Philip,
There is an issue with gravity flux through a closed surface.
For comparison, let's start with Gauss's law used in electrostatics. If I had two co-centric metal spheres both with charge +Q, I could make a closed surface between the inner and outer spheres. A test charge on that closed surface would experience a force due to the electric field. The electric field lines would start at the surface of the inner sphere and end at the induced negative charge on the inside of the outer sphere. Using the correct surface integral I could find the charge +Q on the inner sphere.
If I did the same thing with two co-centric spheres of mass M and a small test mass to find the gravitational field, I could be disappointed. Since there is no anti-gravity there is no induced anti-gravity charge on the inside surface of the outer sphere. I could place the test mass at a radial distance, r, between the inner and outer spheres where the force due to gravity of the two spheres would cancel. A surface integral at the closed surface at radial distance, r, will find no mass for the enclosed inner mass. This means that some of the information contained within a closed surface is not reflected in the flux through that surface.
Gravity does effect time. Since the above is a statics problem and therefore time independent, this effect on time should not be an issue.
Jeff