Web usage mining is to analysis Web log files to discover user accessing patterns of Web pages. The user access log files present very significant information about a web server. This paper is deal with finding information about a web site, top errors, link errors between the pages, etc. from the web server access log files. The aim of this study is to analysis the web server user access logs of Firat University to help system administrator and Web designer to improve their system by determining occured systems errors, corrupted and broken links by using web using mining. We found useful information about activity statistics like top errors, client errors, server errors within the visited pages etc. in a web server. The obtained results of the study will be used in the further development of the web site in order to increase its effectiveness.