We consider the problem of multiuser opportunistic fair transmission scheduling (OFS) in power-controlled CDMA systems. OFS is an important technique in wireless networks to achieve fair and efficient resource allocation. Power control is an effective resource management technique in CDMA systems. Given a certain user subset and the channel states, the optimal power control scheme can be calculated. The multiuser OFS problem then refers to the optimal user subset selection at each interval to maximize instantaneous system throughput subject to some fairness constraint. We propose discrete stochastic approximation algorithms to adaptively select user subsets with the maximal system throughput. We also consider the time-varying channel scenarios where the algorithm can track the time-varying optimal user subset. We present simulation results to show the performance of the proposed algorithms in terms of the throughput optimization, the fast convergence, the excellent time-varying tracking capability, and the fairness.