0% found this document useful (0 votes)
15 views22 pages

Auditing System Design and Fundamentals

Uploaded by

swastiknayak2025
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
15 views22 pages

Auditing System Design and Fundamentals

Uploaded by

swastiknayak2025
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

This is a copy of a conversation between ChatGPT & Anonymous.

Report conversation

Uploaded a file

● Auditing Fundamentals: Define logging and auditing as


parts of determining security violations
and explain the anatomy of an auditing system
.
● Auditing System Design:
○ Design a well-unified auditing system, considering
implementation, syntactic issues, log
sanitization, and system logging
.

Below is a clear, exam-oriented answer based strictly on Module 3 (Auditing). I’ve


structured it with definitions, diagrams-in-words, and headings, which is ideal for
VTU/BEC714B answers.

1. Auditing Fundamentals

Logging

Logging is the recording of events or statistics to provide information about system usage,
activities, and performance.
In security, logging captures security-relevant events such as:

Login and logout attempts


File accesses (read/write/execute)
Privilege changes
System calls and network connections

Definition:

Logging is the recording of events or statistics to provide information about system use
Module 3_1
and performance.
Logs form the raw data needed to determine whether a security policy has been violated.

Auditing

Auditing is the analysis and interpretation of log records to:

Detect security violations


Reconstruct the sequence of events leading to a violation
Evaluate the effectiveness of security controls
Deter misuse by ensuring accountability

Definition:

Auditing is the analysis of log records to present information about the system in a clear
Module 3_1
and understandable manner.

Thus:

Logging = Data collection


Auditing = Data analysis

Together, they help determine whether the system has entered a nonsecure
(compromised) state.

2. Anatomy of an Auditing System


An auditing system has three main components:

1) Logger

Collects and records events based on configuration


Stores data in binary or human-readable form
Examples of logged data:
User ID
Time and date
Action performed
Object accessed
Result (success/failure)
👉 Example: Windows security event log records logins, file access, and privilege use.
Module 3_1

2) Analyzer

Examines log data to:


Detect suspicious behavior
Identify policy violations
Recognize attack patterns
Used by intrusion detection systems (IDS)

👉 Example: Detecting repeated failed login attempts or abnormal query patterns.


Module 3_1

3) Notifier

Reports audit results to:


System administrators
Security tools
May trigger automated actions:
Disable user account
Block network connection
Raise alerts

👉 Example: After three failed login attempts, the notifier disables the account and alerts
Module 3_1
the administrator.

Flow of an Auditing System

sql

System Event → Logger → Analyzer → Notifier → Response

3. Auditing System Design (Well-Unified Design) Copy code


A well-unified auditing system integrates logging and auditing as a core security
mechanism.

3.1 Implementation Considerations


Auditing must be designed based on the security policy
Only actions that can violate policy constraints need to be logged
Logs should include:
Subject (user/process)
Object (file/resource)
Action (read/write/execute)
Result (success/failure)
Initial system state must be recorded to detect violations accurately

👉 Example: In Bell–LaPadula model, log subject level, object level, action, and result to
Module 3_1
detect illegal reads/writes.

3.2 Syntactic Issues in Logging


Problems arise when logs:

Are ambiguous
Lack sufficient context
Use inconsistent formats

Solution: Grammar-Based Logging

Logs are defined using formal grammars (BNF)


Ensures:
Consistency
Unambiguous interpretation
Easy parsing by analyzers

👉 Grammar-based logs eliminate ambiguity and parsing errors. Module 3_1

3.3 Log Sanitization


Logs may contain confidential or private information.

Definition:

A log is sanitized when confidential information is removed according to the security


Module 3_1
policy.

Types of Sanitization:

1) Anonymizing Sanitization
Information is removed permanently
Cannot be reconstructed by anyone

2) Pseudonymizing Sanitization
Information is replaced or encrypted
Can be reconstructed later (e.g., court order)

Sanitization may be applied:

Before logging → protects user privacy


After logging → protects organizational confidentiality

3.4 Application Logging vs System Logging

Application Logging

High-level events
Human-readable
Examples:
Login failures
Email delivery errors

Advantages:

Easy to understand
Smaller log size

System Logging

Low-level kernel events


Includes:
System calls
File I/O
Memory access

Advantages:

Complete and precise


Helps trace root cause of failures

Best Practice

A well-designed auditing system uses both:

Application logs for context


System logs for detailed analysis

Module 3_1
Correlation of both logs helps identify security breaches accurately.

4. Conclusion (Exam-Friendly)
Logging records security-relevant events
Auditing analyzes logs to detect policy violations
An auditing system consists of logger, analyzer, and notifier
A well-unified auditing system requires:
Proper implementation
Clear syntax
Secure log sanitization
Coordinated application and system logging

If you want, I can:

Convert this into 6–8 mark answers


Provide diagram sketches for exams
Create short notes or VTU-style question answers

Just tell me 👍
Sources

Explain the posteriori design for an effective auditing


subsystem
.
● Auditing Types & Mechanisms:
○ Explain state-based auditing, transition-based auditing,
and auditing to detect known policy
violations
.
○ Briefly explain different auditing mechanisms and the six
basic audit browsing
techniques
.
○ Review examples of auditing file systems
.

Below is a clear, structured, VTU-exam–ready explanation of a posteriori auditing design,


auditing types & mechanisms, audit browsing techniques, and file system auditing
examples, exactly aligned with Module-3 (Auditing).

1. A Posteriori Design for an Effective Auditing Subsystem

Meaning of A Posteriori Design

A posteriori auditing refers to designing auditing mechanisms after the system is already
built and deployed, rather than integrating auditing during system design.

This approach is necessary because:

Many systems were not designed with security in mind


Most security breaches occur on existing systems
Security policies may be implicit or incomplete
Goals of A Posteriori Auditing

There are two main goals:

1. Detect violations of a known (stated) policy


2. Detect actions known to indicate security attacks, even if the policy is not explicitly
stated

Unlike a priori design, here the auditor must:

Study the existing system


Identify sensitive actions and states
Add logging and analysis mechanisms externally

Key Challenge

Not all violations are known in advance


Logs may be incomplete or inconsistent
Auditing must work without redesigning the system kernel

2. Auditing Types & Mechanisms

2.1 State-Based Auditing

Definition

State-based auditing examines the current state of the system to determine whether it
violates the security policy.

Focuses on what the system looks like now


Based on snapshots of system state

Characteristics

Uses state-based logging


Compares system state against policy rules
Does not focus on how the state was reached

Example

File system integrity checkers


Comparing current file permissions with a trusted database

Limitation

Difficult to obtain a consistent snapshot in active systems


Can suffer from time-of-check to time-of-use (TOCTTOU) problems

2.2 Transition-Based Auditing

Definition

Transition-based auditing analyzes actions (commands or events) that may move the
system from a secure to a nonsecure state.

Focuses on what is about to happen


Evaluates proposed transitions

Characteristics

Uses transition-based logging


Examines:
Current state
Proposed action
Prevents violations before they occur

Example

Detecting illegal privilege escalation


Blocking unauthorized network connections

Limitation

If the system starts in a nonsecure state, violations may go undetected


Often combined with state-based auditing

2.3 Auditing to Detect Known Policy Violations

Purpose

Used when:

Security policy is implicit


Certain behaviors are clearly dangerous

Approach

Identify known attack patterns


Log events related to those patterns
Report when conditions match attack signatures

Example

Detecting denial-of-service attacks


Detecting repeated failed login attempts
Detecting LAND attack (same source and destination IP and port)

Summary of Auditing Types

Type Focus Detects

State-Based System state Unauthorized states

Transition-Based Actions Unauthorized transitions

Known Violations Attack patterns Specific security attacks

3. Auditing Mechanisms
Auditing mechanisms vary based on system security design.

3.1 Auditing in Secure Systems


Auditing is built into system architecture
Logging occurs at multiple system layers
Supports:
Selective logging
Severity levels
Audit reduction
Features

High reliability
Minimal performance impact
System may halt if auditing fails

Example

VAX VMM system


Compartmented Mode Workstation (CMW)

3.2 Auditing in Nonsecure Systems


Auditing added as an external or add-on module
Primarily designed for accounting
Less detailed security data

Example

SunOS Basic Security Module (BSM)


Uses:
Binary logs
Audit tokens
Event classes
Tools like praudit and auditreduce used for analysis

4. Six Basic Audit Browsing Techniques


Audit browsing helps human analysts examine logs directly.

1. Text Display
Plain textual log viewing
Search by time, user, event

Limitation: No relationship visualization


2. Hypertext Display
Log entries linked using hyperlinks
Enables navigation between related events

Limitation: Local view only

3. Relational Database Browsing


Logs stored in databases
Analyst uses SQL-like queries

Advantage: Powerful correlations


Limitation: Requires preprocessing

4. Replay
Events replayed in temporal order
Shows how events unfolded over time

Advantage: Highlights timing relationships

5. Graphing
Nodes represent entities (processes, files)
Edges represent relationships

Advantage: Visual clarity


Limitation: Scalability issues for large logs

6. Slicing
Extracts minimum set of events affecting an object
Focused analysis

Limitation: Narrow scope


5. Examples of Auditing File Systems

5.1 Auditing NFS (Network File System)

Policy Goals

Only authorized clients access exported file systems


UNIX access controls enforced
No unauthorized device file creation

Logging Requirements

Record:
File handles
User IDs
Client host
Operation results
File names (for LOOKUP)

Auditing Checks

Verify file handle issued to correct user


Check access permissions
Ensure operations behave as local UNIX operations

This is a transition-based auditing mechanism.

5.2 Logging and Auditing File System (LAFS)

Key Idea

Auditing integrated at file system level


No need to modify applications

Features

Logs all file accesses


Uses policy language
Provides virtual files:

%log → access log

%policy → access rules


%audit → violation report

Advantage

Automatic detection of policy violations


Human-readable audit reports

6. Conclusion (Exam-Perfect)
A posteriori auditing is essential for existing systems
Three major auditing approaches:
State-based
Transition-based
Known policy violation detection
Auditing mechanisms differ for secure and nonsecure systems
Audit browsing tools assist human analysts
File system auditing (NFS, LAFS) illustrates practical application

If you want, I can:

Compress this into 10–12 mark answers


Create flow diagrams
Provide short notes / revision sheets
Give likely VTU exam questions

Just tell me 👍

● Intrusion Detection (IDS):


○ Explain the principles and the basic 4-fold goals of
intrusion detection systems
.
○ Explain IDS models: Anomaly, Misuse, and Specification
modeling
.
○ Describe the architecture (agent, director, notifier) and
organization of an IDS

Below is a clear, structured, VTU-exam–oriented answer for Intrusion Detection Systems


(IDS), aligned with Computer & Network Security (BEC714B) expectations.

1. Intrusion Detection Systems (IDS) – Principles & Goals

What is an Intrusion Detection System?


An Intrusion Detection System (IDS) is a security mechanism that monitors system or
network activities to detect attempts to violate security policies, compromise
confidentiality, integrity, or availability.

Unlike prevention mechanisms (firewalls, access control), an IDS focuses on detection,


analysis, and reporting of attacks.

Principles of Intrusion Detection


An IDS works on the following basic principles:

1. System activities can be monitored


User actions, system calls, network packets, file accesses, etc.
2. Attacks produce observable patterns
Intrusions cause abnormal behavior or recognizable misuse patterns.
3. Audit data is the primary source
IDS relies on logs, system events, and network traffic.
4. Policy-based evaluation
Activities are compared against a security policy or expected behavior.
5. Detection ≠ Prevention
IDS detects and reports; response may be manual or automated.

Basic 4-Fold Goals of Intrusion Detection Systems


An effective IDS has four fundamental goals:
1. Detect Intrusions

Identify:
Unauthorized access
Abuse of privileges
Malware and attacks
Detection may be real-time or post-event

2. Provide Accountability

Associate actions with:


Specific users
Processes
Hosts
Supports forensic analysis and legal evidence

3. Assist Damage Control and Recovery

Helps administrators:
Isolate compromised components
Limit damage
Restore system integrity

4. Deter Attacks

Presence of IDS discourages attackers


Increases perceived risk of detection

2. IDS Models
IDS models define how intrusions are detected. There are three major models:

2.1 Anomaly Detection Model


Concept

Builds a model of normal behavior


Flags deviations as possible intrusions

How it Works

1. Learn normal patterns:


CPU usage
Login frequency
Network traffic
2. Compare current behavior with baseline
3. Significant deviation → intrusion alert

Advantages

Can detect unknown (zero-day) attacks


Adaptive to system behavior

Disadvantages

High false positive rate


Difficult to define “normal” precisely

Example

User logs in at unusual hours and accesses large volumes of data

2.2 Misuse Detection Model (Signature-Based)

Concept

Detects intrusions by matching activity against known attack signatures

How it Works

Database of attack patterns (signatures)


Current activity compared with signatures
Match → intrusion detected

Advantages

Low false positives


Accurate for known attacks

Disadvantages

Cannot detect unknown attacks


Requires frequent signature updates

Example

Detecting buffer overflow exploits using known byte patterns

2.3 Specification-Based Detection Model

Concept

Uses formal specifications of correct behavior


Flags violations of specifications as intrusions

How it Works

Define allowed sequences of actions


Monitor system behavior
Any deviation from specification → alert

Advantages

Lower false positives than anomaly detection


Detects some unknown attacks

Disadvantages

Writing specifications is complex


Limited scalability

Example

A program attempts a system call sequence not permitted by its specification

Comparison of IDS Models

Model Basis Detects Unknown Attacks False Positives

Anomaly Normal behavior Yes High


Model Basis Detects Unknown Attacks False Positives

Misuse Attack signatures No Low

Specification Correct behavior Yes (limited) Medium

3. Architecture of an Intrusion Detection System


A typical IDS consists of three core components:

3.1 Agent (Sensor)

Role

Collects data from:


System logs
Audit trails
Network traffic
System calls

Characteristics

Runs on hosts or network devices


Lightweight and continuous
Sends data to the director

Examples

Host-based IDS agent


Network packet sniffer

3.2 Director (Analyzer / Manager)

Role

Central analysis unit of IDS


Correlates data from multiple agents
Applies detection models:
Anomaly
Misuse
Specification

Functions

Data aggregation
Pattern matching
Decision making
Intrusion classification

3.3 Notifier (Response Unit)

Role

Reports detected intrusions

Actions

Alert system administrator


Log the intrusion
Trigger automated response:
Disable account
Block IP address
Terminate process

IDS Architecture Flow

markdown

System / Network Events



Agent

Director

Notifier

Administrator / Response

4. Organization of IDS
IDS can be organized in different ways:

1. Centralized IDS
All data sent to a central director
Easy management
Single point of failure

2. Distributed IDS
Multiple agents and directors
Scalable and fault-tolerant
Complex coordination
Copy code

3. Hybrid IDS
Combination of centralized and distributed
Most practical and widely used

5. Conclusion (Exam-Ready)
IDS detects violations of security policies
Four goals: detect, account, recover, deter
Three IDS models:
Anomaly
Misuse
Specification
IDS architecture consists of:
Agent
Director
Notifier
IDS complements preventive security mechanisms
If you want, I can:

Convert this into 10–15 mark answers


Add neat exam diagrams
Provide short notes / comparison tables
Give previous VTU-type questions

Just tell me 👍

You might also like