Microsoft fixes 89 CVEs on penultimate Patch Tuesday of 2024

Microsoft has issued fixes addressing a total of 89 new Common Vulnerabilities and Exposures (CVEs) – 92 including third-party disclosures – to mark the penultimate Patch Tuesday of 2024, including four critical issues and a number of flaws that could be considered zero-days.

Of these issues, one meets the full traditional definition of a full zero-day, a vulnerability that is both public and known to be exploited. This is CVE-2024-43451, a spoofing vulnerability in New Technology LAN Manager (NTLM) Hash.

NTLM is a set of security protocols used to authenticate users’ identities. It dates back years and has been largely supplanted by vastly more secure protocols – Microsoft has not recommended its use in over a decade, but since it was used in Internet Explorer, it remains supported to some extent and continues to cause problems, not least because at this stage, it is incredibly insecure.

In this instance, successful exploitation of this issue could lead to “total loss of confidentiality”, according to Microsoft, as it discloses a user’s NTLMv2 hash to an attacker who could then use it to authenticate as the user – if the victim can be tricked into minimal interaction with a malicious file, which could include merely selecting or clicking it, not even opening it. This may make it considerably more dangerous than its comparatively low severity score may indicate.

Mike Walters, president and co-founder of Action1, explained: “This issue arises from the mechanism where NTLM authentication credentials, specifically NTLMv2 hashes, are improperly exposed via a maliciously crafted file.

“The root cause of this vulnerability lies in improperly handling file interactions within systems, potentially allowing attackers to extract NTLMv2 hashes without requiring complete file execution,” he told Computer Weekly in emailed commentary.

All supported versions of Microsoft Windows are vulnerable to this issue, said Walters, especially if they use applications reliant on MSHTML and EdgeHTML platforms, while risk is further increased across different system environments thanks to the involvement of other scripting engines.

Walters said the main concern with CVE-2024-43451 is the disclosure of NTLMv2 hashes that can be used to authenticate as the user and leveraged in pass-the-hash attacks, enabling further lateral movement for a canny threat actor.

“This vulnerability is particularly effective in phishing scenarios, where users might be deceived into interacting with malicious files. Once NTLM hashes are obtained, attackers can combine them with other network vulnerabilities to extend their access and compromise additional systems,” he said.

“Organisations that heavily use Windows in environments with substantial network file sharing or legacy applications dependent on Internet Explorer and related platforms face heightened risk. Those lacking robust user training and monitoring systems to detect unusual file interactions may be more susceptible to exploitation.”

Also on the list is CVE-2024-49309, which is exploited but not yet public. This is an elevation of privilege (EoP) vulnerability in Windows Task Scheduler.

This stems from an issue where authentication tokens or credentials are improperly managed and could allow a low-privileged attacker to gain deeper access if they can execute a malicious application designed for the purpose. It impacts multiple versions of Windows that incorporate Task Scheduler as part of their routine task automation processes, and it is thought that environments with shared or multiple-user setups may be particularly vulnerable to it.

“This vulnerability serves as a potential entry point for attackers who have already accessed a system with low privilege. Once privileges are escalated, these attackers can utilise this foothold for further lateral movement within a network or to exploit other vulnerabilities that necessitate higher access levels,” said Walters.

“The nature of this vulnerability is especially concerning in corporate settings where individual users possess specific task automation privileges that could be exploited to gain unauthorised access.”