arXiv:2509.05511v2 Announce Type: replace-cross Abstract: Cloud application services are distributed and have components across the stack working together to deliver the end-user experience. The widespread adoption of microservice architecture exacerbates failure management due to the increased number of service components. To be effective, strategies to enhance application service resilience need to be autonomous and developed at the service granularity, considering its end-to-end components. However, the massive amount of observability data generated by all these components across the service stack poses a significant challenge in reacting to anomalies and restoring the service quality in real time. Towards this end, the article presents the Topology-Aware Root Cause Detection (TA-RCD) algorithm, which leverages the end-to-end application service topology to accurately pinpoint root causes and facilitate swift remediation of cloud service anomalies. Additionally, because effective root-cause localization depends on identifying the most informative observability signals across the cloud service stack, this article presents a topology-guided metric selection strategy that selects relevant metrics from end-to-end layers to enable efficient, explainable, and accurate anomaly remediation. TA-RCD is comprehensively evaluated on the public RCAEval benchmark and a controlled in-house testbed. Experimental results show that the proposed approach consistently achieves higher average Top-3 and Top-5 localization accuracy than state-of-the-art baselines, improving AC@5 by at least 2$\times$ and AC@3 by at least 4$\times$ over standard RCD in the controlled testbed. On the public RCAEval benchmark, TA-RCD further improves metric-level AC@5 by 18\% and Avg@5 by 19\% over the strongest baseline.