Update State Tampering: A Novel Adversary Post-compromise Technique on Cyber Threats

Abstract

With modern cyber threats, attackers should gain persistency in target systems to achieve attack objectives. Once an attacker’s zero-day vulnerabilities on target systems are patched, the attacker may lose control over the system. However, systems remain vulnerable when an attacker manipulates the component resources on a Windows system. We found methods to generate invisible vulnerabilities on a victim’s system. Our findings are as follows: first, we found ways to replace a component to an old vulnerable version while maintaining the current update records; second, we found that the Windows system does not recognize the replaced components. We define the first issue as a package-component mismatch and the second issue as a blind spot issue on the Windows update management. They have been identified on all version of Vista and later, including desktop platforms and server platforms. Based on our findings, we reveal an Update State Tampering technique that can generate invisible security holes on target systems. We also offer corresponding countermeasures to detect and correct package-component mismatches. In this paper, we introduce the problems with the current Windows update management mechanism, the Update State Tampering technique from the attacker’s point of view, and an Update State Check scheme that detects and recovers the package-component mismatches. We stress that our proposed Update State Check scheme should be deployed immediately in order to mitigate large-scale exploitation of the proposed technique.