Security

Patches

Fortinet patches VPN app flaw that could give rogue users, malware a privilege boost

Plus a bonus hard-coded local API key


A now-patched, high-severity bug in Fortinet's FortiClient VPN application potentially allows a low-privilege rogue user or malware on a vulnerable Windows system to gain higher privileges from another user, execute code and possibly take over the box, and delete log files.

The bug is tracked as CVE-2024-47574, and it earned a 7.8 out of 10 CVSS severity rating. It affects FortiClientWindows version 7.4.0, 7.2.4 through 7.2.0, 7.0.12 through 7.0.0, and 6.4.10 through 6.4.0. Fortinet patched the hole on Tuesday, so if you haven't already, upgrade to a fixed release.

Pentera Labs' bug hunter Nir Chako found and reported the flaw to Fortinet, plus a second security oversight that allows someone or something nefarious on a system running the VPN client to alter SYSTEM-level registry keys that would otherwise be off limits.

According to Chako, this latter flaw has been assigned CVE-2024-50564, though the vendor has not yet issued a security alert about it. However, it has also been fixed in the latest version, FortiClient 7.4.1. 

"They said it will be published in the next advisory update," Chako told The Register, adding that advisory is slated for release on the December 10 Patch Tuesday. "From a security perspective, after testing version 7.4.1, we were able to validate that the patch prevented us from executing the techniques."

Neither flaw appears to have been exploited in the wild. Fortinet did not immediately respond to The Register's inquiries. We will update this story if and when we hear back from the vendor.

As Chako explains in this detailed technical write-up, exploiting CVE-2024-47574 involves using Windows named pipes with the FortiClient software to ultimately plant a script so that when a higher-privileged user next uses the VPN, that script is run with their privileges, and thus code execution is achieved with unauthorized powers. This privilege-escalation technique involves a step know as process hollowing.

This could also be abused to delete log files, and make a user connect to an attacker-controlled server. Plus, when combined with the second vulnerability, CVE-2024-50564, a miscreant would be "able to edit SYSTEM level registry values within the HKLM registry hive," Chako said.

Exploiting CVE-2024-50564 involves using a hard-coded local API encryption key that components of Fortinet's software use to exchange commands and data between themselves; it's not a VPN secret. ®

Send us news
Post a comment

Critical security hole in Apache Struts under exploit

You applied the patch that could stop possible RCE attacks last week, right?

Ransomware scum blow holes in Cleo software patches, Cl0p (sort of) claims responsibility

But can you really take crims at their word?

BlackBerry offloads Cylance's endpoint security products to Arctic Wolf

Fresh attempt to mix the perfect cocktail of IoT and Infosec

US reportedly mulls TP-Link router ban over national security risk

It could end up like Huawei -Trump's gonna get ya, get ya, get ya

Microsoft won't let customers opt out of passkey push

Enrollment invitations will continue until security improves

How Androxgh0st rose from Mozi's ashes to become 'most prevalent malware'

Botnet's operators 'driven by similar interests as that of the Chinese state'

Australia moves to drop some cryptography by 2030 – before quantum carves it up

The likes of SHA-256, RSA, ECDSA and ECDH won't be welcome in just five years

Suspected LockBit dev, facing US extradition, 'did it for the money'

Dual Russian-Israeli national arrested in August

Don't fall for a mail asking for rapid Docusign action – it may be an Azure account hijack phish

Recent campaign targeted 20,000 folk across UK and Europe with this tactic, Unit 42 warns

Blocking Chinese spies from intercepting calls? There ought to be a law

Sen. Wyden blasts FCC's 'failure' amid Salt Typhoon hacks

Boffins trick AI model into giving up its secrets

All it took to make an Google Edge TPU give up model hyperparameters was specific hardware, a novel attack technique … and several days

Phishers cast wide net with spoofed Google Calendar invites

Not that you needed another reason to enable the 'known senders' setting