idp-ics
IDS/IPS Overview​
IDS = Intrusion Detection System IPS = Intrusion Prevention System
These are security technologies that monitor network traffic and system activities for malicious behavior and policy violations.
Key Differences​
Aspect | IDS (Detection) | IPS (Prevention) |
---|---|---|
Primary Function | Monitors and Alerts | Monitors and Blocks |
Response | Passive - notifies about threats | Active - stops threats in real-time |
Network Position | Out-of-band (parallel to traffic flow) | In-line (traffic flows through it) |
Impact on Performance | Minimal - doesn't affect traffic flow | Can add latency - all traffic passes through |
False Positives | Alerts only - no service disruption | Can block legitimate traffic |
How They Work​
IDS (Intrusion Detection System)​
Internet → Firewall → Router → [Traffic Copy] → IDS → Alert/Log
↓
Servers (traffic continues normally)
IDS Process:
- Monitors network traffic passively
- Analyzes patterns against known attack signatures
- Generates alerts when suspicious activity detected
- Logs incidents for forensic analysis
- Does NOT block traffic
IPS (Intrusion Prevention System)​
Internet → Firewall → IPS → Router → Servers
↑
(Blocks malicious traffic)
IPS Process:
- All network traffic flows through IPS
- Real-time analysis of packets
- Compares against threat signatures and behavioral patterns
- Automatically blocks malicious traffic
- Allows legitimate traffic to continue
Detection Methods​
1. Signature-Based Detection​
- Matches traffic against database of known attack patterns
- Example signatures:
SQL Injection: "SELECT * FROM users WHERE id='1' OR '1'='1'"
Buffer Overflow: Specific byte patterns that exploit vulnerabilities
Malware Communication: Known command-and-control server communications
2. Anomaly-Based Detection​
- Establishes baseline of "normal" network behavior
- Detects deviations from normal patterns
- Examples:
- Unusual data transfer volumes
- Connections to suspicious geographic locations
- Abnormal login patterns (time, frequency, location)
- Unexpected protocol usage
3. Behavioral Analysis​
- Monitors user and system behavior over time
- Detects insider threats and advanced persistent threats (APTs)