Connect with us

Security

Thunderclap flaws impact how Windows, Mac, Linux handle Thunderbolt peripherals

Published

on


Logo: Markettos et al. // Composition: ZDNet

Windows, Mac, Linux, and FreeBSD systems are all impacted by a new vulnerability that was disclosed this week at the NDSS 2019 security conference.

The vulnerability –named Thunderclap– affects the way Thunderbolt-based peripherals are allowed to connect and interact with these operating systems, allowing a malicious device to steal data directly from the operating system’s memory, including highly sensitive information.

The research team behind this vulnerability says that “all Apple laptops and desktops produced since 2011 are vulnerable, with the exception of the 12-inch MacBook.”

Similarly, “many laptops, and some desktops, designed to run Windows or Linux produced since 2016 are also affected,” as long as they support Thunderbolt interfacing.

What is Thunderbolt?

Thunderbolt is the name of a hardware interface designed by Apple and Intel to allow the connection of external peripherals (keyboards, chargers, video projectors, network cards, etc.) to a computer.

These interfaces became wildly popular because they combined different technologies into one single cable, such as the ability to transmit DC power (for charging purposes), serial data (via PCI Express), and video output (via DisplayPort).

The technology was initially available for Apple devices but was later made available for all hardware vendors, becoming ubiquitous nowadays, especially thanks to the standard’s latest version, Thunderbolt 3.

But according to the research team, all Thunderbolt versions are affected by Thunderclap. This means Thunderbolt 1 and 2 (the interface versions that use a Mini DisplayPort [MDP] connector) and Thunderbolt 3 (the one that works via USB-C ports).

What is Thunderclap?

Thunderclap is a collection of flaws in the way the Thunderbolt hardware interface has been implemented on operating systems.

At the core of this vulnerability, researchers say they are exploiting an OS design issue where the operating system automatically puts faith in any newly connected peripheral, granting it access to all of its memory –a state known as Direct Memory Access (DMA).

Thunderclap flaws allow attackers to create malicious, but fully-working peripherals that when connected via a Thunderbolt-capable port can perform their normal operations, but also run malicious code in the operating system’s background without any restriction from the operating.

This makes the Thunderclap attack highly dangerous, as it can be easily hidden inside any peripheral.

The Thunderclap vulnerabilities are even capable of bypassing an OS security feature known as Input-Output Memory Management Units (IOMMUs) that hardware and OS makers have created in the early 2000s to counter malicious peripherals that abuse their access to the entire OS memory (in what’s known as a DMA attack).

The reason why Thunderclap vulnerabilities work against IOMMU is either because operating systems disable this feature by default, or, in cases the feature has been enabled by the user, the OS leaves user data in the same memory space where the malicious peripheral runs its exploit code, making IOMMU useless.

What’s being done about it?

Researchers from the University of Cambridge, Rice University, and SRI International discovered the Thunderclap issues back in 2016, and they’ve been working with hardware and OS versions for three years in complete silence to have them fixed.

However, despite the almost three-year warning, OS makers have been slow to react, with most of the Thunderclap attack variations described in a research paper published today still working. Here’s the current state of patches, according to researchers:

Windows – “Microsoft have enabled support for the IOMMU for Thunderbolt devices in Windows 10 version 1803, which shipped in 2018. Earlier hardware upgraded to 1803 requires a firmware update from the vendor. This brings them into line with the baseline for our work, however the more complex vulnerabilities we describe remain relevant.”

macOS – “In macOS 10.12.4 and later, Apple addressed the specific network card vulnerability we used to achieve a root shell. However the general scope of our work still applies; in particular that Thunderbolt devices have access to all network traffic and sometimes keystrokes and framebuffer data.”

Linux – “Recently, Intel have contributed patches to version 5.0 of the Linux kernel (shortly to be released) that enable the IOMMU for Thunderbolt and prevent the protection-bypass vulnerability that uses the ATS feature of PCI Express.”

FreeBSD – “The FreeBSD Project indicated that malicious peripheral devices are not currently within their threat model for security response. However, FreeBSD does not currently support Thunderbolt hotplugging.”

As the table below shows, most Thunderclap flaws are still unpatched.

Thunderclap flaws still working

Image: Markettos et al.

In the meantime, users are advised to disable Thunderbolt ports via BIOS/UEFI firmware settings and to avoid plugging in peripherals from untrusted sources.

Technical details about the Thunderclap flaws are available in a research paper entitled “Thunderclap: Exploring Vulnerabilities in Operating System IOMMU Protection via DMA from Untrustworthy Peripherals,” available for download in PDF format from here and here, with more details here.

The research team also released the “Thunderclap platform” on GitHub, which is a collection of ready-made proof-of-concept code to create malicious Thunderclap peripherals.

Extra details are also available on a dedicated website and in this blog post.

As a closing note, Thunderclap vulnerabilities can also be exploited by compromised PCI Express (PCIe) peripherals, such as plug-in cards or chips soldered to the motherboard, but these attacks require compromising the peripheral’s firmware, making the attack much harder to pull off than just plugging in a charger or video projector via a Thunderbolt interface.

Related cybersecurity news coverage:

Source link

Continue Reading
Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *

Security

Managing Vulnerabilities in a Cloud Native World

Published

on

This free 1-hour webinar from GigaOm Research brings together experts in Cloud Native Vulnerability Management, featuring analyst Iben Rodriguez and special guest from Palo Alto Networks, John Morello. The discussion will focus on optimizing cloud security posture and integration with enterprise tool sets.

We will review platforms delivering Security Posture Management and Workload Protection for Microservice based and Hybrid Cloud Workloads.

Registrants will learn how new customers can benefit from Prisma Cloud to better secure their complex multi-cloud environments. Existing customers will learn about new features they can take advantage of and how to optimize their limited resources.

Register now to join GigaOm and Palo Alto Networks for this free expert webinar.

The post Managing Vulnerabilities in a Cloud Native World appeared first on Gigaom.

Continue Reading

Security

Security Tools Help Bring Dev and Security Teams Together

Published

on

Software development teams are increasingly focused on identifying and mitigating any issues as quickly and completely as possible. This relates not only to software quality but also software security. Different organizations are at different levels when it comes to having their development teams and security teams working in concert, but the simple fact remains that there are far more developers out there than security engineers.

Those factors are leading organizations to consider security tooling and automation to proactively discover and resolve any software security issues throughout the development process. In the recent report, “GigaOm Radar for Developer Security Tools,” Shea Stewart examines a roundup of security tools aimed at software development teams.

Stewart identified three critical criteria to bear in mind when evaluating developer security tools. These include:

  • Vendors providing tools to improve application security can and should also enhance an organization’s overall security posture.
  • The prevailing “shift-left” mindset doesn’t necessarily mean the responsibility for reducing risk should shift to development, but instead focusing on security earlier in the process and continuing to do so throughout the development process will reduce risk and the need for extensive rework.
  • Security throughout the entire software development lifecycle (SDLC) is critical for any organization focused on reducing risk.

Figure 1. How Cybersecurity Applies Across Each Stage of the Software Development Lifecycle *Note: This report focuses only on the Developer Security Tooling area

Individual vendors have made varying levels of progress and innovation toward enhancing developer security. Following several acquisitions, Red Hat, Palo Alto Networks, and Rapid7 have all added tooling for developer security to their platforms. Stewart sees a couple of the smaller vendors like JFrog and Sonatype as continuing to innovate to remain ahead of the market.

Vendors delving into this category and moving deeper into “DevSecOps” all seem to be taking different approaches to their enhanced security tooling. While they are involving security in every aspect of the development process, some tend to be moving more quickly to match the pace of the SDLC. Others are trying to shore up existing platforms by adding functionality through acquisition. Both infrastructure and software developers are now sharing toolsets and processes, so these development security tools must account for the requirements of both groups.

While none of the 12 vendors evaluated in this report can provide comprehensive security throughout the entire SDLC, they all have their particular strengths and areas of focus. It is therefore incumbent upon the organization to fully and accurately assess its SDLC, involve the development and security teams, and match the unique requirements with the functionality provided by these tools. Even if it involves using more than one at different points throughout the process, focus on striking a balance between stringent security and simplifying the development process.

Read more: Key Criteria for Evaluating Developer Security Tools, and the Gigaom Radar for Developer Security Tool Companies.

The post Security Tools Help Bring Dev and Security Teams Together appeared first on Gigaom.

Continue Reading

Security

Key Criteria for Evaluating User and Entity Behavior Analytics (UEBA)

Published

on

Cybersecurity is a multidisciplinary practice that not only grows in complexity annually but evolves nearly as quickly. A survey of the security landscape today would reveal concerns ranging from the classic compromised servers to the relatively new DevSecOps practices aimed at securing the rapid deployment of new code and infrastructure. However, some things remain constant no matter how much change is introduced. While technology evolves and complexity varies, there is almost always a human component in
risks presented to an organization.

User Behavior Analysis (UBA) was designed to analyze the actions of users in an organization and attempt to identify normal and abnormal behaviors. From this analysis, malicious or risky behaviors can be detected. UBA solutions identify events that are not detectable using other methods because, unlike classic security tools (an IDS or SIEM for example), UBA does not simply pattern match or apply rule sets to data to identify security events. Instead, it looks for any and all deviations from baseline user activity.

As technology advanced and evolved, and the scope of what is connected to the network grew, the need to analyze entities other than users emerged. In response, entity analysis has been added to UBA to create UEBA or User and Entity Behavior Analysis. The strategy remains the same, but the scope of analysis has expanded to include entities involving things like daemons, processes, infrastructure, and so on.

How to Read this Report

This GigaOm report is one of a series of documents that helps IT organizations assess competing solutions in the context of well-defined features and criteria. For a fuller understanding consider reviewing the following reports:

Key Criteria report: A detailed market sector analysis that assesses the impact that key product features and criteria have on top-line solution characteristics—such as scalability, performance, and TCO—that drive purchase decisions.

GigaOm Radar report: A forward-looking analysis that plots the relative value and progression of vendor solutions along multiple axes based on strategy and execution. The Radar report includes a breakdown of each vendor’s offering in the sector.

Solution Profile: An in-depth vendor analysis that builds on the framework developed in the Key Criteria and Radar reports to assess a company’s engagement within a technology sector. This analysis includes forward-looking guidance around both strategy and product.

The post Key Criteria for Evaluating User and Entity Behavior Analytics (UEBA) appeared first on Gigaom.

Continue Reading

Trending