A_optimize get "primal infeasible" or "dual infeasible" results

[hardsphere]

from diffnet import A_optimize
from cvxopt import matrix
import numpy as np

nsofar = np.zeros((4,4))
sij = np.array([[6.241888, 2.533248, 2.733051, 2.733051],
[2.533248, 6.599607, 1.708107, 1.708107],
[2.733051, 1.708107, 6.63203, 1.001 ],
[2.733051, 1.708107, 1.001, 6.63203 ]])

delta_N = 152000000
nij = A_optimize(matrix(sij), delta_N, matrix(nsofar),delta=None,method='conelp')

pcost dcost gap pres dres k/t
0: 0.0000e+00 -0.0000e+00 3e+08 1e+06 6e+00 1e+00
1: 1.5803e+06 2.2247e+06 4e+07 2e+05 8e-01 6e+05
Certificate of primal infeasibility found.