Private browsing has been a popular privacy feature built into all mainstream browsers since 2005. However, despite the prevalent use, the security of this feature has received little attention from the research community. To the best of our knowledge, no study has existed that systematically evaluates the security of private browsing across all major browsers and from all angles: not only examining the memory, but also the underlying database structure on the disk and the web traffic. In this paper, we present an up-to-date and comprehensive analysis of private browsing across the four most popular web browsers: IE, Firefox, Chrome and Safari. We report that all browsers under study suffer from a variety of vulnerabilities, many of which have not been reported or known before. The problems are generally caused by the following factors: a lax control of permission to allow extensions to run in the private mode with unrestricted privilege; inconsistent implementations of the underlying SQLite database between the private and usual modes; the neglect of the cross-mode interference when the two modes are run in parallel; a lack of attention to side-channel timing attacks, etc. All of the attacks have been experimentally verified with countermeasures proposed.

No abstract

Private browsing has been a popular privacy feature built into all mainstream browsers since 2005. However, despite the prevalent use, the security of this feature has received little attention from the research community. To the best of our knowledge, no study has existed that systematically evaluates the security of private browsing across all major browsers and from all angles: not only examining the memory, but also the underlying database structure on the disk and the web traffic. In this paper, we present an up-to-date and comprehensive analysis of private browsing across the four most popular web browsers: IE, Firefox, Chrome and Safari. We report that all browsers under study suffer from a variety of vulnerabilities, many of which have not been reported or known before. The problems are generally caused by the following factors: a lax control of permission to allow extensions to run in the private mode with unrestricted privilege; inconsistent implementations of the underlying SQLite database between the private and usual modes; the neglect of the cross-mode interference when the two modes are run in parallel; a lack of attention to side-channel timing attacks, etc. All of the attacks have been experimentally verified with countermeasures proposed.

Single-Board Computers (SBC) refer to pocket-sized computers built on a single circuit board. A number of studies have explored the use of these highly popular devices in a variety of domains, including military, agriculture, healthcare, and more. However, no attempt was made to signify possible security risks that misuse of these devices may bring to organizations. In this study, we perform a series of experiments to validate the possibility of using SBCs as an espionage gadget. We show how an attacker can turn a Raspberry Pi device to an attacking gadget and benefit from short-term physical access to attach the gadget to the network in order to access unauthorized data or perform other malicious activities. We then provide experimental results of placing such tools in two realworld networks. Given the small size of SBCs, traditional physical security measures deployed in organizations may not be sufficient to detect and restrict the entrance of SBCs to their premises. Therefore, we reiterate possible directions for network administrators to deploy defensive mechanisms for detecting and preventing such attacks.

No abstract