SELinux/SEAndroid policies used in practice contain tens of thousands of access rules making it hard to analyse them. In this paper, we present an algorithm for reasoning about the consistency of a given policy by analysing the information flows implied by it. For this purpose, we model SELinux policy rules using the Readers-Writers Flow Model (RWFM). Using this model, our method identifies all possible indirect flows due to a given policy that could lead to inconsistency. One of the main features of the method is that it not only identifies inconsistencies in the policy but also traces the rules that lead to inconsistency. To distinguish between benign and vulnerable indirect flows, we further categorise the indirect rules that directly contradict neverallow rules in the policy and hence have a high potential for information leak. We further rank the rules and domains based on the number of policy violations they cause. We have also implemented a tool FlowConSEAL based on the above method and have applied it on various SELinux/SEAndroid policies for providing a succinct feedback to the user.