seclab/MOD4.5_SIEM_Operations.md

FILE: MOD4.5_SIEM_Operations.md

MODULE 4.5: SIEM OPERATIONS & LOG ANALYSIS

Learning Objectives

• Master KQL (Kibana Query Language) for Security Onion
• Build custom dashboards for threat visualization
• Perform log correlation across multiple data sources
• Tune alerts to reduce false positives
• Export data for integration with custom SOC dashboard (React)

---

LAB 4.5.1: KIBANA QUERY LANGUAGE (KQL) FUNDAMENTALS

Basic KQL Syntax

# Search for specific field value
event.module: "suricata"
source.ip: "10.10.2.50"

# Multiple conditions (AND)
event.module: "suricata" AND alert.signature: "GPL EXPLOIT"

# OR condition
source.ip: ("10.10.2.50" OR "10.10.2.51")

# Wildcards
alert.signature: *nmap*

# NOT operator
event.module: "suricata" NOT alert.severity: 3

# Field exists
_exists_: alert.signature

# Range queries
event.duration: >=1000000
@timestamp: >= "2026-02-11T00:00:00"

# Nested field
dns.question.name: "evil.com"

Common SOC Queries

Find failed SSH logins:

event.dataset: "system.auth" AND system.auth.ssh.event: "Failed"

Detect port scans (high SYN packet count):

event.module: "suricata" AND network.transport: "tcp" AND suricata.eve.flow.pkts_toserver: >100

Find DNS queries to suspicious domains:

dns.question.name: (*malware* OR *phishing* OR *bad*)

High-severity alerts:

alert.severity: (1 OR 2) AND event.module: "suricata"

---

LAB 4.5.2: BUILDING CUSTOM DASHBOARDS

Create Attack Detection Dashboard

1. Navigate to: Kibana > Dashboard > Create new dashboard

2. Add Visualization: Top Attack Signatures
   - Click "Create visualization"
   - Type: Vertical bar chart
   - Index pattern: so-*
   - Metrics: Count
   - Buckets: Terms → alert.signature.keyword
   - Top 10
   - Save: "Top Attack Signatures"

3. Add Visualization: Alert Timeline
   - Type: Line chart
   - X-axis: Date Histogram → @timestamp
   - Y-axis: Count
   - Save: "Alert Timeline"

4. Add Visualization: Source IP Geo Map
   - Type: Maps
   - Add layer: Documents
   - Index: so-*
   - Geospatial field: source.geo.location
   - Save: "Attack Source Map"

5. Add Visualization: Alert Severity Breakdown
   - Type: Pie chart
   - Slice: Terms → alert.severity
   - Save: "Severity Distribution"

6. Save Dashboard: "Apophis SOC - Attack Overview"

---

LAB 4.5.3: DASHBOARD INTEGRATION WITH REACT SOC

Export Data for Mock Dashboard

# Query Security Onion for last 24h alerts
# In Kibana Dev Tools:

POST /so-*/_search
{
  "size": 100,
  "query": {
    "bool": {
      "must": [
        { "range": { "@timestamp": { "gte": "now-24h" } } }
      ]
    }
  },
  "sort": [
    { "@timestamp": { "order": "desc" } }
  ]
}

# Export to JSON, then process for React dashboard:

# On Security Onion, create export script:
cat > /home/analyst/export_dashboard_data.sh << 'EOF'
#!/bin/bash
# Export SOC data for dashboard integration

# Export alert counts
curl -X GET "localhost:9200/so-*/_search?pretty" -H 'Content-Type: application/json' -d'
{
  "size": 0,
  "aggs": {
    "severity_count": {
      "terms": { "field": "alert.severity" }
    }
  }
}
' > /tmp/severity_data.json

# Export top threats
curl -X GET "localhost:9200/so-*/_search?pretty" -H 'Content-Type: application/json' -d'
{
  "size": 0,
  "aggs": {
    "top_signatures": {
      "terms": { "field": "alert.signature.keyword", "size": 10 }
    }
  }
}
' > /tmp/top_threats.json

echo "Data exported to /tmp/"
EOF

chmod +x /home/analyst/export_dashboard_data.sh
./export_dashboard_data.sh

# Transfer to dashboard development environment
scp /tmp/*.json user@dashboard-host:~/seclab/dashboard/src/data/live/

Convert to React Mock Data Format

// In dashboard/src/data/mockData.js, add real data integration:

export const importSecurityOnionData = async () => {
  // In production, this would fetch from Security Onion API
  // For now, use exported JSON files

  const severityData = {
    critical: 45,
    high: 128,
    medium: 312,
    low: 891
  };

  const topThreats = [
    { ip: "203.0.113.50", count: 1247, country: "CN", severity: "critical" },
    { ip: "198.51.100.23", count: 892, country: "RU", severity: "high" },
    // ... from exported data
  ];

  return { severityData, topThreats };
};

---

LAB 4.5.4: ALERT TUNING & FALSE POSITIVE REDUCTION

Identify Noisy Rules

# In Kibana, find most frequent alerts:
POST /so-*/_search
{
  "size": 0,
  "aggs": {
    "signatures": {
      "terms": {
        "field": "alert.signature.keyword",
        "size": 20,
        "order": { "_count": "desc" }
      }
    }
  }
}

# Common false positives:
# - Internal vulnerability scanners (Nessus, OpenVAS)
# - Legitimate admin tools (Nmap from IT)
# - Noisy signatures (DNS ANY query, etc.)

Suppress False Positives

# Edit Suricata threshold.config
sudo nano /etc/suricata/threshold.config

# Suppress specific signature for specific IP
suppress gen_id 1, sig_id 2100498, track by_src, ip 10.10.3.50
# Signature 2100498 from source 10.10.3.50 will be suppressed

# Rate limit alerts (max 5 per 60 seconds)
threshold gen_id 1, sig_id 2010937, type threshold, track by_src, count 5, seconds 60

# Restart Suricata
sudo so-suricata-restart

---

LAB 4.5.5: LOG CORRELATION

Multi-Source Event Correlation

Scenario: Detect full attack chain (scan → exploit → command execution)

# Step 1: Find nmap scan in Suricata
alert.signature: "GPL SCAN nmap*" AND source.ip: "10.10.2.50"
# Note timestamp: 2026-02-11T14:30:00

# Step 2: Find exploit attempt within 10 minutes
@timestamp: ["2026-02-11T14:30:00" TO "2026-02-11T14:40:00"]
AND alert.signature: *exploit*
AND destination.ip: "10.10.4.10"

# Step 3: Find Zeek connection logs showing data transfer
event.dataset: "zeek.conn"
AND source.ip: "10.10.2.50"
AND destination.ip: "10.10.4.10"
AND zeek.conn.orig_bytes: >10000
AND @timestamp: ["2026-02-11T14:35:00" TO "2026-02-11T14:45:00"]

# Step 4: Build correlation rule (in Kibana detection rules)
{
  "name": "Attack Chain: Scan to Exploit",
  "description": "Detects nmap scan followed by exploit attempt",
  "index": ["so-*"],
  "query": "sequence by source.ip with maxspan=15m
    [alert.signature: \"GPL SCAN nmap*\"]
    [alert.signature: *exploit*]"
}

---

##DELIVERABLES

• Custom Kibana dashboard with 5+ visualizations
• KQL query library (saved searches for common threats)
• Alert tuning configuration (threshold.config)
• Correlation rule for attack chain detection
• Data export script for React dashboard integration

---

END OF MODULE 4.5

Proceed to MOD5: Active Directory Threat Emulation after mastering SIEM query skills.