Meletis Belsis - Introduction to information security

What is
Security ?
Part I
Meletis A. Belsis
2003, Athens, Greece
Mediterranean College
Computer Crime

Setting the Scene
• Security is one of the oldest problem that governments
,commercial organizations and almost every person
has to face
• The need of security exists since information became a
valuable resource
• Introduction of computer systems to business has
escalated the security problem even more
• The advances in networking and specially in
distributed systems made the need for security even
greater
• The Computer Security Institute report, notes that in
year 2003 computer crime costs where increased to
more than 450 million dollars in the USA alone.

Profiling Adversaries
• Adversaries that target corporate system are
numerous:
• These can be general classified in the
following categories:
– Hackers
– Employees (both malicious and unintentional)
– Terrorists groups
– Governments
– Opposing Industries

Security
• So now we know that we need security.
BUT what is security anyway ?
• Many people fail to understand the meaning of
the word.
• Many corporations install an antivirus software,
and/or a firewall and believe they are protected.
Are they ?

Security through obscurity
• Consider some cases :
– An internal employee wants to revenge the company
and so publishes private corporate information on the
NET.
– The terrorist attack on the twin towers (in USA) had as
a result many corporations to close. Why ?
– An employee forgets his laptop into a café. This laptop
contains all corporate private information.
HOW CAN A FIREWALL PROTECT FROM THE
PREVIOUS ?

Security: easy to understand, difficult to implement
“In the real world, security involves processes. It involves
preventive technologies, but also detection and reaction
processes, and an entire forensics system to hunt down
and prosecute the guilty. Security is not a product; it
itself is a process. …. ”
Bruce Schneier
(Secrets and Lies, Wiley and Sons Inc.)

• Security contains a number of
tools , processes and
techniques.
• These in general cover three
main requirements:
– Confidentiality
– Integrity
– Availability
• Depending on the security
requirements a system has, one
can concentrate only on one of
the previous or all of them.
• A new requirement enforced
by the operation of e-markets
is non-repudiation.

• Computer Security is difficult to implement due to
the following:
– The cost of implementing a security system should not
exceed the value of the data to be secured.
– Industries pay huge amount of money for industrial
espionage.
– Users feel that security is going to take their freedom away
and so often they sabotage the security measures.
– Computer prices have fallen dramatically and the number
of hackers have been multiplied.
– Security managers work under strict money and time
schedule. Criminals do not have any time schedule and
they do not need any specialised software.
– Hackers are often cooperate with known criminals.
That is why, total security is almost infeasible.

The Art of
Hacking
Part II
Attacking Corporate Systems

Information Gathering
• The first step to hacking is to gather as much
information as possible for the target.
• This information is later used to draw a map of
the corporate network.
• This map is used to define and design an attack
methodology as well as identify the needed
attack tools.
• The extreme case of information gathering is
called dumpster diving

Information Gathering : Searching the
Corporate Web site
• Searching the corporate web site for
information:
– Statements like : “This site is best viewed with
Internet Explorer” could uncover that the
company uses Microsoft Web Server.
– Email Addresses. These are used to identify user
names. i.e. username@thecompany.com
– Office Locations: Companies with office locations
in different countries would probably use a VPN to
interconnect.
– Company News

Information Gathering : Searching the Internet
• Searching the WEB can provide valuable information
– Using the link directive. i.e. link: www.somecompany.com
provides information on the sites that link to the corporate web
site.
– Searching the greater WEB using the company’s name
• Searching public WHOIS databases :Provide information
about the domain name of the company.
• Searching the ARIN Whois Database: Provide a
database with all register IP addresses.
• Searching technical forums using either the name of the
administrator or the name of the company.

Information Gathering :Being Polite…..
• When the initial search has finished, it is now time to
ASK the network itself. Believe it or not most networks
are quite polite.
– DNS Interrogation. It can be performed by simple
using the nslookup program.
– Using the PING command (ICMP Echo ). Can unveil
hosts that are connected and are not protected by a
firewall.
– Using the TraceRoute command we can identify
which is the IP of the router that connects the
corporate network to the Internet.

NeoTrace: Windows Based TraceRT

Information Gathering :Identify Running
Services
• Having a map of the internet hosts that are accessible from the
internet, we must now identify the services that they offer and the
operating system that is installed on each host.
• Special programs like nmap and superscanner are used to
interrogate each port in a host.
• Detecting Services
– The Scanner tries to open a connection to each port of the target host (By
sending Syn messages) .
– The open ports that respond show the services that are running.
• Detecting the OS
– The Scanner sends specific erroneous message to the ports. OS response with
different messages.

SuperScan: Windows Based Port Scanner

Information Gathering :Scanning undetected
• Many firewalls can detect these scanning attempts.
So scanners use some alternate techniques:
– Slow Scanning
– Distributed Scanning
– Half Open Connection
– Fragmented packets
– XMAS
– FIN
– FTP Bounce

Password Cracking
• Adversaries use two methods to attack
passwords.
– Brute force: Try all key combination in the
password space.
– Dictionary: Use a dictionary of known words and
try each word along with their combinations.
• These attacks can be performed either locally or
remotely

L0phtCrack: Windows Password Cracking

VIRUSES
• Computer Viruses are
categorised in:
– Normal viruses
– Trojan Horses
– WORMS
• Today there are more than 2,500
virus ready to be downloaded.
• A user can get infected by:
– Running a program
– Opening an email
– Visiting a web site (evil Trojan)
– Opening a .doc file
• Today virus creation and
mutation centres can be freely
downloaded from the Internet

SubSeven: Visual Interface to Control Infected PC

Denial of Service Attack (DoS)
• The idea behind these attacks is to make the
target system unavailable to its authorised users.
• Typical attacks include but not limited to :
– Ping O’ Death (sending packets of size greater that
65,535)
– SYN Flooding Attack (Starting Many half-open
connections)
– Smurf Attack (sending requests to broadcast address
with a spoofed IP address)
– Domain Name Server DoS (Requesting DNS quires
from multiple DNS Servers with a Spoofed IP

Distributed Denial of Service (DDoS)
• Hackers have used the
distributed power
internet offers.
• Tools are now perform
DoS attack from
multiple hosts at the
same time.
• Examples are:
– Tribal Flood Network
– TFN2K
– Stacheldraft

Sniffing
• Ethernet provides the ability to run a network card
in Promiscuous mode. This allows the card to
read any packet travelling on the network.
• Sniffing software are using this to read all data
transmitted in the local net.
• Sniffers can be programmed to steal information
associated only with specific protocols or
programs. i.e. read all information from http
packets only.
• Some sniffers can be even programmed to
transmit sniffed passwords back to the attacker.
• The first and most used sniffer is the TCPDump .

SnifferPro: A windows based Sniffer

System Flaws and Exploits
• Most systems today contain bugs. These are coming either
from the system designers, implementers or the ones that
manage the system.
• Hackers can use these bugs to gain access to systems.
• Examples of such are :
– Default accounts
– Poor User Accounts
– Allowing outside anonymous Telnet connections to the Web
Server
– Allowing trusted connections
– Buffer Overflows
– Allowing Banners in services
– Allowing NetBios over TCP/IP when not needed.
• The Internet has a vast amount of software that test a given
server for a number of such exploits.

Simpsons’: A CGI vulnerability scanner

Social Engineering
• One of the oldest and easiest form of hacking.
<Hacker is calling the administrator >
 Hallo I am <<name of an employee>>. My user name <<user name as seen
on email address>>. I am new to the company but I forgot my system
password <<be very unhappy>> but my manager ask to find him
some files. If I tell him that I forgot my password , I am afraid that
he is going to fire me. Please help <<be persuasive>>>>
<Administrator wants to help a fellow employee>
Ok. Do not cry now. That is why we are here for. I am going to
reset your password to newpassoword. Just do not forget it again.
<Hacker thanks the polite employee>
 Oh thank you so much. I am going to buy the coffee when we
meet. You are a lifesaver….
(The scenario works even better is the hacker is a female and the administrator is a
male.)

IP Spoofing
• Hackers usually change the IP address in their datagrams.
• This happens for two reasons:
– To avoid getting caught.
– To bypass security tools, and systems that allow trusted connections.
• Changing just the IP is called a blind attack, because the
hacker never sees the response from the target.
• In order to see the response the hacker has a number of ways:
– Install a sniffer to the target network.
– Use Source Routing
– Use ICMP redirect
– If both hacker and target are located on the same network use ARP
spoofing.
– DNS cache Poisoning.
• Software programs like A4 proxy allows hackers to use a
number of anonymous servers before they attack. Thus their
real IP is almost untraceable.

A4 Proxy : Using multiple anonymous proxies to
hide the IP address

The Next Step
• So now I am in what am I doing next ?.
1. If you do not already have, try to gain root
access.
2. Find and clear Log Files.
3. Install a Root Kit to ensure that you will have
access in the future

Protecting
Corporate
Systems
PART III
Information Security Measures

Is it possible ?
• Total security is not feasible.
• Systems must be secured depending on their value.
• Security measures are applied according to the threat
level a system has.
• The first step is to understand the threats, to your
corporate systems. This can be done by a risk analysis
process.
• In this stage remember that security is a business
requirement

Creating a DMZ zone
• The first security measure is
to seal the internal network
from the outside world.
• This is performed by
developing a network called
Demilitarized Zone (DMZ).
• The DMZ contains all the
servers that must be
accessible from the outside
world
• NOTE that we must always
assume that servers in the
DMZ are going to be hacked
at some point.

Firewalls
• Firewalls exist into types:
– Packet filters: Are operating on the protocol level. They use
a firewalling policy to allow the packet to pass or to drop the
packet.
– Proxy Servers: They operate at the application level. They
are always located between the user requests and the servers
response. Thus allowing us to enforce policies on which
users can access the internet and on which port.
• Packet Filters are usually located on the router, while
Proxies are installed on computers
• A network may use any number of the previous
depending its size and architecture.
• Known Firewalls are Checkpoint’s Firewall-1, Cisco
PIX, Microsoft’s ISA.

Intrusion Detection Systems (IDS)
• Intrusion detection systems are
used to detect attacks to the
network and inform the
administrator.
• IDS are organised into two
categories :
– Signature based : They hold a
database of known attacks and
they test packets against the data
stored in the database.
– Anomaly based: They test the
traffic against anomalies. I.e.
why does the network has so
heavy traffic at 2 in the
morning ?
• When the IDS detects an attack it
inform the administrator with a
number of ways : email, sms,
pager

Honey Pots
• These are the sacrificed lamps of a network.
• Honey pots are software programs that when installed on a
computer they can simulate a number of systems i.e.:
• Windows NT Server.
• Unix Server.
• Apache Server
• Microsoft Exchange Server
• These simulated systems look unprotected from the outside
world (i.e. open ports, default accounts, known exploits.
• Hackers scanning for victims detect the simulated systems and
try to hack them. The honey pots allow hackers to enter but
record all their moves and inform the administrator.
• Honey pots can be installed either in the DMZ or in the local
network.

Anti sniffing
• The general idea is to make the sniffing host reply to
a message that he should not be able to listen.
– For example creating a packet with a fake MAC address
but with the IP address of the sniffing host. If the host
acknowledges the packet the it is in promiscuous mode.
• Another way is to transmit unencrypted login details
for a fake (honey pot) server to the network. If
someone tries to use this account then someone is
sniffing the network.
• NOTE that using switches instead of hubs will make
a sniffers life much more difficult.

L0pht Antisniff : A windows based program to
detect sniffers

Antivirus
• Antivirus programs are known to most users.
• Such programs can be applied either as
– Standalone : Each copy of the program is
responsible of protecting the specific host on
which it is installed.
– Network based : Each copy of the program is
responsible of protecting the specific host, but they
are all managed by a Antivirus Server.
• Note that using an antivirus program without
updating its virus database does not provide
protection

Security Awareness
• No matter what security tools are going to
be used, if users do not know about security,
hacks are going to be common.
• There are many ways to educate users on the
issues of security:
– Use of seminars
– Use of posters
– Use of e-mail messages
– Enforce penalties

Penetration Testing and Security analyzers
• Security systems must be regularly tested for
flaws.
• These flaws are usually created from bugs in the
software programs, or from bad management (i.e.
bad passwords)
• The process of testing a system is called
penetration testing.
• The process uses a number of hacking / security
programs that test a system for a number of known
flaws and provide advice on securing these flaws

Microsoft Baseline Security Analyzer: Tests the systems
for known bugs

Additional Security Measures
• Encryption/ Decryption
• Digital Signatures / PKI
• AAA
• Security Protocols
• Physical Security
– The Jaguar Paradigm
– The polite Employees paradigm
• Security Policy

Questions ?
In Accordance
Meletis A. Belsis

Meletis Belsis - Introduction to information security

Recommended

More Related Content

What's hot (20)

Viewers also liked (20)

Similar to Meletis Belsis - Introduction to information security (20)

More from Meletis Belsis MPhil/MRes/BSc (7)

Recently uploaded (19)

Meletis Belsis - Introduction to information security