Summary
TOTOLINK T6 contains an incorrect access-control check in the meshInfoKick function, allowing unauthenticated attackers to manipulate mesh state through crafted MQTT messages. Successful abuse may remove stale mesh information or trigger regeneration of mesh metadata, potentially disrupting mesh operations.
Affected context
The issue occurs in the TOTOLINK T6 4.1.5cu.748_B20211015 mesh-management functionality when the cs_broker component processes MQTT messages for the meshInfoKick function. Risk is present where the MQTT attack surface is reachable and the function does not adequately authenticate or authorize requests before changing mesh information or state.
Exploit preconditions
- Attack surface: MQTT messaging through the device's cs_broker component
- Required access: No authenticated account may be required when the relevant broker endpoint is reachable
- Trigger point: Processing of a crafted message by the meshInfoKick function
- Why exploitation is possible: The function performs mesh-state changes without sufficiently enforcing access-control requirements
Why this happens
The vulnerability results from an authorization failure in a management operation that modifies mesh information. The meshInfoKick function accepts state-changing MQTT requests without reliably verifying that the requester is permitted to perform the operation, mapping to CWE-284, Improper Access Control. The insecure pattern is exposing privileged device-management behavior through a message interface without enforcing authentication and authorization at the operation boundary.
Potential impact
- An unauthenticated attacker may remove or clean mesh information and state
- Mesh connectivity or device-management behavior could be disrupted
- Mesh metadata may be regenerated without authorization
- Dependent wireless or network services may experience instability
- Repeated requests could contribute to denial-of-service conditions
General mitigation guidance
Mitigation generally focuses on enforcing authentication and authorization for every state-changing MQTT operation, including mesh maintenance functions. Secure-by-default designs should keep management messaging inaccessible from untrusted networks, validate message origin and structure, and separate read-only telemetry from privileged control actions. Reducing trust in broker-supplied input, applying explicit allowlists for permitted operations, limiting device-management privileges, and isolating broker interfaces can reduce exposure. Where feasible, management services should run with only the privileges required for mesh maintenance.
How teams detect issues like this
Detection can include static analysis and code review for MQTT handlers that invoke state-changing functions before authentication or authorization checks. Reviewers should look for missing ownership checks, broker topics exposed beyond trusted management networks, and authorization decisions based only on message content. Runtime monitoring can alert on unauthenticated or unusual meshInfoKick activity, repeated mesh-state resets, and unexpected metadata regeneration. Network policy and broker ACL reviews can verify that only approved principals and interfaces can publish management messages.

