We would like to bring your attention to the following critical root privilege escalation vulnerability that we have recently become aware of. This vulnerability affects all major Linux distributions and is considered to be easy to exploit. Proof of concept exploits are publicly available for this vulnerability.
The vulnerability is tracked as CVE-2022-0847 (also known as Dirty Pipe) and allows for privilege escalation. A non-privileged user is able to inject and overwrite data in read-only files, including SUID processes that run as root. This vulnerability has been reported to affect Linux Kernel 5.8 (released August 2020) and later.
The vulnerability was officially fixed in Linux Kernel versions 5.16.11, 5.15.25 and 5.10.102, however, many distributions may also back-port this fix to earlier kernel versions.
Am I Affected?
Running one of the following commands will help identify the current Kernel
-
uname -r
-
cat /proc/version
-
hostnamectl | grep Kernel
Example – uname -r
$ uname -r 5.13.19-3-generic
This host is likely to be vulnerable as the kernel version is 5.13.9 which is newer than 5.8 and older than any of the fixed versions 5.16.11, 5.15.25 and 5.10.102 (assuming now back ports have been applied).
Understanding Linux Version Numbers
Example Linux kernel version is 5.13.19-3-generic, where:
-
5 : Kernel version
-
13 : Major revision
-
19 : Minor revision
-
3 : Patch level or number
-
generic : Linux distro/kernel specific additional info
Proof of Concept
Multiple proof of concepts now exists within the public domain which we will continue to monitor and update when new information becomes available.
We are aware of the following proof of concept exploit code; however we would not recommend running these without performing your own due diligence:
This first proof of concept video demonstrates a malicious attacker with local access escalating to root.
In this second video CovertSwarm demonstrates a malicious action with local access removing roots password requirement.
Remediation
At this time, many distributions have yet to releases patches against this issue, this is likely to change within the near future. For latest updates please review your respective vendors update page.
References
When Your IDE Becomes An Insider: Testing Agentic Dev Tools Against Indirect Prompt Injection
Agentic development tools don’t need to bypass your firewall. They’re already inside. And if an attacker can control what they read, they can control what they…
What Moltbook reveals about AI agent security
The Moltbook launch exposed a critical gap: organizations deploying AI agents faster than they can secure them. Research shows 22% of enterprises currently host unauthorised AI…
Cloud security vulnerabilities: the unpatchable risks hiding in your SaaS stack
The cloud gives businesses scalability, agility, and built-in resilience. But it’s created a dangerous illusion: that moving workloads to SaaS platforms and managed services automatically makes…