The World Wide Web is the most wide known information source that is easily available and searchable. It consists of billions of interconnected documents Web pages are authored by millions of people. Accesses made by various users to pages are recorded inside web logs. These log files exist in various formats. Because of increase in usage of web, size of web log files is increasing at a much faster rate. Web mining is application of data mining technique to these log files. It can be of three types Web usage mining, Web structure mining and Web content mining. Web Usage mining is mining of usage patterns of users which can then be used to personalize web sites and create attractive web sites. It consists of three main phases: Preprocessing, Pattern discovery and Pattern analysis. In this paper we focus on Data cleaning and IP Address identification stages of preprocessing. Methodology has been proposed for both the stages. At the end conclusion is made about number of users left after IP address identification.