June 27, 2017 By Shane Schick 2 min read

Security researchers discovered a haunting technique that they’re calling a GhostHook attack, which uses a feature in Intel central processing units (CPUs) to take over 64-bit Windows systems. Windows has traditionally been safe from most cybercriminals trying to install rootkits, but the GhostHook attack can bypass PatchGuard, which was specifically developed to protect its operating system at the kernel level.

Experts at CyberArk described how Intel Processor Trace (IPT), which captures information about software on a PC to deal with debugging and other issues, could open a way to hook rootkits onto Windows. This would be very difficult to detect, hence the “ghost” moniker.

Hooking is not always used for malicious purposes, Dark Reading noted, and fraudsters would need to already have malware present on the system to exploit a rootkit. The GhostHook attack is particularly scary, however, because it’s happening on the kernel level. That means it’s impervious to many security products, including personal firewalls and even intrusion prevention systems (IPS).

GhostHook Attack Degrades Defenses

PatchGuard isn’t effective here. As Bleeping Computer explained, CyberArk researchers were able to allocate extremely small buffers to packets of code in IPT. Eventually, the CPU needs additional space and attempts to deal with the code by opening a performance monitoring interrupt (PMI) handler.

The GhostHook attack is extremely clever because PatchGuard wasn’t designed to look at what happens within PMI handlers. That’s also why it’s possible to plant a rootkit as the system is being patched.

Although there haven’t been any public accounts of cybercriminal activity using the GhostHook attack, a CyberArk researcher told iTWire that Windows admins should be on guard. If used in combination with disk-wiping malware, for instance, cybercriminals could do a lot of damage before they’ve ever discovered by authorities.

The Industry Responds

Microsoft has been informed about the GhostHook attack. According to CyberArk, however, the software giant said the flaw likely would not be addressed, except possibly in a future version of Windows.

It’s also possible that Intel will make it more difficult to pursue the technique in its next generation of IPT technology. Since attackers would need to have malware on the PC already, this is not yet a full-blown emergency — more like a phantom menace.

More from

CVE-2023-20078 technical analysis: Identifying and triggering a command injection vulnerability in Cisco IP phones

7 min read - CVE-2023-20078 catalogs an unauthenticated command injection vulnerability in the web-based management interface of Cisco 6800, 7800, and 8800 Series IP Phones with Multiplatform Firmware installed; however, limited technical analysis is publicly available. This article presents my findings while researching this vulnerability. In the end, the reader should be equipped with the information necessary to understand and trigger this vulnerability.Vulnerability detailsThe following Cisco Security Advisory (Cisco IP Phone 6800, 7800, and 8800 Series Web UI Vulnerabilities - Cisco) details CVE-2023-20078 and…

X-Force data reveals top spam trends, campaigns and senior superlatives in 2023

10 min read - The 2024 IBM X-Force Threat Intelligence Index revealed attackers continued to pivot to evade detection to deliver their malware in 2023. The good news? Security improvements, such as Microsoft blocking macro execution by default starting in 2022 and OneNote embedded files with potentially dangerous extensions by mid-2023, have changed the threat landscape for the better. Improved endpoint detection also likely forced attackers to shift away from other techniques prominent in 2022, such as using disk image files (e.g. ISO) and…

The compelling need for cloud-native data protection

4 min read - Cloud environments were frequent targets for cyber attackers in 2023. Eighty-two percent of breaches that involved data stored in the cloud were in public, private or multi-cloud environments. Attackers gained the most access to multi-cloud environments, with 39% of breaches spanning multi-cloud environments because of the more complicated security issues. The cost of these cloud breaches totaled $4.75 million, higher than the average cost of $4.45 million for all data breaches.The reason for this high cost is not only the…

Topic updates

Get email updates and stay ahead of the latest threats to the security landscape, thought leadership and research.
Subscribe today