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  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]
Date: Thu, 19 Jan 2017 14:45:38 +0100
From: "Curesec Research Team (CRT)" <>
Subject: [FD] Tap 'n' Sniff

Content Table

1. Introduction
2. Failsafe mode
3. Installing Openwrt
4. Configuring Openwrt
5. Testing

1. Introduction

The goal of this guide is to provide a reliable and fast way for creating a lan
tap for red team assessments of networks. While this was our main target this
tap is also quite helpful if you want to have a great device for your daily
analysis of network attached computers. Before we started with our
implementation we made a list of things which were mandatory. The hardware had
to be small, have at least two lan ports and wifi, cheap and opensource
included or available.

After some research we choose the TL-WR810N, a 20 euro Pocket Router which
should be available in most electronic stores. It features two lan ports and a
wifi card, which allows us to bridge the lan interfaces and create a hidden AP
to connect to the device. It should be said that the device only supports
Fastlan (100 Mbit/sec) and not Gigabit lan (1000 Mbit/sec) but at this size you
can't be picky and it's quite difficult to find something better even online
when ordering from a foreign country so there is that. After we are finished we
want to be able to listen to the network traffic between the taped sources,
manipulate packets or directly pivot into the network. For our setup we are
going to use openwrt instead of the default TP-Link firmware. We are currently
working on creating an image that will make the configuration of openwrt
obsolete so stay tuned for info regarding this. And this is how it actually

[wr810n_front] [wr810n_back] [wr810n_ports] [wr810n_led_switch]

On the inside we find a SoC (System on Chip), namely the Qualcomm Atheros
QCA9533 which is capable of wireless ABGN communication and has a clock speed
of 560 MHz according to wikidevi. There is also 64 MB of Ram and we can use 4.6
MB of flash storage with 1.1 still availiable after finishing this guide. Below
is the output of cpuinfo, free and df. It is interesting that when we opened
the device later on we actually found the cpu to be a different one, the
Qualcom QCA9531-BL3A but apparently they are identical. Basic information found
via commandline:

root@...nWrt:~# cat /proc/cpuinfo system type : Qualcomm Atheros QCA9533 ver 2
rev 0 machine : TP-LINK TL-WR810N processor : 0 cpu model : MIPS 24Kc V7.4
BogoMIPS : 432.53 wait instruction : yes microsecond timers : yes tlb_entries :
16 extra interrupt vector : yes hardware watchpoint : yes, count: 4, address/
irw mask: [0x0ffc, 0x0ffc, 0x0ffb, 0x0ffb] isa : mips1 mips2 mips32r1 mips32r2
ASEs implemented : mips16 shadow register sets : 1 kscratch registers : 0
package : 0 core : 0 VCED exceptions : not available VCEI exceptions : not
available -------------------- root@...nWrt:~# free total used free shared
buffers cached Mem: 60220 17032 43188 20 1504 4828 -/+ buffers/cache: 10700
49520 Swap: 0 0 0 -------------------- root@...nWrt:~# df -h -T Filesystem Type
Size Used Available Use% Mounted on /dev/root squashfs 2.0M 2.0M 0 100% /rom
tmpfs tmpfs 29.4M 20.0K 29.4M 0% /tmp /dev/mtdblock3 jffs2 4.6M 3.5M 1.1M 76% /
overlay overlayfs:/overlay overlay 4.6M 3.5M 1.1M 76% / tmpfs tmpfs 512.0K 0
512.0K 0% /dev

2. Failsafe mode

Before we begin you should know about the built-in failsafe mode of openwrt.
This exists just in case you make a mistake and loose the connection to the
router by something else than a reboot. It is activated by pressing the reset
button rapidly on startup till the led blinks more frequently than usual. Now
you have to give your ethernet interface an ip like and connect to
the WAN/LAN port. Then you should be able to ssh to as root without
password. In the ssh session you can mount the filesystem with


and reverse the changes that made your system fail. If the worst-case scenario
happens and you have no clue why your system behaves abnormal, you can always
reinstall openwrt. To do so, we copy our image via scp in the tmp directory of
the router:

scp /path/to/image/ root@....168.1.1:/tmp

Now we use the sysupgrade command on the router to install the bin file:

sysupgrade -n /tmp/binary

The -n flag means that we also erase all config files. Don't worry if you loose
the connection, first the router will reboot and then you have to switch your
lan cable to the lan port again.

3. Installing Openwrt

Installing openwrt is as easy as it gets, you just have to configure the
router, preferrably by connecting via lan, using its webinterface and download
the respective firmware image from the Openwrt wiki. We used the EU Version
1.1, also availible at our github(TODO) since we can't guarantee this procedure
to work with other versions that might be published in the future. The easiest
way to install openwrt is via the webinterface and its option Firmware-Upgrade.
Íf, for whatever reason, this fails you can also install openwrt via the serial
console or TFTP. A guide can be found at the wiki. After doing so, you will
loose the connection to the router and will have to reconnect via an ethernet
cable. Make sure to use the lan port on the router, not the lan/wan one.

4. Configurating Openwrt

Now the interesting part begins: To start we open a ssh session to,
the default ip of openwrt. The first thing we do is setting a password with
passwd. Ideally we also connect the router to the internet via the other
ethernet port afterwards and run

opkg update

to get our system up to date. We also want to install tcpdump, python-light
(python is to big) and the normal netcat because the netcat provided by busybox
can't run in server mode. Make sure to remove wpad-mini before installing
hostapd or you will get an error when installing hostapd. If you do get the
error, do not worry, just remove wpad-mini and install hostapd again.

We do all this by running:

opkg remove wpa-mini and opkg install tcpdump python-light hostapd netcat

Secondly we disable a number of services normally ran at boot to reduce
unneccessarity or possible traffic or because they would interfere with our
scripts. To be more precisely, we disable firewall, led, network and odhcpd
with these commands:

/etc/init.d/firewall disable /etc/init.d/led disable /etc/init.d/network
disable /etc/init.d/odhcpd disable

To maintain access to the router we have to set up our AP because since we just
placed it somewhere between interesting devices we can't access the router via
lan. To do so we write a hostapd.conf like this:

vim /etc/config/hostapd.conf interface=wlan0 ssid=sniffntap hw_mode=g wpa=2
wpa_passphrase=networksniffer wpa_key_mgmt=WPA-PSK ignore_broadcast_ssid=1

Now we write a script which will run hostapd on startup, tells dropbear to
listen on the wifi interface for ssh and bridges the two lan interfaces. So,
our actual sniffer.

vim /etc/init.d/sniffer #!/bin/sh /etc/rc.common START=99 #tells openwrt in
which order the scripts in /etc/init.d/ are to be started, 99 is per default
the last entry STOP=99 #same as start but stop start() { echo start fconfig
wlan0 up #the wifi is not activated by default ifconfig wlan0 #
since we disabled dhcp we have to give our interface an IP manually ifconfig
eth0 up #we need to bring our lan interfaces up manually since we disabled
network ifconfig eth1 up brctl addbr br0 #create a new bridge interface
ifconfig br0 up #bring our new bridge interface up brctl addif br0 eth0 #add
interface eth0 to bridge br0 brctl addif br0 eth1 #add interface eth1 to bridge
br0 dropbear -p #tell dropbear to listen on the wifi interface
on port 22 for incoming ssh connections hostapd /etc/config/hostapd.conf #start
our AP } stop() { echo stop ifconfig wlan0 down #take our wifi interface down
killall hostapd #stop hostapd }

Last but not least, make it executable and tell openwrt to run it at boot:

chmod +x /etc/init.d/sniffer /etc/init.d/sniffer enable

5. Testing

To test our configuration just reboot, connect to the wifi network and ssh to in order to log in with your previously set password. In order to
check if everything went as it should you can run ifconfig to check if all
inerfaces are up and brctl show to check your bridge. You should see something
like this:

root@...nWrt:~# ifconfig br0 Link encap:Ethernet HWaddr 98:DE:D0:23:42:42 inet6
addr: fe80::709f:78ff:fec9:cd43/64 Scope:Link UP BROADCAST RUNNING MULTICAST
MTU:1500 Metric:1 RX packets:752 errors:0 dropped:0 overruns:0 frame:0 TX
packets:5 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000
RX bytes:38624 (37.7 KiB) TX bytes:518 (518.0 B) eth0 Link encap:Ethernet
HWaddr 98:DE:D0:23:42:43 inet6 addr: fe80::9ade:d0ff:fe20:9bd7/64 Scope:Link UP
BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:762 errors:0 dropped:0
overruns:0 frame:0 TX packets:12 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000 RX bytes:50678 (49.4 KiB) TX bytes:1276 (1.2 KiB)
Interrupt:4 eth1 Link encap:Ethernet HWaddr 98:DE:D0:23:42:42 UP BROADCAST
MULTICAST MTU:1500 Metric:1 RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:0 errors:0 dropped:0 overruns:0 carrier:0 collisions:0
txqueuelen:1000 RX bytes:0 (0.0 B) TX bytes:0 (0.0 B) Interrupt:5 wlan0 Link
encap:Ethernet HWaddr 98:DE:D0:23:42:44 inet addr:
Bcast: Mask: inet6 addr: fe80::9ade:d0ff:fe20:9bd6/64
Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:140
errors:0 dropped:0 overruns:0 frame:0 TX packets:127 errors:0 dropped:0
overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:17691 (17.2 KiB) TX
bytes:18844 (18.4 KiB) -------------------- root@...nWrt:~# brctl show bridge
name bridge id STP enabled interfaces br0 8000.98ded0234242 no eth0 eth1

After ensuring that the bridge does work we can test our device by placing it
between two devices. Now, the last question we have to answer is how to capture
and store the traffic on a device with so small storage. And the Answer is: we
don't. In the screenshots below you can see that we pipe the output of tcpdump
on our remote Device to wireshark on our local device.

On our local machine we use mkfifo and tell wireshark to listen there:


And on our remote machine we run tcpdump (for copy pasters: ssh
root@....168.2.1 "tcpdump -U -s0 -w - -i br0 'port not 22'" > /tmp/sharkfin):


If you plan on monitoring networks with a lot of traffic consider to give
tcpdump more filters to reduce load on the remote machine.

With this we conclude this article and wish you all happy hacking!

Blog Reference:

Curesec GmbH
Curesec Research Team
Josef-Orlopp-Straße 54
10365 Berlin, Germany

Sent through the Full Disclosure mailing list
Web Archives & RSS:

Powered by blists - more mailing lists