Developed and integrated auto-balance scheduling validation tests for the longhorn/longhorn-manager repository, focusing on verifying best-effort behavior in node selection under constrained conditions. Leveraged Go and Kubernetes expertise to design tests that simulate edge-case scenarios, ensuring the scheduling algorithm reliably selects appropriate nodes when resources are limited. The new tests were incorporated into the existing suite, enhancing regression safety and supporting future maintainability. This work laid the foundation for improved reliability of the auto-balance feature, reducing the risk of suboptimal volume placements in production environments and contributing to overall quality assurance through robust, automated validation of scheduling logic.