In today’s interconnected world, ensuring the security of computer networks is crucial and massively important. One of the major threats faced by network administrators is the occurrence of Denial-of-Service (DoS) and Distributed Denial-of-Service (DDoS) attacks. Detecting and containing these attacks in a timely manner is crucial to maintaining network availability and preventing service disruptions. This paper presents a comprehensive approach for detecting DoS and DDoS attacks using the eXtreme Gradient Boosting (XGBoost) algorithm on a powerful and recent dataset. We focus on a subset of a dataset containing only normal DoS and DDoS attacks, and we perform extensive preprocessing and feature selection techniques. Our approach incorporates variable encoding, elimination of columns with only one value, normalization, besides correlation-based feature selection. Finally, we apply the XGBoost algorithm to classify the attacks. The experimental results demonstrate promising and insightful findings in terms of various performance metrics.