Threat Advisories

Critical NGINX HTTP/2 Remote Code Execution Vulnerability (CVE-2026-123007)

Written by Integrity360 | Nov 19, 2026, 10:20:19 AM

Researchers have disclosed a critical remote code execution (RCE) vulnerability affecting NGINX HTTP/2 implementations, tracked as CVE-2026-123007 and disclosed as part of a set of flaws dubbed "nginx2root." The vulnerability has existed within NGINX codebases since September 2015 and allows unauthenticated attackers to execute code remotely on servers with HTTP/2 enabled.

Security researchers from Wiz, Rapid7, and Xint successfully demonstrated exploitation, with Xint reportedly achieving RCE against cloud-hosted systems exposed via public IP addresses across AWS, Azure, and GCP environments.

While code execution initially occurs in the context of an NGINX worker process, Xint reports that further privilege escalation is possible through a separate Linux kernel vulnerability, potentially leading to full system compromise. This kernel issue is unpatched and has not yet been confirmed in a vendor advisory.

F5 released fixes on 18 November 2026. The same updates address four further critical NGINX vulnerabilities, so patching should not be limited to systems running HTTP/2.

Key Findings

  • Critical unauthenticated RCE vulnerability in NGINX HTTP/2 functionality.
  • Tracked as CVE-2026-123007 with a CVSS v4.0 score of 9.2 (8.1 under CVSS v3.1).
  • Vulnerable code has existed since NGINX version 1.9.5 (September 2015).
  • Public proof-of-concept exploit code has been published on GitHub by Rapid7.
  • Successful exploitation demonstrated against cloud-hosted environments including AWS, Microsoft Azure, and Google Cloud Platform (GCP).
  • Exploitation does not require authentication when HTTP/2 is enabled.
  • F5 states that code execution requires ASLR to be disabled or bypassed. Researchers report bypassing it.
  • Four further critical vulnerabilities are fixed in the same releases: CVE-2026-123633 (HTTP/2), CVE-2026-121810 (HTTP/3), CVE-2026-123634 (DNS resolver) and CVE-2026-113425 (mail proxy).
  • No confirmed in-the-wild exploitation has been reported by F5, Rapid7, or Wiz, and the vulnerability is not in CISA's Known Exploited Vulnerabilities (KEV) catalogue.

Technical Overview

The vulnerability resides within NGINX's handling of HTTP/2 traffic and can permit remote attackers to trigger memory corruption conditions leading to arbitrary code execution. According to researchers, exploitation can occur by interacting with a reachable HTTP/2 service exposed by the target server.

Although initial code execution occurs under the security context of an NGINX worker process, Xint highlighted scenarios where attackers could leverage an additional, currently unpatched Linux kernel vulnerability to achieve root-level access, including from within Docker-based deployments. This element is researcher-reported and not vendor-confirmed.

The issue is particularly concerning because:

  • HTTP/2 is widely enabled across modern web infrastructure.
  • NGINX is one of the most widely deployed web servers globally.
  • Exploitation can be performed remotely without credentials.
  • Public exploit code lowers the barrier to entry for threat actors.

Affected Versions

NGINX Open Source

Version Range

Status

Fixed In

1.31.0 - 1.31.8 (mainline)

Vulnerable

1.31.9

1.9.5 - 1.30.6

Vulnerable

1.30.7

NGINX Plus

Version Range

Status

Fixed In

37.1.0 - 37.2.0.1

Vulnerable

37.3.0.1

37.0.0 - 37.0.7.1

Vulnerable

37.0.8.1

R34 - R36

Vulnerable

No fix available. Upgrade to a fixed 37.x release

 

F5 also lists NGINX Instance Manager, F5 WAF for NGINX, NGINX App Protect WAF, F5 DoS for NGINX, NGINX App Protect DoS, NGINX Gateway Fabric and NGINX Ingress Controller as affected. Fixed versions for these products are listed in F5 advisory K000163926. Linux distributions and third-party products that bundle NGINX publish their own fixed versions.

Threat Assessment

Likelihood: High

Several factors increase the likelihood of exploitation:

  • Public exploit availability.
  • Internet-facing attack surface.
  • No authentication requirement.
  • Broad deployment of vulnerable software.
  • Media coverage likely to accelerate scanning and exploitation activity.

Potential Impact: Critical

Successful exploitation may result in:

  • Remote code execution.
  • Web server compromise.
  • Container escape opportunities through chained vulnerabilities.
  • Lateral movement within cloud or enterprise environments.
  • Data theft and service disruption.
  • Deployment of ransomware or web shells.

Indicators of Potential Exploitation

No vendor-published indicators of compromise are available. CERT-EU and the reporting researchers recommend monitoring for:

  • NGINX worker processes spawning unexpected child processes.
  • Abnormal worker crashes or core dumps.
  • Long-lived HTTP/2 sessions generating excessive inbound traffic.

Recommended Mitigations

Immediate Actions

  1. Identify exposed NGINX systems. Inventory all internet-facing NGINX and NGINX Plus instances and verify whether HTTP/2 is enabled.
  2. Apply vendor patches. Upgrade to NGINX Open Source 1.30.7 or 1.31.9 or later, or NGINX Plus 37.0.8.1 or 37.3.0.1 or later.
  3. Restart NGINX after upgrade. A configuration reload alone is insufficient. Perform a full restart or live binary upgrade as recommended by the vendor.

Temporary Mitigation

F5 lists no mitigation other than upgrading. Where patching is not immediately possible, CERT-EU and the reporting researchers recommend the following:

  • Disable HTTP/2 support on affected NGINX instances. This addresses CVE-2026-123007 and CVE-2026-123633 only, and will break services that rely on HTTP/2, such as gRPC.
  • Disable HTTP/3 if not required.
  • Verify HTTP/2 removal using the command below, and confirm the server no longer negotiates the h2 protocol.

openssl s_client -alpn h2 -connect <host>:443

References

If you are worried about any of the threats outlined in this bulletin or need help in determining what steps you should take to protect yourself from the most material threats facing your organisation, please contact your account manager, or alternatively get in touch to find out how you can protect your organisation.