Automations
LogZilla documentation for Automations
LogZilla's automation capabilities provide Security Orchestration, Automation, and Response (SOAR) functionality that enables organizations to respond to events automatically, reducing manual intervention and improving response times. Users can create automated workflows that execute scripts, send webhooks, generate notifications, and orchestrate responses across multiple systems.
How Automation Works
LogZilla automation operates through triggers that execute actions when specific events occur. When a trigger matches an event, it can:
- Execute custom scripts with full event context
- Send HTTP webhooks to external systems
- Generate email notifications with event details
- Create internal notifications and alerts
- Mark events for classification and filtering
All automation actions receive complete event information through environment variables, enabling context-aware responses.
Practical Automation Examples
The following examples demonstrate real-world automation scenarios users can implement:
Network Interface Recovery
Scenario: A Cisco interface goes down unexpectedly.
Automation: A trigger detects the %LINK-3-UPDOWN
event and executes a
Python script that:
- Uses python to SSH into the device
- Runs
no shutdown
on the affected interface - Sends a Slack notification confirming the remediation
Implementation: The script receives the device hostname via the
EVENT_HOST
environment variable and interface details parsed from
EVENT_MESSAGE
.
Duplex Mismatch Correction
Scenario: A CDP duplex mismatch is detected on a network link.
Automation: When LogZilla receives a %CDP-4-DUPLEX_MISMATCH
event, a
trigger executes a script that:
- Connects to both devices involved in the mismatch
- Configures both interfaces to use the same duplex setting
- Logs the corrective action taken
Service Restart with Threshold
Scenario: A critical service fails and needs automatic recovery with escalation limits.
Automation: Using event correlation rules, the system:
- Detects service failure events
- Attempts to restart the service automatically
- Tracks restart attempts (maximum 3 times in 5 minutes)
- Escalates to human operators if automatic recovery fails
Flapping Interface Detection
Scenario: A network interface is unstable, repeatedly going up and down.
Automation: Event correlation identifies the pattern:
- Correlates interface down/up event pairs within a time window
- Counts "short outage" occurrences
- Generates an alert when flapping exceeds threshold (e.g., 10 times in 6 hours)
- Notifies network team of unstable interface requiring investigation
Security Incident Response
Scenario: Multiple failed login attempts detected from a single source.
Automation: A trigger identifies the pattern and:
- Automatically disables the affected user account
- Creates a security incident ticket in ServiceNow
- Sends alert to security team with user and source IP details
- Blocks the source IP at the firewall if attempts continue
Malware Communication Detection
Scenario: A host makes numerous outbound connections to suspicious destinations.
Automation: When firewall logs show high-volume connection attempts:
- Cross-reference destination IPs with threat intelligence feeds
- Automatically block confirmed malicious IPs at the firewall
- Quarantine the source host via NAC integration
- Create high-priority security incident for investigation
ServiceNow Integration
Scenario: Automatically create and update incident tickets for critical events.
Automation: A trigger sends HTTP POST requests to ServiceNow's REST API:
- Creates incident tickets with event details
- Uses correlation IDs to prevent duplicate tickets
- Updates existing tickets when related events occur
- Includes event context, severity, and affected systems
Implementation: The webhook payload includes LogZilla variables like
{{event:host}}
, {{event:message}}
, and {{event:severity}}
to populate
ServiceNow fields automatically.
Interactive Slack Automation
Scenario: Network interface down with human-in-the-loop remediation.
Automation: When an interface goes down:
- Sends Slack message with interface details and "Remediate" button
- Operator clicks button to authorize automatic remediation
- LogZilla receives postback and executes interface recovery script
- Sends confirmation message with remediation results
Implementation Methods
Script Execution
Scripts receive complete event context through environment variables:
bashEVENT_HOST=router01.company.com EVENT_MESSAGE=%LINK-3-UPDOWN: Interface GigabitEthernet0/1, changed state to down EVENT_SEVERITY=3 EVENT_FACILITY=23 EVENT_PROGRAM=LINK
Scripts can use this information to make intelligent, context-aware decisions about remediation actions.
Event Correlation
Complex automation scenarios use the Simple Event Correlator (SEC) to:
- PairWithWindow: Match related events within time windows
- SingleWithThreshold: Count occurrences and trigger on thresholds
- EventGroup: Correlate multiple event types for complex patterns
Webhook Integration
Send structured data to external APIs using HTTP POST requests with JSON payloads containing event variables and custom fields.
LogZilla SOAR Advantages
LogZilla has provided SOAR capabilities for years before the term "Security Orchestration, Automation, and Response" became widely adopted. This established foundation offers several advantages:
- Proven Reliability: Battle-tested automation engine with years of production deployment
- High Performance: Purpose-built for high-volume event processing and real-time response
- Flexible Integration: Extensive API and webhook support for security tool orchestration
- Cost Effective: Comprehensive SOAR functionality without the complexity and licensing costs of dedicated SOAR platforms
- Unified Platform: Combines log management, SIEM capabilities, and SOAR automation in a single solution
Getting Started
1. Identify Automation Opportunities
Look for repetitive manual tasks triggered by specific events:
- Service restarts after failures
- Network device configuration issues
- Security incidents requiring immediate response
- Ticket creation for critical alerts
2. Start Simple
Begin with basic automations before building complex workflows:
- Email notifications for critical events
- Slack alerts with event details
- Simple script execution for common fixes
3. Build Context-Aware Scripts
Use environment variables to create intelligent automations:
EVENT_HOST
for device-specific actionsEVENT_MESSAGE
for parsing specific detailsEVENT_SEVERITY
for response prioritizationEVENT_USER_TAGS
for enriched context
4. Implement Safeguards
Include error handling and limits in automation scripts:
- Maximum retry attempts with backoff
- Validation before making changes
- Logging of all automated actions
- Escalation paths when automation fails
Advanced Scenarios
Domain Trust Enumeration Detection
Detect reconnaissance activity by correlating events from multiple domain controllers when a single source attempts to enumerate domain trusts across several systems within a short timeframe.
Windows Event Log Service Monitoring
Monitor for Windows Event Log service failures and automatically restart the service, escalating to administrators if the service fails to stay running.
Technical Implementation
For detailed technical information on implementing automation scripts, including environment variables, script requirements, and Docker container setup, see the Trigger Scripts documentation.
For complex event correlation scenarios, see the Event Correlation documentation.