Preventing injection attacks through automated randomization of keywords
Author(s)
Willenson, Daniel M
DownloadFull printable version (2.361Mb)
Other Contributors
Massachusetts Institute of Technology. Dept. of Electrical Engineering and Computer Science.
Advisor
Jeff Perkins and Martin Rinard.
Terms of use
Metadata
Show full item recordAbstract
SQL injection attacks are a major security issue for database-backed web applications, yet the most common approaches to prevention require a great deal of programmer effort and attention. Even one unchecked vulnerability can lead to the compromise of an entire application and its data. We present a fully automated system for securing applications against SQL injection which can be applied at runtime. Our system mutates SQL keywords in the program's string constants as they are loaded, and instruments the program's database accesses so that we can verify that all keywords in the final query string have been properly mutated, before passing it to the database. We instrument other method calls within the program to ensure correct program operation, despite the fact that its string constants have been mutated. Additionally, we instrument places where the program generates user-visible output to ensure that randomized keyword mutations are never revealed to an attacker.
Description
Thesis (M. Eng. and S.B.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2012. Cataloged from PDF version of thesis. Includes bibliographical references (p. 47-48).
Date issued
2012Department
Massachusetts Institute of Technology. Department of Electrical Engineering and Computer SciencePublisher
Massachusetts Institute of Technology
Keywords
Electrical Engineering and Computer Science.