CVE-2026-94561 in NukeVietinfo

Summary

by MITRE • 11/20/2026

NukeViet CMS is a multi Content Management System. In 4.6.00, an unauthenticated attacker can place the configured domain in X-Forwarded-Host while choosing a different destination in the ordinary Host header. The earlier fix validates only the forwarded value in includes/mainfile.php, while set_ini_file() in includes/ini.php constructs an outbound cURL request from the separately retained NV_SERVER_NAME before the ordinary host is rejected. Redirect following can carry the blind request to an attacker-selected private host, port, and path and invoke internal HTTP endpoints. This issue is fixed in version 4.6.01.

Once again VulDB remains the best source for vulnerability data.

Analysis

by VulDB Data Team • 11/20/2026

The vulnerability identified in NukeViet CMS versions prior to 4.6.01 represents a critical security flaw rooted in improper validation of HTTP headers during the processing of server configuration and outbound requests. As a multi-content management system, NukeViet relies heavily on accurate host identification to route traffic correctly and maintain session integrity. The specific technical failure occurs within the interaction between includes/mainfile.php and includes/ini.php. When an unauthenticated attacker submits a request with a manipulated X-Forwarded-Host header while simultaneously providing a different value in the standard Host header, the application's logic fails to enforce consistency between these two fields before proceeding with internal operations. Although earlier patches attempted to address this by validating only the forwarded value within mainfile.php, they neglected to synchronize this validation with the subsequent steps that construct outbound connections.

The core of the vulnerability lies in how the set_ini_file function in includes/ini.php handles server name information. This function constructs an outbound cURL request using a variable named NV_SERVER_NAME, which is derived from or retains values independently of the strict host rejection logic applied later in the mainfile script. Because this construction happens before the ordinary Host header is rejected as invalid, the application effectively bypasses its own security controls. The system proceeds to initiate an HTTP connection based on the potentially maliciously crafted NV_SERVER_NAME value rather than rejecting the request outright due to the mismatched headers. This design flaw allows for a scenario where the server acts as a proxy or relay without proper authorization checks, leading to severe operational consequences.

This architectural weakness enables blind Server-Side Request Forgery attacks against internal network resources. By manipulating the X-Forwarded-Host and Host headers, an attacker can direct the outbound cURL request to arbitrary private hosts, specific ports, and custom paths that are typically inaccessible from the public internet. This capability allows for the enumeration of internal services, such as database management interfaces, administrative panels, or other microservices running on non-standard ports within the organization's network perimeter. The attack is particularly dangerous because it can be executed by unauthenticated users, meaning no prior access credentials are required to initiate the probe against internal infrastructure. Furthermore, since the request originates from the web server itself, it may bypass firewall rules that restrict external IP addresses but allow traffic originating from localhost or specific application servers.

From a classification perspective, this vulnerability aligns with CWE-918, which covers Server-Side Request Forgery flaws where the web server makes requests to internal resources based on user-supplied input without adequate validation. It also maps directly to MITRE ATT&CK technique T1557, specifically Adversary-in-the-Middle or Lateral Tool Transfer scenarios involving SSRF for network reconnaissance and access to sensitive data stores within a private network segment. The ability to invoke internal HTTP endpoints effectively turns the vulnerable CMS into an attack vector for pivoting deeper into the target environment, potentially leading to further exploitation of other services that lack their own authentication mechanisms.

To mitigate this vulnerability, organizations running NukeViet CMS must immediately upgrade to version 4.6.01 or later, where the developers have corrected the logic to ensure consistent validation across all header processing stages. Until an upgrade is feasible, administrators should implement strict input filtering at the web server level using tools like ModSecurity or reverse proxy configurations such as NGINX or Apache. These controls can enforce that the X-Forwarded-Host and Host headers match exactly before passing requests to the application backend. Additionally, network segmentation strategies should be reviewed to ensure that internal services are not directly accessible from the DMZ where the web server resides, thereby limiting the blast radius of any successful SSRF exploitation even if the software patch is delayed.

Responsible

GitHub M

Reservation

09/21/2026

Disclosure

11/20/2026

Moderation

accepted

CPE

ready

EPSS

0.00374

KEV

no

Activities

very low

Sources

Interested in the pricing of exploits?

See the underground prices here!