In this technical note, an adaptive failure compensation problem has been studied for a class of nonlinear uncertain systems subject to stochastic actuator failures and unknown parameters. The stochastic functions related to Markovian variables have been introduced to denote the failure scaling factors for each actuators which is much more practical and challenging. Firstly, by taking into account of the Markovian variables existing in the system, some preliminary knowledges have been established. Then, by employing backstepping strategy, an adaptive failure compensation control scheme has been proposed, which ensures the boundedness in probability of all the closed-loop signals in the presence of stochastic actuator failures. A simulation example is presented to show the effectiveness of the proposed scheme.