice Systems

2026-07-22

AI Summary

This work addresses the challenge of costly erroneous repairs in existing automated remediation systems, which often lack the ability to assess intervention necessity and thus rely on manual approval for safety. The authors formulate safe repair as an intervention decision problem under risk constraints and introduce a three-dimensional risk decomposition framework encompassing impact scope, reversibility, and epistemic uncertainty. They further design a context-adaptive human-in-the-loop gating strategy that enables interpretable, workload-aware safety interventions. Built upon constrained Markov decision processes (CMDPs), offline policy learning, Chaos Mesh fault injection, and the RCAEval classification framework, the proposed approach reduces erroneous repair rates by 39% and improves repair success rates by 2.5 percentage points on the Train Ticket benchmark, while decreasing on-call escalation burden by 17% compared to fixed-threshold baselines.

Abstract

In modern IT operations (IT-Ops), the cost of an incorrect repair often exceeds the cost of no action at all. Yet existing automated remediation systems are designed to generate actions rather than to decide whether intervention is warranted, leaving safety as an afterthought enforced by manual approval. This paper makes three contributions to close this gap: (i) we reformulate safe remediation as a risk-constrained intervention decision problem and cast it as a Constrained Markov Decision Process (CMDP), in which the agent maximizes repair success subject to a bounded false remediation rate (FRR); (ii) we introduce a three-dimensional risk decomposition comprising blast radius, reversibility, and epistemic uncertainty, providing operators with an interpretable per-action safety interface; and (iii) we design a context-adaptive human-in-the-loop (HITL) gate that turns escalation from a binary failsafe into a bandwidth-aware control layer responsive to on-call load and business criticality. The full policy is learned offline from historical incident logs, enabling explicit control of the expected FRR. Experiments on the Train Ticket microservice benchmark with Chaos Mesh fault injection and an RCAEval-aligned fault taxonomy show that our framework reduces FRR by 39% while improving repair success by 2.5 points over a strong runbook baseline, and reduces on-call escalation load by 17% relative to a fixed-threshold variant.

Problem

• safe remediation
• risk-constrained intervention
• microservice systems
• false remediation rate
• automated decision-making

Innovation

• Constrained Markov Decision Process
• risk decomposition
• human-in-the-loop
• false remediation rate
• microservice remediation