Slope failure and debris flow cause a lot of casualties and loss. The early-warning of slope collapse is essential to ensure safety of human beings and assets. Based on fiber optic sensing technology and internet of things, a new sensing transducer for internal earth pressure in the soil slope is proposed, fabricated, and tested in the study. The working principles, theoretical analysis, laboratory calibrations and discussions of the proposed pressure transducers are elaborated. Extensive evaluations of the resolutions, physical properties and response to the applied pressures have been performed through modeling and experimentations. Finally, a field soil slope was instrumented with the developed fiber optic sensors and other fiber sensors. Through IoT and cloud computing platform, the stability of the soil slope was analyzed. In the cloud platform, the numerical simulation is carried out by considering the slope internal deformations, rainfall infiltration and limit equilibrium state of the soil slope. With the new real-time fiber optic sensing technology and rainfall data, the factor of safety for the soil slope was calculated which was further used to determine health condition of the instrumented slope. The performance was evaluated and classified into three categories for the application layer. It proves that the proposed early-warning system is efficient for health monitoring and preformation evaluation of soil slopes.