The pathophysiology of cluster headache (CH) is not well-known. For several years, the most widely accepted theory was that CH was triggered by hypothalamus with secondary activation of the trigeminal-autonomic reflex. However, it was recently suggested that the posterior hypothalamus might be an actor of the pain modulating network more involved in terminating rather than triggering attacks. To investigate this hypothesis, resting state fMRI could provide valuable information on functional connectivity between brainstem and hypothalamus, as well as other brain structures that could be involved in CH pathophysiology. In this framework, here we review recent studies investigating functional connectivity by means of resting state fMRI. Despite the important findings of these studies, we suggest that important steps in the comprehension of CH pathophysiology will be done when the scientific community will use the new methodological approaches recently suggested to study functional connectivity in the brainstem.