when i give a zero strain at the first step

when i give a [0,0,0] strain to the Yade, the Yade gives back a stress that is not zero, and then give a zero strain to Yade, and Yade will give back a different stress.
the question is when i set different young of the balls, i think the stress is no doubt different, however, when i give a strain like[1e-7,0,0],[0,1e-7,0],[0,0,1e-7] at the beginning of steps, the stress Yade get is different.
part of the code is as shown below:

.................................................................................................

O.load('/tmp/cccc.gz')
O.dt=1e-8

# see the example of peri3dcontroller, 1.5 is used to speed up the computation
EnlargeFactor=1.5
EnlargeFactor=1.0
O.engines=[
 ForceResetter(),
 InsertionSortCollider([Bo1_Sphere_Aabb(aabbEnlargeFactor=EnlargeFactor,label='bo1s')]),
 InteractionLoop(
  [Ig2_Sphere_Sphere_ScGeom(interactionDetectionFactor=EnlargeFactor,label='ig2ss')],
  [Ip2_CpmMat_CpmMat_CpmPhys()],[Law2_ScGeom_CpmPhys_Cpm()]),
 NewtonIntegrator(),

O.cell.velGrad=utils.Matrix3(0,0,0,0,0,0,0,0,0)
O.step()
bo1s.aabbEnlargeFactor=ig2ss.interactionDetectionFactor=1.

......................................................................................................................................................................

  for b in O.bodies:
   b.material.young = 1.0001*17e10 #change it to get different stress

   ag = [0]*3
   ag[0] = arr[0] - arrcc[0] #arr means current step stain and arrcc means last step strain
   ag[1] = arr[1] - arrcc[1]
    ag[2] = arr[2] - arrcc[2]

   ns=100

   dstrain = utils.Matrix3(ag[0],ag[2],0,0,ag[1],0,0,0,0) # the goal strain xx xy xz, yx yy yz, zx zy zz
   O.cell.velGrad=dstrain/(ns*O.dt)

   O.run(ns,True)

   stressValue=utils.getStress()
   stressxx=stressValue[0,0] # the return stress xx
   stressyy=stressValue[1,1] # the return stress yy
   stressxy=(stressValue[0,1]+stressValue[1,0])/2 # the return stress xy = (xy + yx)/2