A barrier alignment iterative learning controller is designed for spacecraft attitude stability control with both unbounded perturbation of moment of inertia and external disturbance. Under the condition that the expected attitude is const, the non-repetitive attitude tracking process is decomposed into periodic, continuous state alignment repetitive process, and a barrier alignment iterative learning controller is designed to guarantee the attitude stabilization. For the physical actuators saturation limiting, a barrier projection is utilized to estimate the unknown moment of inertia and external disturbance, and the asymptotic stability of the closed-loop system and the boundness of the attitude tracking error are analyzed. The simulation results show that the controller has the advantages of simple structure, strong robustness and bound attitude tracking error.