Security Terms/Detection

Intrusion Detection System (IDS)

ayosecu 2025. 1. 16. 15:25

Intrusion Detection System (IDS)

An IDS is designed to monitor a network or system for abnormal activities and potential intrusions or attacks. There are two main methods of intrusion detection: Signature-Based Detection and Behavior-Based Detection.

  • Signature-Based IDS:
    • Signature-based detection identifies known attack patterns. A signature refers to a rule or pattern that defines the unique characteristics of an attack, which is stored in a database. The IDS checks whether events or traffic in the system match any known signatures.
    • Advantages: It can detect known attacks quickly and accurately.
    • Disadvantages: It struggles to detect new or zero-day attacks. If an attacker can bypass the signature, the system won’t detect the threat.
    • Examples: Snort and Suricata are well-known signature-based IDS tools that analyze network traffic based on predefined patterns.
  • Behavior-Based IDS:
    • Behavior-based detection learns what normal system behavior looks like and detects deviations from these norms. It monitors real-time activity on the network or system and flags any anomalies as potential threats.
    • Advantages: It can detect new types of attacks by recognizing abnormal behavior patterns. It’s also useful for detecting zero-day attacks.
    • Disadvantages: It may generate many false positives since not all abnormal behavior is malicious.

Snort/Suricata/YARA Rule Writing:

For IDS tools like Snort and Suricata, creating detection rules is essential for defining specific patterns of behavior or signatures to identify threats.

  • Snort/Suricata Rule Writing:

    • Snort and Suricata are signature-based network IDS tools. Rules written for these systems define patterns to be detected in network packets, such as certain strings, ports, or protocols.
    • Example rule:
      alert tcp any any -> 192.168.1.100 80 (msg:"Possible HTTP Attack"; content:"/cmd.exe"; sid:1001;)
      This rule triggers an alert if any packet directed to IP 192.168.1.100 on port 80 contains the string /cmd.exe, which is common in certain attack types.
      Writing effective rules requires a solid understanding of attack scenarios and how to define the signatures of these attacks.
  • YARA Rule Writing:

    • YARA is a tool designed to identify malware by writing rules that look for specific patterns in files, processes, or memory. These rules can include strings, byte patterns, and other conditions.
    • Example rule:
      rule MyMalware
      {
      strings:
      $a = "malicious_string"
      $b = { 6A 40 68 00 30 00 00 6A 14 8D 91 }
      

    condition:
    $a or $b
    }

    This rule detects a file or memory pattern containing the defined string or byte sequence, identifying potential malware.
    

Host-based Intrusion Detection System (HIDS):

A HIDS focuses on monitoring and analyzing the activities on individual hosts (like servers or PCs), unlike network-based IDS that monitors network traffic. HIDS typically examines log files, checks file integrity, and monitors system calls to detect intrusions.

  • OSSEC:
    • OSSEC is an open-source HIDS that performs real-time log analysis, file integrity monitoring, rootkit detection, and alerting. It tracks changes in critical files and directories, ensuring their integrity.
    • Key Features:
      • Log Analysis: OSSEC analyzes logs from various operating systems to detect attacks.
      • File Integrity Monitoring: It tracks changes to important files, alerting if unauthorized modifications occur.
      • Rootkit Detection: It checks for modifications or tampering with core system files and processes.

HIDS provides detailed monitoring at the host level and can be paired with network-based IDS for a more comprehensive security strategy.

'Security Terms > Detection' 카테고리의 다른 글

Anomaly or Behavior-based Detection  (0) 2025.01.16
Signatures  (0) 2025.01.16
Security Signals  (0) 2025.01.16
IOC (Indicator of Compromise)  (0) 2025.01.16
Security Information and Event Management (SIEM)  (0) 2025.01.16