In this article
Nginx2Root - Critical Vulnerability in Nginx Leads to Privilege Escalation
Introduction
Several vulnerabilities have been chained together to create a working exploit that, if executed, can lead to a remote unauthenticated user gaining root privilege on any system running Nginx. Nginx is the internet swiss army tool that most organisations big and small rely on, leading to it running a third of the world-wide-web. This BridgeLynk security advisory provides an overview of the vulnerabilities and aims to provide practical tips that can be used to resolve the vulnerability and avoid exploitation by threat agents and actors.
Vulnerability Summary
Vulnerability Summary
TLP: Green
CVSS Score: 9.2
CVSSv3.1: AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:H/A:H
MITRE ATT&CK: Tactic: TA0002-Execution, TA004- Privilege Escalation | Technique: T1674-Input Injection
CVE: CVE-2026-123007
Severity: Critical
Title: Nginx2Root - Critical Vulnerability in Nginx Leads to Privilege Escalation
Detail Analysis
Security researchers at Xint, Wiz, AND Rapid7 have found and reported multiple vulnerabilities that could be chained together to create a highly effective attack chain through a vulnerable instance of Nginx. These vulnerabilities include CVE-2026-123007, CVE-2026-123634, CVE-2026-123633, CVE-2026-121810, and CVE-2026-113425. The security weakness is primarily made possible due to a critical flaw in the implementation of HTTP/2 within Nginx. The impact is specifically a flow control heap overflow flaw that allows an unauthenticated user or agent to corrupt and leak the memory address of an Nginx worker through a segmentation fault. Additionally, the leaked memory address “memcpy” can then be overwritten with encoded shellcode which grants the user remote code execution capabilities within the privilege of the Nginx worker service account.
The diagram below that was inspired by the original advisory from Xint, illustrates the process of chaining all fixed vulnerabilities together to achieve code execution.

Figure -1 Nginx2Root exploit-chain visualiser. Original inspiration credit: nginx2root.com.
Once remote command execution capability is achieved, malicious actors can then leverage a local kernel privilege escalation flaw to elevate privileges to the root or NT SYSTEM user which are the highest level of privilege within Linux, Unix and Windows based operating systems. See original disclosure report for more in-depth analysis.
According to Rapid7, CVE-2026-123007 requires the victim to download approximately 128GB of data using HTTP/2 within a short period of time. This important detail infers that the attack is limited by network bandwidth and nginx network connection timeout.
Affected versions
Product | Versions |
|---|---|
NGINX Plus | Affected - 37.1.0 < 37.3.0.1 Affected - 37.0.0 < 37.0.8.1 Affected -R33 < * |
NGINX Open Source | Affected 1.31.0 < 1.31.9 (semver) Affected 1.30.0 < 1.30.7 (semver) Affected 1.9.5 < * (semver) |
The fixed version is Nginx 1.31.9 and Nginx OS 1.30.7 - > https://nginx.org/download/nginx-1.31.9.tar.gz
Mitigation Strategies & Recommendations:
Due to Nginx wide usage for internet facing assets, cybercriminals are already carrying out reconnaissance activities including scanning the web using tools such as Shodan.io and censys.io to detect Nginx banners. Organisations are thereby advised to take immediate action to resolve the critical flaw starting from internet-facing servers that accept HTTP/2 requests.

Image 1 — censys.io search query showing alleged NGINX products active on the internet. The screenshot alone does not establish that these systems are vulnerable.
As at the writing of this advisory, there is no known exploitation in the wide, but it is expected that cybercriminals and malicious agents will likely begin offensive operations aimed at detecting and exploiting vulnerable servers reachable on the internet.
Our recommendation is for the available patched version of Nginx to be installed immediately. Where the installation of a new version of Nginx is not feasible, system administrator and security personnel can elect to disable HTTP/2.
BridgeLynk experts provided highly specialist attack informed defence cybersecurity and AI security capability aimed at exposing hidden weaknesses and supporting your team to resolve them collaboratively. Contact us to learn more about how BridgeLynk can partner with you to improve organisational security maturity and risk reduction. Book a free 60 minutes expert threat assessment consultation and work with our expert to achieve better business resilience.
References:
Under attack or want to get ahead of one?
Speak to a BridgeLynk expert
Get a free threat assessment consultation and find out where your organisation is exposed before an attacker does.
Speak To An Expert
Our Articles

Nginx2Root - Critical Vulnerability in Nginx Leads to Privilege Escalation
BridgeLynk advisory on the reported Nginx2Root vulnerability chain, affected versions and recommended mitigations.

Practical Ransomware Response Guide for SMBs
Ransomware attacks can disrupt your business, lock you out of critical systems, and put sensitive data at risk. Learn how SMBs can respond to an attack, preserve evidence, recover safely, and prepare for future incidents with the right security measures and support

Critical WordPress "wp2shell" Vulnerabilities Exploited to Deploy Webshells
Critical vulnerabilities collectively known as "wp2shell" are being actively exploited to compromise vulnerable WordPress websites, enabling attackers to achieve unauthenticated remote code execution (RCE) and deploy persistent webshells. By chaining flaws in the WordPress Core REST API and SQL injection mechanisms, threat actors can take full control of affected sites without requiring plugins, themes, or valid credentials. This article examines how the wp2shell exploit chain works, the WordPress versions at risk, real-world exploitation activity, the security implications for website owners, and the essential mitigation steps administrators should take to prevent complete server compromise.

Top 10 AI Chatbot Breach Survival Guide for CISOs & CIOs
Discover the top 10 AI chatbot security risks every CISO and CIO must prepare for. This comprehensive breach survival guide explores emerging AI threats, real-world attack techniques, incident response strategies, governance best practices, and proactive security measures to help organisations protect sensitive data, maintain compliance, and build resilient AI-powered environments.

Critical Remote Code Execution (RCE) in AI Automation Tool n8n — "Ni8mare"
A critical unauthenticated RCE vulnerability (CVE-2026-21858, CVSS 10.0) in the popular AI workflow tool n8n allows attackers to read sensitive server files and escalate to full remote code execution. Here is what you need to know.

Critical vulnerability affecting Cisco Catalyst SD-WAN - CVE-2026-20127
A maximum-severity vulnerability (CVSS 10.0) in Cisco Catalyst SD-WAN controllers is being actively exploited in the wild. Attackers are bypassing authentication, escalating to root, and establishing persistent access across enterprise network infrastructure.

Notepad++ Update Feature Hijacked by Chinese State Hackers for Months
Between June and December 2025, Chinese state-linked hackers from the Lotus Blossom group hijacked the Notepad++ update mechanism to deliver a custom backdoor to millions of users. Here is a full breakdown of what happened and what organisations must do now.

GitHub Confirms Breach of 4,000 Repositories Through Malicious VSCode Extension
GitHub has confirmed a significant security breach involving nearly 4,000 internal repositories after a malicious Visual Studio Code extension was installed on an employee's device. The trojanised Nx Console extension harvested credentials including GitHub tokens, AWS keys, and 1Password vault data.