Explicitly addressing fault-tolerance during the requirements analysis phase facilitates the early detection of inconsistencies between functional and fault-tolerance requirements, which could potentially reduce the overall development costs. Most existing approaches use redundancy of services as a means to mask faults, where it is difficult to provide a systematic approach for modeling and analyzing the effect of faults on functional requirements during use case analysis. Moreover, providing masking fault-tolerance could be costly or impractical. This paper overviews a systematic approach for use case-based modeling of faults and failsafe fault-tolerance, where a failsafe fault-tolerant system at least meets its safety requirements when faults occur.