/

CVE-2026-51740

Severity:

CRITICAL RISK

(9.8)

Broken Access Control (CWE-284) in TOTOLINK T6 killProcess

CVE-2026-51740

Deferred

Vulnerability type

Category:

Broken Access Control

CWE:

CWE-284

Impact:

Unauthorized Service Termination and Denial of Service

Severity:

CRITICAL RISK

(9.8)

CVSS version:

3.1

Catch this CVE while code is being written

Detect vulnerable code directly in pull requests, so CVEs are fixed before merge.

Summary

A critical access-control flaw in the killProcess function of the TOTOLINK T6 router allows unauthenticated attackers to terminate critical services. The issue becomes risky when the device exposes the affected CGI endpoint and fails to require authorization for crafted POST requests.

Affected context

The vulnerability occurs in TOTOLINK T6 devices using the affected firmware component that handles requests through /cgi-bin/cstecgi.cgi. It may be present when the killProcess function is reachable without authentication and does not adequately restrict which services a caller can terminate.

Exploit preconditions

  • Attack surface: HTTP POST request handling through /cgi-bin/cstecgi.cgi
  • Required access: No authentication may be required when the vulnerable function is exposed
  • Trigger point: A crafted request invoking the killProcess functionality
  • Why exploitation is possible: The function performs a sensitive process-management operation without enforcing appropriate authorization checks

Why this happens

The underlying issue is an authorization failure in a privileged process-control function. The request handler permits an unauthenticated caller to invoke an operation intended to be restricted to trusted administrative users, mapping to CWE-284: Improper Access Control. The insecure pattern is exposing a sensitive management action without consistently validating identity, privilege, and permitted target scope before execution.

Potential impact

  • Critical router services may be terminated
  • Network availability may be disrupted
  • Administrative and monitoring functions could become unavailable
  • Dependent devices or connected users may experience service interruption
  • Repeated abuse can cause persistent denial-of-service conditions

General mitigation guidance

Mitigation generally focuses on enforcing authorization at the server-side handler for every process-management operation.

Common defensive patterns include:

  • Requiring authenticated, authorized administrative sessions before invoking service-control functions
  • Applying deny-by-default policy to CGI management endpoints
  • Restricting operations to an explicit allowlist of safe service targets
  • Separating privileged process control from unauthenticated request handling
  • Running management components with the minimum privileges necessary
  • Limiting administrative interfaces to trusted network segments where practical
  • Adding request validation, audit logging, and rate controls for sensitive operations

How teams detect issues like this

Organizations can look for:

  • Static-analysis findings where CGI handlers invoke process termination or service-control APIs without nearby authorization checks
  • Code-review signals such as missing session validation, role checks, or target allowlists around administrative actions
  • Dynamic tests that send unauthenticated requests to management endpoints and verify that sensitive operations are denied
  • Web-server and router logs showing repeated POST requests to /cgi-bin/cstecgi.cgi or unexpected service termination events
  • Policy enforcement requiring authentication and authorization controls for all exposed device-management functions
  • Automated security testing and code-review tools that identify unauthenticated access to privileged operations

Sources

SourceLink
github.comgithub.com/DarkBoulder/CVE-Vendor-Coordination/blob/main/TOTOLINK/README.md
github.comgithub.com/ShengWu00/CVE-Vendor-Coordination/blob/main/TOTOLINK/README.md
totolink.nettotolink.net/
totolink.nettotolink.net/home/menu/detail/menu_listtpl/download/id/190/ids/36.html

Frequently asked

questions

It may be remotely exploitable when the affected CGI endpoint is reachable over a network and the device accepts the request without authentication. Exposure depends on router access settings and network placement.

The described issue allows unauthenticated access, so an account may not be required when the vulnerable endpoint is reachable.

Enforce server-side authentication and authorization before allowing any process termination or service-management operation, using a deny-by-default policy and an explicit target allowlist.

Treat every management endpoint as untrusted until identity and privilege are verified, review privileged handlers for missing access checks, restrict administrative interfaces, and test unauthenticated and low-privilege request paths.

Find out what's already
exploitable in your codebase.