Know Thy Neighbor: Crypto Library Detection in Cloud

Abstract

Software updates and security patches have become a standard method to fix known and recently discovered security vulnerabilities in deployed software. In server applications, outdated cryptographic libraries allow adversaries to exploit weaknesses and launch attacks with significant security results. The proposed technique exploits leakages at the hardware level to first, determine if a specific cryptographic library is running inside (or not) a co-located virtual machine (VM) and second to discover the IP of the co-located target. Nevertheless, one of the main concerns that is slowing the widespread usage of such Infrastructure as a Service (IaaS) technologies are potential security vulnerabilities and privacy risks of cloud computing. Usually, CSPs employ virtual machines (VMs), allowing multiple tenants to share the same computing hardware. While this resource sharing maximizes utilization and hence drastically reduces cost, ensuring isolation of potentially sensitive data between VMs instantiated by different and untrusted tenants can be a challenge. Indeed, the main security principle in the design and implementation of virtual machine managers (VMMs) has been that of process and data isolation achieved through sandboxing. Although logical isolation ensures security at the software level, a malicious tenant might still extract private information due to leakage coming from side channels such as shared hardware resources. In short, hardware sharing create an opening for various side channel attacks developed for non-virtualized environments. These powerful attacks are capable of extracting sensitive information, e.g. passwords and private keys, by profiling the victim process. ignoring leakages of information through subtle side-channels shared by the processes running on the same physical hardware. In non-virtualized environments, a number of effective side-channel attacks were proposed that succeeded in extracting sensitive data by targeting the software layer only.

Flush+Flush: A Fast and Stealthy Cache Attack

“…In contrast to Prime+Probe, Flush+Reload exploits the availability of shared memory and especially shared libraries between the attacker and the victim program. Applications of Flush+Reload have been shown to be reliable and powerful, mainly to attack cryptographic algorithms [12,17,18,48].…”

Defeating Hardware Prefetchers in Flush+Reload Side-Channel Attack

“…Gruss et al [9] have shown that a variant of Flush+Reload without the clflush instruction is possible without a significant loss in accuracy. Applications of Flush+Reload have been shown to be reliable and powerful, mainly to attack cryptographic algorithms [12,13,14,15]. Flush+Reload is also being used to compromise user privacy.…”

Cache Misses and the Recovery of the Full AES 256 Key

“…The attack was first introduced in [33], and was later extended to target the LLC to retrieve cryptographic keys, TLS protocol session messages or keyboard keystrokes across VMs [2,34,35]. It has also demonstrated its power against AES T-table based implementations [36], RSA implementations [2], or ECDSA [37] among others, and also detecting cryptographic libraries [38]. Further, Zhang et al [39] showed that it is applicable in several commercial PaaS clouds, where it is possible to achieve co-residency with a victim [13].…”