Abstract:
Security is one of the key elements of any financial model. An OSI model
perspective we have seen an upward progression inside the layer of insecurity. We
started from physical layer attacks and gradually moved to Layer 2 and so on and so
forth. At the moment we are at a point where most of the security issues lie at the
application layer.
Most of the application layer attacks we witness are fell inside the category of “Input
Validation Attacks” i.e. not properly validating the users input before its parsed. To
minimize the attack surface at application layer, vendors have come up with
application layer firewalls also known as WAF (Web Application Firewall).
A web application firewall uses two basic approaches for detecting malicious input, a
positive model (Allow Known Good) or otherwise known as a White list or a
negative model (Reject Known Bad), due to the simplicity of design, almost all of the
“WAF” vendors rely upon “Blacklisting” Approach to filter the input. However,
there is a fundamental problem inside the blacklist approach is that different
browsers parse different inputs in a different way, which makes it easier for us to
bypass blacklist based Web application firewalls
The objective of this project is the Implementation of Web Application Firewall over
Network for Maximum Protection it will be fulfilling the needs over the security
area. This document has the detailed description of project; this project actually
works for the server side protection.
VIII
Our solution is a light weight, and contains a user friendly interface which allows
even non-technical people to configure and modify the rules. The implementation
would be based upon a strong blacklist with minimal number of false positives. The
second objective of this project is to create a tool which allows vendors to test the
effectiveness of their Web Application Firewall as well as for webmasters to analyze
the current solutions they have implemented
