The Internet has brought significant changes to the world, but it has also brought numerous cyberattacks. One of the most dangerous attacks is the (DDoS) Distributed Denial of Services attack, which can halt the normal functionality of services in all computing environments. The DDoS attacks are classified by the position of pivotal attacks that undermine the network's functionality. These attacks have become sophisticated and continue to grow rapidly, making it challenging to describe and address them. These malicious attacks are caused by a network of computers infected with special malware, known as a "botnet," which bombards a server with traffic until it collapses under the strain. The Mirai botnet is one such example, which is largely made up of internet-connected devices such as digital cameras and the DVR players. In our proposed system, we propose machine learning grounded attack discovery from traffic data. This module contains three corridors first, preprocessing; second, detecting the traffic as licit or not using machine learning (CATBOOST and Random Forest); and third, classifying the attack using machine learning (CATBOOST and Random Forest). Eventually, estimate and compare the performances of CATBOOST and Random Forest for traffic classification and attack classification. In the preprocessing step of traffic data, the data is cleaned and filtered to remove any noise or irrelevant information. The performance of the model is evaluated based on metrics such as accuracy, precision, recall, and F1-score. Appropriate data preprocessing can improve the accuracy of the machine learning model. The evaluation of the performance of the model includes metrics such as accuracy, precision, recall, and F1-score.