Note that Peskin and Schroeder write in fact the correct equation, i.e.,
$$(\Box+m^2) D_{R}(x-y)=-\mathrm{i} \delta^{(4)}(x-y).$$
Since by definition ##D_R(x-y) \propto \Theta(x^0-y^0)## this function is the retarded Green's function of the Klein-Gordon operator (modulo the usual conventional factor ##-\mathrm{i}## on the right-hand side).