This paper deals with the problem of output feedback tracking control for a class of Euler-Lagrange (EL) systems under denial-of-service (DoS) attack and external disturbance. A new adaptive observation scheme is devised for state reconstruction during the attack, and the adaptive observation function included in it can achieve the effect of suppressing interference. With the help of estimated values, an adaptive neural network security control protocol is constructed under the backstepping control framework to ensure the bounded output tracking of EL system under DoS attack and disturbance. Finally, the efficacy of the presented observation and control scheme for nonlinear EL system is validated through numerical simulation.