Comp System Sec

Page 1

Lecture One - T|M

GOALS AND ADVERSARY

Security, in general, is all about achieving some goal when there is an adversary present, The best way to think about security is to break it up into three parts

Policy (CIA Triad)

Confidentiality

Measures are undertaken to ensure systems are designed to prevent sensitive information from reaching the wrong people while making sure that the right people can, in fact, gain access to that information

Integrity

Integrity involves maintaining the consistency, accuracy, and trustworthiness of data over its entire lifecycle. Data must not be changed in transit and steps must be taken to ensure that data cannot be altered by unauthorized people. These measures include file permissions and user access controls.

Means should also be put in place to detect any changes in data that might occur as a result of non-human-caused events such as server crashes

Availablity

Availablity is best maintained by rigorously maintaining all hardware, performing hardware repairs immediately when needed, maintaining all software systems free from conflicts and applying all necessary system upgrades regularly

Threat Model

A way of narrowly thinking about the sorts of protection you want for your data. It's impossible to protect against every kind of attack and attacker, so it's better to concentrate on which people are most likely or would likely want to get hold of the data being held, where hey might want it from and ways in which they may attempt to get it.

Coming up with a set of possible attack you plan to protect against is called threat modeling. Once you have a threat model, you can then conduct some risk analysis.

Documents and works out assumptions about potential adversaries

Mechanics

Attack Prevention

can be defined as a series of security mechanism implemented to prevent or defend against various types of attack before they can actually reach and affect the target systems. An important mechanism is access control which is defined as the process of limiting the access to the resources of the Information System. Access can be implemented at different levels such as the operating system the network and the application layer.

A firewall is also an important access control system that is implemented at the network layer the concept behind the firewall is to separate the trusted network from the entrusted network known as the internet. The firewall prevents the attack from the outside world against the machines inside the internal network by preventing connection s attempts from the unauthorized entities located outside.

Attach Avoidance

The expansion of connectivity of computers makes the need of protecting the message and message from tampering reading important. This is the technique in which the information is modified in a way that makes it unusable for the attacker. This is performed under the assumption that the attacker may have access to the subject system/information. The sender preprocesses the information before it is sent to the unsecured system and the same is again post-processed on the receiver end systems. This encryption and decryption is perfumed by cryptography mechanism, they are further divided in the following forms

Secret Key Cryptography

Public Key Cryptography

Hash Functions

Attack Detection

In this mechanism, it is assumed that the attacker is able to bypass the installed security measures to access the desired target/information. When such incidents happen attack detection takes the responsibility to report someone that something went wrong somewhere in the system.

Attack detection is not an applicable mechanism instead of that it’s a check/measure which will make sure that if anything happened badly in the system then someone should be notified of the same

Software and hardware systems

EXAMPLES OF WHEN THINGS GO WRONG

Policy

Account recovery questions were introduced to allow users to recover their account if they have forgotten their password. Introducing a feature like this changes the policy from:

You can log in with your password to You can log in with your password or by answering correct recovery questions.

Another example is the policy Gmail used to use where they would send emails to a recovery email address.

Gmail send email to recovery email @ me.com (owned by Apple)

Me.com allowed you to rested your password if you knew your billing address and the last four digits of your credit card number.

The user under attack had an account at Amazon.

Amazon allowed people to buy accounts without logging in and pay with new credit cards. Following this

Amazon also allowed people to retest their password using a valid credit card which had made purchases for that account, meaning once an attacker had bought something for that account they could then reset the password and take control of the account.

Once an account had been taking control of, Amazon did have a mechanism in place which would only display the last four digits of every credit card used, which was all that was need to reset the account @ me.com.

Threat Models

Human factors are one of the weakest areas when looking at threat models as we are quite bad at assuming the thing people will do., such a picking a good strong password, not fall for phishing attacks. With this in mind, it is always best to avoid making threat models based on assumptions of actions or things users will do.

MIT in the 1980's developed a computer network authentication system named Kerberos. When they initially developed this it was using 56-bit encryption to generate their keys, they employed a number of people to test how secure this was and it was deemed to be extremely strong.

Years later a number of security students were able to successfully hack the system using some specific hardware, simply because technology had moved on so much 56 bits was trivial.

So the lesson to be learned here is to always take into account that thing evolves especially technology.

SSL/TLS has 300+ certified authorities around the world