0% found this document useful (0 votes)
18 views57 pages

Information Sys Security

Pdf

Uploaded by

Ray Rose
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
18 views57 pages

Information Sys Security

Pdf

Uploaded by

Ray Rose
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 57

Network Security (Computer Security)

Addresses three requirements:


 Confidentiality. Prevent unauthorized users
from accessing or utilizing resources
 Availability. Ensure that resources are
available when users need access to them
 Integrity. Verify the integrity of the resources
(make sure things have not been corrupted or
maliciously altered)
Security Aspects

Three aspects of security:


 Vulnerability - identify the parts of a network or
system that can cause problems in terms of
overall security
 Threats - persons, entities, organizations, or
events that can exploit a vulnerability and cause a
system to fail
 Countermeasures - steps or actions taken by
system administrators to reduce vulnerabilities or
to prevent security threats from becoming realized
Attacks

 "Attacks" are specific instances of a threat being


realized, i.e., when somebody attempts to break
into a system or circumvent security policies, this
is called an "attack"
Network Security Threats

 Two categories of threats


 Passive threats such as eavesdropping,
monitoring of transmissions, etc.
 Active threats involve some modification of
transmitted data or creation of false
transmissions
Passive Attacks

 Involve no modification of data stream


 Eavesdropping or monitoring of transmissions
 Two types:
 Release of message contents
 getting access to confidential or sensitive information in an
e-mail message or in a transferred file
 Traffic analysis
 Very subtle
 Attacker captures messages to analyze the nature of
different ongoing transmissions
 Very difficult to detect
An Example of Passive Attack

Packet sniffing:
 broadcast media
 promiscuous NIC reads all packets passing by
 can read all unencrypted data (e.g. passwords)
 e.g.: C sniffs B’s packets

A C

src:B dest:A payload


B
Active Attacks

 Involve some modification of data stream or


creation of false stream
 Can be divided into four categories
 Masquerade
 When one pretends to be a different entity i.e
impersonating an entity that has extra privileges
 Replay
 Involves passive capture of a data unit and its subsequent
retransmission to produce an unauthorized effect
Active Attacks (Cont’d)

 Modification of messages
 Some potion of a legitimate message is altered
 Messages are delayed or reordered to produce an
unauthorized effect
 Denial of service attack
 Prevents or inhibits the normal use or management of
communication facilities
 Examples:
 Attacker may suppress all messages directed to a
particular destination
 Disabling a network or overloading a network with
messages
Example 1: Active Attack

IP Spoofing:
 can generate “raw” IP packets directly from application, putting
any value into IP source address field
 receiver can’t tell if source is spoofed
 e.g.: C pretends to be B

A C

src:B dest:A payload

B
Internet security threats

Denial of service (DOS):


 flood of maliciously generated packets “swamp” receiver
 Distributed DOS (DDOS): multiple coordinated sources swamp
receiver
 e.g., C and remote host SYN-attack A

A C
SYN
SYN
SYN SYN SYN

B
SYN
SYN
Passive versus Active Attacks

 Passive attacks are difficult to detect


 Active attacks are easy to detect
 Passive attacks are easy to prevent such as by
using encryption techniques
 Active attacks are difficult to prevent as the
process would require physical protection of all
network facilities and paths all times
Absolute Security?

 No computer or network system can ever be


considered absolutely secure
 The more secure a system needs to be, the more
it costs to implement additional security
measures
Important Note about Computer Security

 Computer security has become an activity of


identifying threats and vulnerabilities in a
system, estimating the risk of threats being
realized, determining the value of the resources
that are at risk, and comparing the benefits to be
achieved by securing a resource against the cost
of implementing that security
Examples of vulnerabilities

 Physical vulnerabilities
 Example: parts of a computer may be physically
damaged by a number of different threats
 Natural vulnerabilities
 Example: Computers can be affected by natural events
(flood, fire, heat, humidity, dust, etc.)
 Hardware/software vulnerabilities
 Example: Computer components fail to work properly
(wear out, incorrect design, implementation problems,
conflicts with other systems)
Examples of vulnerabilities (continued)

 Media vulnerabilities
 Examples: disks and tapes wear out, erasure
vulnerabilities, other physical characteristics of media
that can fail, magnetism
 Emanation vulnerabilities
 Example: Computers and computer systems "leak”
information in ways we don't want them to (radio
frequency leakage from network wiring, video monitors)
 Communications vulnerabilities
 Communications can be intercepted or faked.
Examples of vulnerabilities (continued)

 Human vulnerabilities
 Vulnerabilities related to the fact that computers are
typically used by people; people give out information
they shouldn't, people often forget things, delete files or
enter in incorrect commands, improper usage of a
system, bribery/sabotage
Example Sources of Threats

 Natural and Physical Threats


 Threats that are inherent from being physical devices
and subject to disasters (e.g., earthquakes, fires, floods,
lightning, hurricane, power spikes,etc.)
 Unintentional Threats
 Dangers due to ignorance or accident (dropping a disk
drive, spilling a drink into a computer, using equipment
improperly, entering the wrong command)
Example Sources of Threats (Cont’d)

 Intentional Threats
 Outsiders vs. Insiders
 Outsiders threats
 From outside an organization
 Foreign Intelligence Agents (spies), terrorists, criminals,
corporate Raiders, crackers
 Insiders threats
 From inside an organization
 Fired or upset employees (or students), coerced employees
(blackmail, bribery, threats to employment), lazy or untrained
users
 One of the most effective attacks on a system is to
combine an outsider threat with an insider threat
Examples of Countermeasures

 Backups  Audit Trails / System Logs


 User Training  Inquisitive
 Uninterruptible Power Users/Administrators
Supplies (UPS)  Locks/Physical Security
 Mirrors  Network Analyzers/Reporting
 Security/System Tools
Administration Tools  Secure Hubs/Equipment with
 Enforcement of good security features
password schemes  Virus Detection/Eradication
 Alternate Authentication systems
Mechanisms  Firewalls
 Encryption
Passwords, encryption, and authentication

 Mechanisms of determining identity of requestor


and degree of access
 Identification - mechanism used so that a user
requesting a service can identify himself/herself to the
system. The most common identification mechanism is
a username, although others can exist.
 Authentication - mechanism used to verify the
identity of a person. The most common authentication
mechanism is the use of a password.
 Authorization - mechanisms used to determine what
actions a user is allowed to perform on a system. The
most common authorization mechanisms are
permission controls and/or access control lists.
Authentication

 Can be broken down into three broad categories


 Something a user knows - examples: passwords,
birthdates, mother's maiden name, PIN numbers, SSNs,
birthplace pet names, etc
 Something a user has - examples: keys, tokens,
credit cards, identification cards
 Something the user is -- a physical characteristic
unique to the person being authenticated. examples:
fingerprints, retina scans, DNA scans, voice print
analysis
Password authorization schemes

 Encryption schemes
 An "encryption scheme" is some method that is used to
encode information such that the original contents of
the information are not easily seen or determined by
unauthorized persons
 Two broad classes of encryption schemes
 Two-way encryption schemes
 allows information to be encrypted (encoded) into a format
that can be transmitted and then later decrypted
(decoded) into its original format
 One-way encryption schemes
Terms used for encryption schemes

 Cleartext
 Information in its original, unencrypted form
 Ciphertext
 Information after it has been encrypted
 Key
 Something used in the encryption and decryption
process to convert cleartext to ciphertext and vice-versa
 Note: A good encryption algorithm is one in which it is
nearly impossible to deduce the cleartext form of
information from the ciphertext without having the
appropriate key
Normal Process for Two-Way Encryption

 cleartext --encrypt(key)--> ciphertext --


transmit-to-receiver-->
 ciphertext --decrypt(key)--> cleartext
The language of cryptography

plaintext K K plaintext
A B
ciphertext

Figure 7.3 goes here

symmetric key cryptography:


sender, receiver keys identical and secret
public-key cryptography:
encrypt key public, decrypt key secret
How good is an Encryption Scheme?

 Difficulty of "breaking a code or encrypted


message" depends on the size of the key
 In general, a longer key value will make a
particular encryption more secure
 For most encryption methods, the only way a
cracker can break the encryption is to try every
possible key (this is known as a "brute-force
attack")
 The greater the number of possible keys, the less
likely a brute-force attack will succeed
Encryption Key

 Most keys are measured in terms of the number of


bits
 A 16-bit key means that there are 65K possible keys
(216=65536)
 A 32-bit key means that there are 4 billion possible keys
 A 64-bit key means that there are 16 quadrillion possible
keys
 A system that can test 1000 key values per second
would take
 65 seconds to test a 16-bit key
 49 days to test a 32-bit key
 Thousands of years to test a 64-bit key
One-way Encryption Scheme

 A cleartext string is converted into a ciphertext


form
 Cannot be converted back to the cleartext form
 Often used for authenticating passwords
Example: One-way Encryption Scheme

 A user enters a password into a system to set


his/her password
 The system then encrypts the password using a
one-way encryption scheme, and stores the
ciphertext version of the password in a database
 Then, when the user wishes to authenticate
himself/herself to the system, he/she supplies the
same password to the system
 The system encrypts the user-supplied
authentication password, which should match the
encrypted-form stored in the system's password
database
One-way Encrytion Schemes (Cont’d)

 Irreversible encryption schemes like some


mathematical operations (Ex. modulus operation)

 Difficult to break to determine the original


password
Password attacks

 One-way encryption scheme is effective but can


fail
 Many users will not choose good passwords
 Crack programs
 Do not need to attempt all combination of passwords
 Can use common words in an attempt to find a poorly
chosen password
 Most crackers look for any password to gain access
to a system
 Rarely some crackers look for a password for
administrative access
Countermeasures for preventing password attacks

 Train and/or require users to use good passwords


 A good password is not a word, a name, a birthdate or a
social security number. Why?
 Best passwords are random sequences of characters
and non-letter with mixed case
 Example: Using initial letters of an easy-to-remember
phrase. A password "ttl*hiwwya" is derived from phrase
"twinkle twinkle little star, how I wonder what you are"
 Configure systems to refuse to allow users to set
a bad password
Countermeasures for preventing password attacks

 Secure the password databases so that the


encrypted forms are not available to users (on
UNIX systems, this known as "shadow passwords"
since the actual passwords are kept in a
"shadow" file which is not accessible to average
users)

 Don't use passwords as the primary


authentication mechanism (there are others)
Problem with Symmetric Key Cryptography

 Requires sender, receiver know shared secret


key
 Questions:
 How to agree on key in first place (particularly if
never “met”)?
 How to tell about the key to other party over
Internet, a publicly accessible medium?
Public Key system encryption methods

 Two-way encryption system


 A complex mathematical algorithm generates keys
with two halves
 Whatever is encrypted by one half of the key can
only be decrypted by the other half
 Has the capability of allowing messages to be sent
securely without having to keep a key secret from
the rest of the world
 Two halves of the key are known as the public
key and private key depending on who receives
the keys
Public Key Cryptography

Figure 7.7 goes here


Public Key System Encryption Methods(Cont’d)

 A person wanting to receive secure messages


 Will generate a public/private key pair
 Supply the public key to anyone who wants to send them
a secure message
 The receiver keeps the private key secret
 When a sender wants to send a secure message
to a receiver
 Sender encrypts the message using receiver's public key
 Encrypted message is sent to receiver
 Receiver decrypts the message using his/her private key
 Message delivery is secured
 No one but the receiver knows the private key
corresponding to the public key
Authentication Mechanism: Digital Signatures

 Digital signature
 Mechanism for verifying the authenticity of the sender of a
message
 Can use public key cryptography to implement
 Sender encrypts the message using his/her private key
 Recipients decrypt the message using the sender's public
key
 The message is authentic if readable after decryption
 Can be used to validate that a document has not
been altered in transit if the encrypted form is
altered in any way, then it will not decrypt with the
sender's public key
Digital Signatures

Cryptographic technique Simple digital signature for


analogous to hand- message m:
written signatures.  Bob encrypts m with his
 Sender (Bob) digitally public key dB, creating
signs document, signed message, dB(m).
establishing he is
document owner/creator.  Bob sends m and dB(m)
 Verifiable, nonforgeable: to Alice.
recipient (Alice) can
verify that Bob, and no
one else, signed
document.
Pretty Good Privacy (PGP) Protocol

 Popular protocol
 Used for sending/receiving encrypted and
digitally signed electronic mail over the internet
 Is known as PGP (Pretty Good Privacy)
 Used for interactive Internet communications
 Usually known as "Secure Sockets Layer” (SSL)
 With SSL, every packet sent over the network is
encrypted
 A new protocol is being developed for use on the
Internet that is known as IPsec (Internet Protocol
w/security)
Side Note

There are U.S. export restrictions on the export of


any form of cryptography (other than those
approved by the
NSA) outside of the United States/Canada. This
includes
technologies such as PGP and public key
cryptography
Programmed Threats

 Programmed Threat
 An item of software or hardware that is intended to (or
has the result of) compromising system security
 Categories of programmed threats
 Security tools and toolkits -- programs designed to
assist with securing a system by exposing vulnerabilities
 Back doors -- secret codes or routines that allow
unauthorized access to a system
 Logic bombs -- a program that is set to do something
specific (often destructive) whenever a pre-programmed
event occurs. Example events: dates, a certain
combination of keystrokes, etc.
Programmed Threats (Cont’d)

 Categories of programmed threats (Cont’d):


 Viruses -- program that attaches itself to other
programs and is executed whenever the host program
is executed (so it can copy itself to other programs)
 Worms -- programs that are designed to attack as
many computers as possible through a network (a worm
is usually a standalone program)
 Trojan Horses -- program that appears to have one
function but actually has another hidden purpose
 Bacteria or Rabbit attacks -- programs that rapidly
multiply in order to bring down a system
Programmed Threats (cont’d)

 Bacteria and rabbit programs exhibit a


special type of attack that is known as a "Denial
of Service" (DoS) attack, preventing other
(legitimate) users from being able to use the
system effectively
 One form of denial of service attack is to flood a
machine or network with ping requests
 Email spam often has the effect of being a
denial of service attack
Side Note

 In the past few years, some organizations (e.g.,


the Pentagon) have been experimenting with
reverse denial-of-service attacks. When a
computer detects that someone is attempting to
breach its security, it retaliates by flooding the
attacking computer with denial-of-service
attacks. (Note that this is illegal.)
Firewalls

 Firewall
 A general name for hardware, software, or
combination of the two
 Used to protect internal network from
intruders
 Works as a transit device that examines traffic
and decides what traffic (packet) should pass
or should not pass
Typical Firewall Components

 Hardware
 Routers
 Dedicated computers
 Special hardware device such as Cisco PIX
(Private Internet eXchange)
 Software
 Firewall software
Role of Routers

 Operate at the network level


 Understand source and destination IP addresses
 Understand packet types
 Can be configured to
 allow only certain type of packets to pass
 allow connections to be established only from a certain
set of internal hosts
 block any attempts to access internal hosts from the
external, unsafe Internet
 Process is called packet filtering
A Simple Firewall System with a Router

Internal Network Internet


Router
Limitations of Routers

 Routers understand only the information in the


TCP/IP headers
 Routers do not understand data in the application
protocol level (such as HTTP)
 Routers cannot enforce authentication of users
 Routers access control capabilities are limited to
IP address level
 Routers are not able to provide meaningful logs
of transactions that occur
Application-Level Firewalls

 Perform more complex tasks than the capabilities of router


hardware
 Understand the application
 Understand the application content
 Can perform intelligent tasks such as filtering mail based
on its source and contents
Application-Level Proxy Servers

 Software programs that are familiar with a


specific or several protocols
 Application-level gateways
 Clients on internal networks make requests to the
proxy server instead of connecting directly to a
remote device
 Proxy server performs the actual request on
behalf of the client
 Relay authorized traffic between interfaces
 Block unauthorized traffic
Important Features of Application-Level Proxy Servers

 Latency Reduction, Bandwidth Conservation


 Cache data and service requests from their cache
 Advanced Access Control
 Can perform authentication and use authentication
information in access control
 Advanced Filtering
 Understand the internals of the protocol
 Filter out sensitive information
 Insert additional information
 Remap request to a mirror site
 Logging and auditing
 Able to provide extensive logs of transactions
 Can be used for auditing of Web traffic
A Firewall with a Proxy Server and a Router

Proxy
Server

Router
Internal Internet
Network

•Routers and proxy servers are used together


•Provide a more secure solution
•Router protects the proxy server from Internet
Problem

 We want to prevent people from accessing internal


networks
 Also we want to provide services to outside world
 Services must be visible to external networks
 How is it possible?
 Solution:
 Create what is commonly known as a "DMZ"
(Demilitarized Zone), a third network attached to firewall.
DMZ is a subnet.
 Place servers and other computers that need to be visible
to users outside of the organization's on DMZ
 These services are vulnerable. Hence place a second firewall
that protects the internal LAN from the computers in DMZ
A Firewall with a Proxy Server Enclosed in a Subnet
Protected by Two Routers

Proxy
Internal Server
Network Router Router Internet

DMZ

 Proxy server is surrounded by a router on both sides


 More secured
 Proxy server is on its own subnet
 Subnet is referred to as demilitarized zone (DMZ)
Demilitarized Zone (DMZ)

 Part of network which lies between Internet and


internal network (intranet)
 Shields rest of intranet from threats of Internet
 More exposed to threats of Internet than internal
hosts
 More stringent security measure must be taken
on this zone

You might also like