In webserver security, a DDoS request attack is a significant risk in the application layer. Harmul request is a fast developing issue. These attacks avoid most interruption anticipation frameworks by sending various HTTP calls. Since the vast majority of these attacks post quickly and thoughtfully, a quick interruption avoidance framework is attracting to identify and alleviate these attacks straightaway. We propose a robust security framework, named rahuldefencesystem, which rapidly recognizes and help application-layer Harmul request attacks. A manual human test performs to choose whether there is an attack or a human. At whatever point a request is coming from a blacklisted IP address, it won't permit getting to the worker. Norma or Greylisted IP addresses are real IP addresses. This improves the adequacy by keeping away from the opposite reckoning of harmful. The test results show that application can rapidly lessen damaging request while representing a restricted sway on ordinary clients. We have also included a file sharing and password mutation System for record security in the cloud and password safety. The client can be ready to store the record into shared cloud storage

Keywords : DDoS attack, AES encryption, CAPTCHA test (Completely Automated Public Turing test tell Computers and Humans Apart).