For a one-dimensional Brownian motion starting from the origin, the cumulative distribution of the occupation time V staying above the origin obeys the celebrated arcsine law. In this work, we show how the law is modified for a resetting Brownian motion, where the Brownian is reset to the position x
r
at random times but with a constant rate r. When x
r
is exactly equal to zero, we derive the exact expression of the probability distribution P
r
(V∣0, t) of V during time t, and the moments of V as functions of r and t. P
r
(V∣0, t) is always symmetric with respect to V = t/2 for arbitrary value of r, but the probability density of V at V = t/2 increases with the increase of r. Interestingly, P
r
(V∣0, t) at V = t/2 changes from a minimum to a local maximum at a critical value R
* ≈ 0.742 338, where R = rt denotes the average number of resetting during time t. Moreover, we consider the case when x
r
is a random variable and is distributed by a function g(x
r
), where g(x
r
) is assumed to be symmetric with respect to zero and possesses its maximum at zero. We derive the general expressions of the moments of V when the variance of x
r
is low. The mean value of V is always equal to t/2, but the fluctuation in x
r
leads to an increase in the second and third moments of V. Our results provide a quantitative understanding of how stochastic resetting destroys the persistence of Brownian motion.