Do not use decoy files as the sole ransomware defense. They are a detection mechanism, not a prevention mechanism, and should complement backups, EDR, and access controls.
watchdog library for cross-platform file system monitoringPlan file placement for maximum detection coverage:
Canary File Placement Strategy:
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Naming Convention:
- Use names that sort FIRST and LAST alphabetically in each directory
- Ransomware typically enumerates directories A-Z or Z-A
- Examples: _AAAA_budget_2024.docx, ~zzzz_report_final.xlsx
Placement Locations:
- Root of every file share (\\server\share\_AAAA_canary.docx)
- Desktop, Documents, Downloads on each endpoint
- Department-specific shares (Finance, HR, Legal)
- Backup staging directories
- Home directories of high-privilege accounts
File Types:
- .docx, .xlsx, .pdf (most targeted by ransomware)
- .sql, .bak (database files, high value)
- Mix of file types to detect ransomware that targets specific extensions
Create decoy files with realistic content and metadata:
import os
import time
def create_canary_docx(filepath, content="Q4 Financial Summary - Confidential"):
"""Create a realistic .docx canary file using python-docx."""
from docx import Document
doc = Document()
doc.add_heading("Financial Report - CONFIDENTIAL", level=1)
doc.add_paragraph(content)
doc.add_paragraph(f"Generated: {time.strftime('%Y-%m-%d')}")
doc.save(filepath)
def create_canary_txt(filepath):
"""Create a simple text canary with known content for hash verification."""
content = "CANARY_TOKEN_DO_NOT_MODIFY\n"
content += f"Created: {time.strftime('%Y-%m-%dT%H:%M:%S')}\n"
content += "This file is monitored for unauthorized changes.\n"
with open(filepath, "w") as f:
f.write(content)
Monitor canary files for any modification, rename, or deletion:
from watchdog.observers import Observer
from watchdog.events import FileSystemEventHandler
class CanaryHandler(FileSystemEventHandler):
def __init__(self, canary_paths, alert_callback):
self.canary_paths = set(canary_paths)
self.alert_callback = alert_callback
def on_modified(self, event):
if event.src_path in self.canary_paths:
self.alert_callback("MODIFIED", event.src_path)
def on_deleted(self, event):
if event.src_path in self.canary_paths:
self.alert_callback("DELETED", event.src_path)
def on_moved(self, event):
if event.src_path in self.canary_paths:
self.alert_callback("RENAMED", event.src_path)
Define automated responses when canary files are triggered:
Alert Response Matrix:
━━━━━━━━━━━━━━━━━━━━━
Event: Canary MODIFIED
→ Severity: CRITICAL
→ Action: Alert SOC, identify modifying process (PID), isolate endpoint
Event: Canary DELETED
→ Severity: HIGH
→ Action: Alert SOC, check for ransomware note in same directory
Event: Canary RENAMED (new extension added)
→ Severity: CRITICAL
→ Action: Alert SOC, check extension against known ransomware extensions
→ Automated: Kill modifying process, disable network interface
Event: Multiple canaries triggered within 60 seconds
→ Severity: EMERGENCY
→ Action: Network-wide isolation, activate incident response plan
Test that canary files detect actual ransomware behavior:
# Simulate ransomware encryption (safe test - modifies canary content)
echo "ENCRYPTED_BY_TEST" > /path/to/canary/_AAAA_budget.docx
# Simulate ransomware rename (adds extension)
mv /path/to/canary/report.xlsx /path/to/canary/report.xlsx.locked
# Verify alerts were generated in SIEM/alerting system
| Term | Definition |
|---|---|
| Canary File | A decoy file placed in a directory that is monitored for any access or modification, serving as a tripwire for unauthorized activity |
| Honeytoken | A broader category of deception artifacts (files, credentials, database records) designed to alert when accessed |
| File Integrity Monitoring | Continuous monitoring of file attributes (hash, size, permissions, timestamps) to detect unauthorized changes |
| ReadDirectoryChangesW | Windows API for monitoring file system changes in a directory; used by the watchdog library on Windows |
| inotify | Linux kernel subsystem for monitoring file system events; provides near-instant notification of file changes |