non zero momentum flux in uniformly initialized domain
 

Hello,
I am confused with non-zero momentum flux when the domain is uniformly initialized with identical values in each cell. I am using in house CFD code based on FVM, Euler equations, explicit Euler time integration and AUSM and HLLC solvers. I have initialized the domain with the same values everywhere (initially, there is no flow -> M=0), and, as you may imagine nothing changes in following time steps. But, the flux is not zero actually (but only the momentum flux). The momentum flux is equal to the pressure term in AUSM solver (only this term is non zero), and equals the average pressure from neigbouring cells. The flux is the same in each volume, and compensates, so that the values in each cell remains constant. But is it normal? Moreover, HLLC gives the same results. I thought, that when the states in two neighbouring cells are the same, there will be no flux at all. Could anyone explain this to me, please?