Felix Schuster

VC3: Trustworthy Data Analytics in the Cloud Using SGX

We present VC3, the first system that allows users to run distributed MapReduce computations in the cloud while keeping their code and data secret, and ensuring the correctness and completeness of their results. VC3 runs on unmodified Hadoop, but crucially keeps Hadoop, the operating system and the hyper visor out of the TCB, thus, confidentiality and integrity are preserved even if these large components are compromised. VC3 relies on SGX processors to isolate memory regions on individual computers, and to deploy new protocols that secure distributed MapReduce computations. VC3 optionally enforces region self-integrity invariants for all MapReduce code running within isolated regions, to prevent attacks due to unsafe memory reads and writes. Experimental results on common benchmarks show that VC3 performs well compared with unprotected Hadoop: VC3s average runtime overhead is negligible for its base security guarantees, 4.5% with write integrity and 8% with read/write integrity.

Counterfeit Object-oriented Programming: On the Difficulty of Preventing Code Reuse Attacks in C++ Applications

Code reuse attacks such as return-oriented programming (ROP) have become prevalent techniques to exploit memory corruption vulnerabilities in software programs. A variety of corresponding defenses has been proposed, of which some have already been successfully bypassed -- and the arms race continues. In this paper, we perform a systematic assessment of recently proposed CFI solutions and other defenses against code reuse attacks in the context of C++. We demonstrate that many of these defenses that do not consider object-oriented C++ semantics precisely can be generically bypassed in practice. Our novel attack technique, denoted as counterfeit object-oriented programming (COOP), induces malicious program behavior by only invoking chains of existing C++ virtual functions in a program through corresponding existing call sites. COOP is Turing complete in realistic attack scenarios and we show its viability by developing sophisticated, real-world exploits for Internet Explorer 10 on Windows and Fire fox 36 on Linux. Moreover, we show that even recently proposed defenses (CPS, T-VIP, vfGuard, and VTint) that specifically target C++ are vulnerable to COOP. We observe that constructing defenses resilient to COOP that do not require access to source code seems to be challenging. We believe that our investigation and results are helpful contributions to the design and implementation of future defenses against control flow hijacking attacks.

Scriptless attacks: stealing the pie without touching the sill

Due to their high practical impact, Cross-Site Scripting (XSS) attacks have attracted a lot of attention from the security community members. In the same way, a plethora of more or less effective defense techniques have been proposed, addressing the causes and effects of XSS vulnerabilities. NoScript, and disabling scripting code in non-browser applications such as e-mail clients or instant messengers. As a result, an adversary often can no longer inject or even execute arbitrary scripting code in several real-life scenarios. In this paper, we examine the attack surface that remains after XSS and similar scripting attacks are supposedly mitigated by preventing an attacker from executing JavaScript code. We address the question of whether an attacker really needs JavaScript or similar functionality to perform attacks aiming for information theft. The surprising result is that an attacker can also abuse Cascading Style Sheets (CSS) in combination with other Web techniques like plain HTML, inactive SVG images or font files. Through several case studies, we introduce the so called scriptless attacks and demonstrate that an adversary might not need to execute code to preserve his ability to extract sensitive information from well protected websites. More precisely, we show that an attacker can use seemingly benign features to build side channel attacks that measure and exfiltrate almost arbitrary data displayed on a given website. We conclude this paper with a discussion of potential mitigation techniques against this class of attacks. In addition, we have implemented a browser patch that enables a website to make a vital determination as to being loaded in a detached view or pop-up window. This approach proves useful for prevention of certain types of attacks we here discuss.

It's a TRaP: Table Randomization and Protection against Function-Reuse Attacks

Code-reuse attacks continue to evolve and remain a severe threat to modern software. Recent research has proposed a variety of defenses with differing security, efficiency, and practicality characteristics. Whereas the majority of these solutions focus on specific code-reuse attack variants such as return-oriented programming (ROP), other attack variants that reuse whole functions, such as the classic return-into-libc, have received much less attention. Mitigating function-level code reuse is highly challenging because one needs to distinguish a legitimate call to a function from an illegitimate one. In fact, the recent counterfeit object-oriented programming (COOP) attack demonstrated that the majority of code-reuse defenses can be bypassed by reusing dynamically bound functions, i.e., functions that are accessed through global offset tables and virtual function tables, respectively. In this paper, we first significantly improve and simplify the COOP attack. Based on a strong adversarial model, we then present the design and implementation of a comprehensive code-reuse defense which is resilient against reuse of dynamically-bound functions. In particular, we introduce two novel defense techniques: (i) a practical technique to randomize the layout of tables containing code pointers resilient to memory disclosure and (ii) booby trap insertion to mitigate the threat of brute-force attacks iterating over the randomized tables. Booby traps serve the dual purpose of preventing fault-analysis side channels and ensuring that each table has sufficiently many possible permutations. Our detailed evaluation demonstrates that our approach is secure, effective, and practical. We prevent realistic, COOP-style attacks against the Chromium web browser and report an average overhead of 1.1% on the SPEC CPU2006 benchmarks.

Evaluating the Effectiveness of Current Anti-ROP Defenses

Recently, many defenses against the offensive technique of return-oriented programming (ROP) have been developed. Prominently among them are kBouncer, ROPecker, and ROPGuard which all target legacy binary software while requiring no or only minimal binary code rewriting.

Towards reducing the attack surface of software backdoors

Backdoors in software systems probably exist since the very first access control mechanisms were implemented and they are a well-known security problem. Despite a wave of public discoveries of such backdoors over the last few years, this threat has only rarely been tackled so far. In this paper, we present an approach to reduce the attack surface for this kind of attacks and we strive for an automated identification and elimination of backdoors in binary applications. We limit our focus on the examination of server applications within a client-server model. At the core, we apply variations of the delta debugging technique and introduce several novel heuristics for the identification of those regions in binary application that backdoors are typically installed in (i.e., authentication and command processing functions). We demonstrate the practical feasibility of our approach on several real-world backdoors found in modified versions of the popular software tools ProFTPD and OpenSSH. Furthermore, we evaluate our implementation not only on common instruction set architectures such as x86-64, but also on commercial off-the-shelf embedded devices powered by a MIPS32 processor.

Scriptless attacks: Stealing more pie without touching the sill

Due to their high practical impact, Cross-Site Scripting (XSS) attacks have attracted a lot of attention from the members of security community worldwide. In the same way, a plethora of more or less effective defense techniques have been proposed, addressing both causes and effects of XSS vulnerabilities. As a result, an adversary often can no longer inject or even execute arbitrary scripting code in several real-life scenarios. In this article, we examine an attack surface that remains after XSS and similar scripting attacks are supposedly mitigated by preventing an attacker from executing JavaScript code. We address the question of whether an attacker really needs to execute JavaScript or similar functionality to perform attacks aiming for information theft. The surprising result is that an attacker can abuse Cascading Style Sheets (CSS) in combination with other Web techniques like plain HTML, inactive SVG images, or font files. Having employed several case studies, we discuss so called scriptless attacks and demonstrate that an adversary might not need to execute code to preserve his ability to extract sensitive information from well-protected websites. More precisely, we show that an attacker can use seemingly benign features to build side-channel attacks that measure and exfiltrate almost arbitrary data displayed on a given webpage. We conclude this article with a discussion of potential mitigation techniques against this class of attacks. In addition, we have implemented a browser patch that enables a website to make a vital determination as to being loaded in a detached view or a pop-up window. This approach proves useful for prevention of certain types of attacks we here discuss.

Preventing backdoors in server applications with a separated software architecture

We often rely on system components implemented by potentially untrusted parties. This implies the risk of backdoors, i.e., hidden mechanisms that elevate the privileges of an unauthenticated adversary or execute other malicious actions on certain triggers. Hardware backdoors have received some attention lately and we address in this paper the risk of software backdoors. We present a design approach for server applications that can --- under certain assumptions --- protect against software backdoors aiming at privilege escalation. We have implemented a proof-of-concept FTP server to demonstrate the practical feasibility of our approach.