In this paper, classical as well as Bayesian estimation of stress strength reliability (η) of Dagum distribution under progressive type-II censored sample is done. Maximum likelihood estimators (MLEs) of η are also obtained along with asymptotic, bootstrap-p (boot-p) and bootstrap-t (boot-t) confidence intervals. Bayes estimators of η along with highest posterior density (HPD) credible intervals based on informative and non-informative priors are also obtained. Symmetric as well as asymmetric loss functions are considered for the Bayesian estimation. Monte Carlo simulation study is carried out to check the performance of estimators. Real life data sets are considered for illustration purpose.