Execution Modes

LMDA Composer supports four execution modes, each designed for different LogicModule scripting scenarios. The selected mode determines how output is parsed and validated.

---

Freeform

General Scripting & Testing

Best for: Testing, debugging, and exploration without output validation.

Freeform mode runs your script without any output format requirements. Use it when:

• Testing API calls or data retrieval
• Debugging script logic step-by-step
• Running maintenance or utility scripts
• Exploring collector capabilities

// Freeform example - just print output
def devices = Snmp.walk("192.168.1.1", "1.3.6.1.2.1.1")
devices.each { oid, value ->
    println "${oid} = ${value}"
}
return 0

In Freeform mode:

• Parsed and Validation tabs are disabled
• Only Raw Output is available
• No format requirements apply

---

Active Discovery

Instance Discovery

Best for: Writing and testing Active Discovery (AD) scripts for DataSources.

Active Discovery scripts return instance data that LogicMonitor uses to create monitoring instances.

Output Format

Each line represents one discovered instance:

instance_id##instance_name
instance_id##instance_name##description
instance_id##instance_name##description####prop1=value1&prop2=value2

Field	Required	Max Length	Rules
Instance ID	Required	1024 chars	No spaces, =, :, \, or #
Instance Name	Required	255 chars	Display name for the instance
Description	Optional	—	Optional description text
Properties	Optional	—	Auto-properties as key=value&key=value

Example Script

// Discover network interfaces
def host = hostProps.get("system.hostname")
def interfaces = Snmp.walk(host, "1.3.6.1.2.1.2.2.1.2")

interfaces.each { oid, name ->
    def index = oid.tokenize('.')[-1]
    // Format: id##name##description####auto.props
    println "${index}##${name}##Interface ${index}####auto.interface.index=${index}"
}

return 0

Validation Checks

The Validation tab reports:

• Valid — Instances with all required fields
• Error — Invalid instance IDs (empty, too long, invalid characters)
• Warning — Instance names exceeding 255 characters
• Total instance count in the tab header

---

Collection

Single-Instance Metrics

Best for: Writing collection scripts for single-instance DataSources.

Collection scripts output datapoint values as simple key-value pairs:

datapoint_name=numeric_value

Example Script

// Collect CPU metrics
def host = hostProps.get("system.hostname")
def cpuOid = "1.3.6.1.2.1.25.3.3.1.2"
def cpuValues = Snmp.walk(host, cpuOid)

def total = 0
def count = 0
cpuValues.each { oid, value ->
    total += value.toInteger()
    count++
}

println "CpuUsage=${count > 0 ? total / count : 0}"
println "CpuCount=${count}"

return 0

Validation Rules

Rule	Severity
Must be key=value format	Error
Value must be numeric	Error
Invalid datapoint name characters	Warning
Non-matching lines ignored	Warning

---

Batch Collection

Multi-Instance Metrics

Best for: Writing batch collection scripts that handle multiple instances in a single execution.

Batch scripts output data for all instances at once, prefixed with the instance ID (wildvalue):

wildvalue.datapoint=numeric_value

Example Script

// Batch collect for all interfaces
def host = hostProps.get("system.hostname")
def instances = datasourceinstanceProps.keySet()

instances.each { wildvalue ->
    def inOctets = Snmp.get(host, "1.3.6.1.2.1.2.2.1.10.${wildvalue}")
    def outOctets = Snmp.get(host, "1.3.6.1.2.1.2.2.1.16.${wildvalue}")

    println "${wildvalue}.InOctets=${inOctets}"
    println "${wildvalue}.OutOctets=${outOctets}"
}

return 0

Execution Context

When running in Collection or Batch Collection mode, a dialog prompts for additional context:

• Collection Mode: Enter the wildvalue (instance ID) to simulate
• Batch Mode: Enter the datasourceId to load all instance properties

This populates instanceProps (Collection) or datasourceinstanceProps (Batch) for your script.

Validation Rules

Same rules as Collection mode, plus:

• Wildvalue prefix is required on each line
• Wildvalue follows instance ID rules (no spaces, =, :, \, #)
• Maximum 1024 characters for wildvalue

---

Choosing the Right Mode

By Task

Task	Recommended Mode
Testing SNMP/WMI queries	Freeform
Building discovery logic	Active Discovery
Single-instance DataSource	Collection
Multi-instance DataSource	Batch Collection
PropertySource script	Freeform
Debugging any script	Freeform

By Module Type

Module Type	Primary Mode
DataSource (AD script)	Active Discovery
DataSource (Collect script)	Collection or Batch
ConfigSource	Collection
PropertySource	Freeform
EventSource	Freeform
TopologySource	Freeform (Graph tab)

---

Mode Indicators

The current mode is displayed in multiple locations:

• Toolbar — Mode selector dropdown shows the active mode with its icon
• Status Bar — Mode badge displayed at the bottom of the editor window
• Tab Bar — Each tab shows its mode; switching tabs switches the active mode context

Per-Tab Modes

Each tab remembers its own mode setting. Work on AD and Collection scripts simultaneously without switching modes manually.

---

Next Steps

Output Panel Learn how to interpret parsed output and validation results.