Overview
A vulnerability in the user of hard-coded Platform Keys (PK) within the UEFI framework, known as PKfail, has been discovered. This flaw allows attackers to bypass critical UEFI security mechanisms like Secure Boot, compromising the trust between the platform owner and firmware and enabling manipulation of sensitive system settings.
Description
The UEFI standard establishes trust relationships using Public Key Infrastructure (PKI) between the platform owner, the platform firmware, and the operating system. Central to this process is the Platform Key (PK), which is designed to secure the connection between the platform owner and the platform firmware.

The platform owner enrolls the public half of the key (PKpub) into the platform firmware. The platform owner can later use the private half of the key (PKpriv) to change platform ownership or to enroll a Key Exchange Key. For UEFI, the recommended Platform Key format is RSA-2048.
(Section 7.2.1 of the UEFI 2.3.1 Errata C standard)

The PKFail vulnerability highlights a critical flaw in the UEFI ecosystem. While the Platform Key is expected to originate from the Original Equipment Manufacturer (OEM) using a secure hardware security module (HSM), in practice, much of the UEFI software and drivers are developed by a complex network of supply-chain partners and Independent BIOS Vendors (IBVs). These components are often shared across multiple OEMs. In some cases, temporary test software keys, or “softkeys,” which are hard-coded for ease of build and testing, inadvertently make their way into production firmware.
These softkeys, intended solely for compatibility testing and performance evaluation, are supposed to be untrusted and restricted in their usage. The current UEFI’s key verification process is limited – it only checks against the keys in the local database, with no verification against the root Certificate Authority (CA) or special validation of extended attributes. Although keys cannot be self-signed, the lack of stringent verification allows these untrusted keys to be mistakenly included in production firmware.
Recent audits have uncovered that many OEM devices shipped with hard-coded, untrusted keys in their production UEFI firmware. Despite these keys often having attributes like “DO NOT TRUST,” there is no programmatic safeguard or other validations (say attribute-based) to prevent their inclusion in final products. The compromise or leak of these private keys could have bad consequences, allowing attackers to sign malicious modules that execute with high privileges during the boot process, even if Secure Boot is enabled. This undermines the very purpose of signed software verification, leaving systems vulnerable to untrusted and malicious modules.
Compounding the issue, UEFI firmware is largely invisible to most Endpoint Detection and Response (EDR) software, making it difficult to audit and detect the use of compromised keys. Moreover, many UEFI implementations lack Remote Measurement or Auditing capabilities that could dynamically check the integrity of the key database via network resources.
Impact
An attacker with access to an undesired-yet-trusted test Platform Key’s private portion can exploit it to sign malicious UEFI software, enabling the execution of code with the highest privileges during the early boot phases of a UEFI Secure Boot-protected system. A successful attack could lead to the following impacts:

Invalidation or bypass of UEFI security features like SecureBoot.
Installation of persistent software that cannot be easily detected or erased, that can also persist across reboots and potentially surviving OS reinstalls.
Creation of backdoors and back communications channels to exfiltrate sensitive data.
Interruption of system execution leading to device damage or permanent shutdown.

Solution

Update UEFI Firmware: Ensure you install the latest stable version of UEFI firmware provided by your PC vendor or the reseller of your computing environment. Refer to the Vendor Information section below for specific resources and updates from vendors addressing these vulnerabilities.
Use Researcher Tools for Impact Assessment: Utilize tools and information provided by Binarly to assess the impact of untrusted Platform Keys on your systems. These resources can help you conduct a thorough analysis of affected systems.
Leverage Automatic Firmware Updates: If your operating system supports automatic or managed firmware updates (e.g., Linux Vendor Firmware Service, LVFS), regularly check for updates using fwupdmgr get-updates and apply them with fwupdmgr update or use Windows OEM supported mechanisms as appropriate. Keeping your firmware up to date is crucial in mitigating the risks associated with PKfail.

Acknowledgements
Thanks to Binarly for disclosing this vulnerability. This document was written by Vijay Sarvepalli.

Overview
Multiple hosted, outbound SMTP servers are vulnerable to email impersonation. This allows authenticated users and certain trusted networks to send emails containing spoofed sender information. Two vulnerabilities were identified that reduce the authentication and verification of the sender, provided by the combination of Sender Policy Framework (SPF) and Domain Key Identified Mail (DKIM). Domain-based Message Authentication, Reporting, and Conformance (DMARC) builds on SPF and DKIM, adding linkage to the author (FROM:) domain name, published policies for recipient handling of authentication failures, and reporting from receivers to senders to improve and monitor protection of the domain from fraudulent email (DMARC.org). An authenticated remote attacker can spoof the identity of a sender when sending emails using a hosted service provider.
Description
As identified in RFC 5321 #7.1, the SMTP protocol is inherently insecure and susceptible to spoofing the sender identity that is present in the various parts of the SMTP transaction. Various facilities, such as SPF and DKIM, continued to evolve to address these issues. SPF records identify the IP networks that are allowed to send email on behalf of a domain. Receiving servers can check SPF records to verify that incoming messages that appear to be from an organization are sent by permitted (allowed) networks. DKIM goes further in email security by providing a digital signature that verifies specific portions of the SMTP-relayed message, allowing to digitally assert specific information that is part of a message such as the FROM: address, subject, and date fields. While SPF verifies the network source of an email transaction, DKIM looks into an email message to prevent message tampering. DMARC is an email authentication, policy, and reporting protocol that builds on the widely deployed SPF and DKIM protocols. As a useful combination of these two capabilities, DMARC helps both email senders and receivers work together to better secure emails, protecting users and brands from costly abuse.
A set of vulnerabilities were discovered by researchers in the practical usage of these capabilities exposing the potential abuse of sender trust in email communications. Many of the hosted, email services provide hosting for multiple domains and use a wide range of network resources to deliver emails from their domain addresses. The hosting service providers typically provide a way to authenticate before allowing emails to be sent on behalf of the sender. However, due to the nature of their shared hosting, many of them do not verify the authenticated sender against their allowed domain identities. Hosting providers who have published SPF records, and, in some cases, also add DKIM signatures, do not sufficiently verify the trust relationship of authenticated user against the allowed domains. This allows an authenticated attacker to spoof an identity in the email Message Header to send emails as anyone in the hosted domains of the hosting provider, while authenticated as a user of a different domain name.
Any remote email receiving services may incorrectly identify the sender’s identity as it passes the cursory check of DMARC policy adherence. The DMARC policy is thus circumvented, allowing spoofed messages to be seen as an attested and a valid message.
CVE-2024-7208
A vulnerability in multi-tenant hosting allows an authenticated sender to spoof the identity of a shared, hosted domain, thus bypass security measures provided by DMARC (or SPF or DKIM) policies.
CVE-2024-7209
A vulnerability exists in the use of shared SPF records in multi-tenant hosting providers, allowing attackers to use network authorization to be abused to spoof the email identify of the sender.
Impact
An authenticated attacker using network or SMTP authentication can spoof the identity of a shared hosting facility, circumventing any DMARC policy and sender verification provided by a domain name owner.
Solution
Hosting providers
Domain hosting providers that provide email relay should verify the identity of an authenticated sender against authorized domain identities. The email service providers should use reliable ways to verify that the network sender identity (MAIL FROM) and the Message Header (FROM:) are the same or related. As much as SMTP software does not verify the Message Header with the network sender, identity mail filter software, such as (Milter) Milterfrom, may provide ways to enforce such requirements.
Domain owners
Domain owners should use strict measures to ensure their domain, DNS-based DMARC policy (DKIM and SPF) protects their sender identity and their users and brands from abuse caused by spoofing. If a domain is expected to provide high assurance of identity, the domain owner should use their own DKIM facility, independent of the hosting provider, to reduce the risk of spoofing attacks.
Email Senders
Email senders that require high fidelity of their identity can use facilities such as S/MIME and PGP, as suggested in RFC 5321 #7.1.
Acknowledgements
Thanks to the reporters, Caleb Sargent and Hao Wang, for raising awareness of these vulnerabilities. This document was written by Dr. Elke Drennan, Vijay Sarvepalli, and Timur Snoke.

Update 10/8/2023 Data carrier indicated that there was a failure of one of their core routers. They have replaced it […]