Security Hacks

1536 Articles

This Week In Security: PostHog, Project Zero Refresh, And Thanks For All The Fish

December 19, 2025 by 1 Comment

There’s something immensely satisfying about taking a series of low impact CVEs, and stringing them together into a full exploit. That’s the story we have from [Mehmet Ince] of Prodraft, who found a handful of issues in the default PostHog install instructions, and managed to turn it into a full RCE, though only accessible as a user with some configuration permissions.

As one might expect, it all starts with a Server Side Request Forgery (SSRF). That’s a flaw where sending traffic to a server can manipulate something on the server side to send a request somewhere else. The trick here is that a webhook worker can be primed to point at localhost by sending a request directly to a system API.

One of the systems that powers a PostHog install is the Clickhouse database server. This project had a problem in how it sanitized SQL requests, namely attempting to escape a single quote via a backslash symbol. In many SQL servers, a backslash would properly escape a single quote, but Clickhouse and other Postgresql servers don’t support that, and treat a backslash as a regular character. And with this, a read-only SQL API is vulnerable to SQL injection.

These vulnerabilities together just allow for injecting an SQL string to create and run a shell command from within the database, giving an RCE and remote shell. The vulnerabilities were reported through ZDI, and things were fixed earlier this year. Continue reading “This Week In Security: PostHog, Project Zero Refresh, And Thanks For All The Fish” →

This Week In Security: Hornet, Gogs, And Blinkenlights

December 12, 2025 by 12 Comments

Microsoft has published a patch-set for the Linux kernel, proposing the Hornet Linux Security Module (LSM). If you haven’t been keeping up with the kernel contributor scoreboard, Microsoft is #11 at time of writing and that might surprise you. The reality is that Microsoft’s biggest source of revenue is their cloud offering, and Azure is over half Linux, so Microsoft really is incentivized to make Linux better.

The Hornet LSM is all about more secure eBPF programs, which requires another aside: What is eBPF? First implemented in the Berkeley Packet Filter, it’s a virtual machine in the kernel, that allows executing programs in kernel space. It was quickly realized that this ability to run a script in kernel space was useful for far more than just filtering packets, and the extended Berkeley Packet Filter was born. eBPF is now used for load balancing, system auditing, security and intrusion detection, and lots more.

This unique ability to load scripts from user space into kernel space has made eBPF useful for malware and spyware applications, too. There is already a signature scheme to restrict eBPF programs, but Hornet allows for stricter checks and auditing. The patch is considered a Request For Comments (RFC), and points out that this existing protection may be subject to Time Of Check / Time Of Use (TOCTOU) attacks. It remains to be seen whether Hornet passes muster and lands in the upstream kernel. Continue reading “This Week In Security: Hornet, Gogs, And Blinkenlights” →

This Week In Security: React, JSON Formatting, And The Return Of Shai Hulud

December 5, 2025 by 3 Comments

After a week away recovering from too much turkey and sweet potato casserole, we’re back for more security news! And if you need something to shake you out of that turkey-induced coma, React Server has a single request Remote Code Execution flaw in versions 19.0.1, 19.1.2, and 19.2.1.

The issue is insecure deserialization in the Flight protocol, as implemented right in React Server, and notably also used in Next.js. Those two organizations have both issued Security Advisories for CVSS 10.0 CVEs.

There are reports of a public Proof of Concept (PoC), but the repository that has been linked explicitly calls out that it is not a true PoC, but merely research into how the vulnerability might work. As far as I can tell, there is not yet a public PoC, but reputable researchers have been able to reverse engineer the problem. This implies that mass exploitation attempts are not far off, if they haven’t already started. Continue reading “This Week In Security: React, JSON Formatting, And The Return Of Shai Hulud” →

One-Way Data Extraction For Logging On Airgapped Systems

November 27, 2025 by 50 Comments

If you want to protect a system from being hacked, a great way to do that is with an airgap. This term specifically refers to keeping a system off any sort of network or external connection — there is literally air in between it and other systems. Of course, this can be limiting if you want to monitor or export logs from such systems. [Nelop Systems] decided to whip up a simple workaround for this issue, creating a bespoke one-way data extraction method.

The concept is demonstrated with a pair of Raspberry Pi computers. One is hooked up to critical industrial control systems, and is airgapped to protect it against outside intruders. It’s fitted with an optocoupler, with a UART hooked up to the LED side of the device. The other side of the optocoupler is hooked up to another Raspberry Pi, which is itself on a network and handles monitoring and logging duties.

This method creates a reliable one-way transmission method from the airgapped machine to the outside world, without allowing data to flow in the other direction. Indeed, there is no direct electrical connection at all, since the data is passing through the optocoupler, which provides isolation between the two computers. Security aficionados will argue that the machine is no longer really airgapped because there is some connection between it and the outside world. Regardless, it would be hard to gain any sort of access through the one-way optocoupler connection. If you can conceive of a way that would work, drop it down in the comments.

Optocouplers are very useful things; we’ve seen them used and abused for all sorts of different applications. If you’ve found some nifty use for these simple parts, be sure to drop us a line!

This Week In Security: Cloudflare Wasn’t DNS, BADAUDIO, And Not A Vuln

November 21, 2025 by 8 Comments

You may have noticed that large pieces of the Internet were down on Tuesday. It was a problem at Cloudflare, and for once, it wasn’t DNS. This time it was database management, combined with a safety limit that failed unsafe when exceeded.

Cloudflare’s blog post on the matter has the gritty details. It started with an update to how Cloudflare’s ClickHouse distributed database was responding to queries. A query of system columns was previously only returning data from the default database. As a part of related work, that system was changed so that this query now returned all the databases the given user had access to. In retrospect it seems obvious that this could cause problems, but it wasn’t predicted to cause problems. The result was that a database query to look up bot-management features returned the same features multiple times.

That featurelist is used to feed the Cloudflare bot classification system. That system uses some AI smarts, and runs in the core proxy system. There are actually two versions of the core proxy, and they behaved a bit differently when the featurelist exceeded the 200 item limit. When the older version failed, it classified all traffic as a bot. The real trouble was the newer Rust code. That version of the core proxy threw an error in response, leading to 5XX HTTP errors, and the Internet-wide fallout. Continue reading “This Week In Security: Cloudflare Wasn’t DNS, BADAUDIO, And Not A Vuln” →

Browser Fingerprinting And Why VPNs Won’t Make You Anonymous

November 19, 2025 by 89 Comments

Amidst the glossy marketing for VPN services, it can be tempting to believe that the moment you flick on the VPN connection you can browse the internet with full privacy. Unfortunately this is quite far from the truth, as interacting with internet services like websites leaves a significant fingerprint. In a study by [RTINGS.com] this  browser fingerprinting was investigated in detail, showing just how easy it is to uniquely identify a visitor across the 83 laptops used in the study.

As summarized in the related video (also embedded below), the start of the study involved the Am I Unique? website which provides you with an overview of your browser fingerprint. With over 4.5 million fingerprints in their database as of writing, even using Edge on Windows 10 marks you as unique, which is telling.

In the study multiple VPN services were used, each of which resulted in exactly the same fingerprint hash. This is based on properties retrieved from the browser, via JavaScript and other capabilities exposed by the browser, including WebGL and HTML5 Canvas.

Continue reading “Browser Fingerprinting And Why VPNs Won’t Make You Anonymous” →

This Week In Security: Landfall, Imunify AV, And Sudo Rust

November 14, 2025 by 7 Comments

Let’s talk about LANDFALL. That was an Android spyware campaign specifically targeted at Samsung devices. The discovery story is interesting, and possibly an important clue to understanding this particular bit of commercial malware. Earlier this year Apple’s iOS was patched for a flaw in the handling of DNG (Digital NeGative) images, and WhatsApp issued an advisory with a second iOS vulnerability, that together may have been used in attacks in the wild.

Researchers at Unit 42 went looking for real-world examples of this iOS threat campaign, and instead found DNG images that exploited a similar-yet-distinct vulnerability in a Samsung image handling library. These images had a zip file appended to the end of these malicious DNG files. The attack seems to be launched via WhatsApp messaging, just like the iOS attack. That .zip contains a pair of .so shared object files, that are loaded to manipulate the system’s SELinux protections and install the long term spyware payload.

The earliest known sample of this spyware dates to July of 2024, and Samsung patched the DNG handling vulnerability in April 2025. Apple patched the similar DNG problem in August of 2025. The timing and similarities do suggest that these two spyware campaigns may have been related. Unit 42 has a brief accounting of the known threat actors that could have been behind LANDFALL, and concludes that there just isn’t enough solid evidence to make a determination.

Not as Bad as it Looks

Watchtowr is back with a couple more of their unique vulnerability write-ups. The first is a real tease, as they found a way to leak a healthy chunk of memory from Citrix NetScaler machines. The catch is that the memory leak is a part of an error message, complaining that user authentication is disabled. This configuration is already not appropriate for deployment, and the memory leak wasn’t assigned a CVE.

There was a second issue in the NetScaler system, an open redirect in the login system. This is where an attacker can craft a malicious link that points to a trusted NetScaler machine, and if a user follows the link, the NetScaler will redirect the user to a location specified in the malicious link. It’s not a high severity vulnerability, but still got a CVE and a fix. Continue reading “This Week In Security: Landfall, Imunify AV, And Sudo Rust” →