Resilient Test Automation Framework for Microservices: Addressing Integration, Scalability, and Reliability

Abstract

The increasing adoption of microservices architecture has introduced significant challenges for software test automation due to distributed service dependencies, independent deployment lifecycles, heterogeneous technologies, asynchronous communication, and the difficulty of reproducing production-like failure conditions. Conventional testing approaches, particularly those that rely heavily on end-to-end testing, may become slow, brittle, and difficult to scale as the number of services increases. This paper proposes a Resilient Test Automation Framework (RTAF) designed to address three critical dimensions of microservices testing: integration, scalability, and reliability. RTAF integrates consumer-driven contract testing, service virtualization, integration testing, and automated chaos-based fault injection within a CI/CD-embedded testing pipeline. A central test orchestrator coordinates these testing layers and aggregates results from functional, performance, and resilience assessments. The framework additionally incorporates observability through metrics and distributed tracing to evaluate indicators such as contract verification success, test execution time, request latency, Mean Time to Recovery (MTTR), failover success rate, and error-budget consumption. The proposed framework is intended to reduce dependency-related test failures, improve the detection of service incompatibilities, and enable systematic verification of system behaviour under controlled failure conditions. An experimental evaluation using a containerised microservices reference application, Kubernetes, Pact, WireMock, Testcontainers, Chaos Mesh, Prometheus, OpenTelemetry, and Grafana is proposed to assess the framework's effectiveness. The study is expected to demonstrate that integrating interaction validation and resilience testing into a layered, automated pipeline can provide a more scalable and reliable approach to testing modern microservices systems.