Fast-charging stations that supply energy to electrical vehicles at high power rates may incorporate energy storage to avoid high currents to the grid and reduce peak-demand costs. We demonstrate the potential for utilizing such infrastructure for the provision of ancillary services (AS) to power grids. The main challenge we address is the coordination between the stochastic demand for the real-time delivery of AS while charging the randomly arriving electrical vehicles. The method presented here, employing two-stage stochastic optimization, calculates both the maximum capacity of AS the system can provide and handles optimally the real-time delivery of control-energy to the grid upon request by integrating intraday trades. To showcase our concept we have employed the ABB TOSA electrical bus charging network and demonstrated, through simulation studies, with realistic models and energy prices, that the method is technically feasible, can deliver a considerable level of control energy, offers substantial economic benefits, and has only a minor impact on the lifetime of the storage system.