After several years of research, even today estimation of software reliability is extremely challenging. The undeterministic nature of the factors that contribute to the reliability estimation has led to the development of numerous statistical models for this purpose. One of the important factors that contribute to software reliability is testing. The aim of this paper is to use mutation testing for estimating the true reliability of a system and provide a framework for estimation of uncertainty bounds associated with testing and a measure for factors such as test coverage, time between failures and code coverage that contribute to the estimation of reliability and use mutation testing as a tool to create a software failure dataset. The approach is experimented with an open source search tool.