One of the important elements of the new generation of the Web is the emergence of blogs. Currently a considerable number of users are creating content using blogs. Although Persian blogs have a short history, they have improved significantly during this short period. Because of fundamental differences between Persian and other languages, limited work has been done to analyze Persian blogs. In this work, a system named BlogDisc for automatic discovery and accumulation of Persian blogs is developed. This system uses content as well as link structure of the blogs. As an important part of this research, we propose an algorithm to recognize blogs that are not hosted on special blog hosts.