Deadlocks should be eliminated in resource allocation systems such as flexible manufacturing systems. An iterative deadlock control policy is usually considered to be a natural solution with reasonable computational cost for a large-scale system where direct methods would be prohibitively expensive (and in some cases impossible) even with the best available computing power. This paper reviews the existing iterative deadlock prevention policies for discrete event systems that are modeled with Petri nets. A number of technical problems in the existing iterative deadlock control approaches are formulated and discussed. Their solutions are illustrated through case studies. We conclude that the suitability, effectiveness, and efficiency of an iterative deadlock control approach are sensitive to specific examples, and no general algorithm is found in the literature, which works well for all cases.