performing-cloud-storage-forensic-acquisition

name	performing-cloud-storage-forensic-acquisition
description	通过收集 API 远程数据和端点设备本地同步客户端制品，对 Google Drive、OneDrive、Dropbox 和 Box 等云存储服务执行取证获取和分析。
domain	cybersecurity
subdomain	digital-forensics
tags	["cloud-forensics","google-drive","onedrive","dropbox","box","cloud-acquisition","api-forensics","sync-client","endpoint-artifacts","magnet-axiom"]
version	1.0
author	mahipal
license	Apache-2.0

执行云存储取证获取

概述

云存储取证获取（Cloud storage forensic acquisition）涉及通过 API 远程获取和本地端点制品分析两种方式，从 Google Drive、OneDrive、Dropbox 和 Box 等服务收集数字证据。现代调查必须应对一个挑战：云同步文件可能存在多种状态——本地已同步、仅云端（按需下载）、已缓存和已删除。与云存储同步过的端点设备包含大量元数据，涵盖本地同步文件、仅存于云端的文件，甚至可从缓存文件夹恢复的已删除项目。使用特定服务 API 进行基于 API 的获取，在拥有有效凭据和适当法律授权的情况下，可直接访问远程数据。

前置条件

• 访问云数据的法律授权（搜查令、同意书或企业政策）
• 有效的用户凭据或管理员访问令牌
• Magnet AXIOM Cloud、Cellebrite Cloud Analyzer 或同等工具
• 带有云存储目标文件的 KAPE
• Python 3.8+ 及 google-api-python-client、msal、dropbox SDK
• 用于 API 获取的网络连接

获取方法

方法 1：基于 API 的远程获取

Google Drive API 获取

from google.oauth2.credentials import Credentials
from googleapiclient.discovery import build
from googleapiclient.http import MediaIoBaseDownload
import io
import os
import json
from datetime import datetime


class GoogleDriveForensicAcquisition:
    """通过 API 对 Google Drive 文件和元数据进行取证获取。"""

    def __init__(self, credentials_path: str, output_dir: str):
        self.creds = Credentials.from_authorized_user_file(credentials_path)
        self.service = build("drive", "v3", credentials=self.creds)
        self.output_dir = output_dir
        os.makedirs(output_dir, exist_ok=True)
        self.acquisition_log = []

    def list_all_files(self, include_trashed: bool = True) -> list:
        """列出所有文件，包括回收站中的项目。"""
        files = []
        page_token = None
        query = "" if include_trashed else "trashed = false"

        while True:
            results = self.service.files().list(
                q=query,
                pageSize=1000,
                fields=
                       
                       
                       ,
                pageToken=page_token
            ).execute()

            files.extend(results.get(, []))
            page_token = results.get()
              page_token:
                

         files

     () -> :
        
        output_path = os.path.join(.output_dir, file_name)

         mime_type.startswith():
            export_formats = {
                : ,
                : ,
                : ,
            }
            export_mime = export_formats.get(mime_type, )
            request = .service.files().export_media(fileId=file_id, mimeType=export_mime)
        :
            request = .service.files().get_media(fileId=file_id)

         io.FileIO(output_path, )  fh:
            downloader = MediaIoBaseDownload(fh, request)
            done = 
              done:
                _, done = downloader.next_chunk()

        .acquisition_log.append({
            : datetime.utcnow().isoformat(),
            : file_id,
            : file_name,
            : output_path,
            : 
        })
         output_path

     () -> :
        
        revisions = .service.revisions().(
            fileId=file_id,
            fields=
        ).execute()
         revisions.get(, [])

     () -> :
        
        report_path = os.path.join(.output_dir, )
         (report_path, )  f:
            json.dump({
                : .acquisition_log[][]  .acquisition_log  ,
                : datetime.utcnow().isoformat(),
                : (.acquisition_log),
                : .acquisition_log
            }, f, indent=)
         report_path

OneDrive / Microsoft 365 API 获取

import msal
import requests
import os
import json
from datetime import datetime


class OneDriveForensicAcquisition:
    """通过 Microsoft Graph API 对 OneDrive 文件和元数据进行取证获取。"""

    def __init__(self, client_id: str, tenant_id: str, client_secret: str, output_dir: str):
        self.output_dir = output_dir
        os.makedirs(output_dir, exist_ok=True)

        authority = f"https://login.microsoftonline.com/{tenant_id}"
        self.app = msal.ConfidentialClientApplication(
            client_id, authority=authority, client_credential=client_secret
        )
        token_result = self.app.acquire_token_for_client(
            scopes=["https://graph.microsoft.com/.default"]
        )
        self.access_token = token_result.get("access_token")
        self.headers = {"Authorization": f"Bearer {self.access_token}"}
        self.base_url = "https://graph.microsoft.com/v1.0"

    def list_user_files(self, user_id: str) -> list:
        """列出用户 OneDrive 中的所有文件。"""
        url = f"{self.base_url}/users/{user_id}/drive/root/children"
        files = []
        while url:
            response = requests.get(url, headers=.headers)
            data = response.json()
            files.extend(data.get(, []))
            url = data.get()
         files

     () -> :
        
        url = 
        response = requests.get(url, headers=.headers, stream=)
        output_path = os.path.join(.output_dir, filename)
         (output_path, )  f:
             chunk  response.iter_content(chunk_size=):
                f.write(chunk)
         output_path

     () -> :
        
        url = 
        response = requests.get(url, headers=.headers)
         response.json().get(, [])

方法 2：本地端点制品收集

KAPE 云存储目标

# 使用 KAPE 收集所有云存储制品
kape.exe --tsource C: --tdest C:\Output\CloudArtifacts --target GoogleDrive,OneDrive,Dropbox,Box

# OneDrive 制品位置
# %USERPROFILE%\AppData\Local\Microsoft\OneDrive\logs\
# %USERPROFILE%\AppData\Local\Microsoft\OneDrive\settings\
# %USERPROFILE%\OneDrive\

# Google Drive 制品位置
# %USERPROFILE%\AppData\Local\Google\DriveFS\
# 包含元数据 SQLite 数据库和缓存文件

# Dropbox 制品位置
# %USERPROFILE%\AppData\Local\Dropbox\
# %USERPROFILE%\Dropbox\.dropbox.cache\
# 包含 filecache.dbx（加密 SQLite）、host.dbx、config.dbx

OneDrive 本地数据库分析

import sqlite3
import os

def analyze_onedrive_sync_engine(db_path: str) -> list:
    """分析 OneDrive SyncEngineDatabase 以获取文件元数据。"""
    conn = sqlite3.connect(db_path)
    cursor = conn.cursor()

    # 查询所有被追踪的文件，包括仅云端项目
    cursor.execute("""
        SELECT fileName, fileSize, lastChange,
               resourceID, parentResourceID, eTag
        FROM od_ClientFile_Records
        ORDER BY lastChange DESC
    """)

    files = []
    for row in cursor.fetchall():
        files.append({
            "filename": row[0],
            "size": row[1],
            "last_change": row[2],
            "resource_id": row[3],
            "parent_id": row[4],
            "etag": row[5]
        })

    conn.close()
    return files

云存储制品汇总

服务	本地数据库	缓存位置	日志文件
OneDrive	SyncEngineDatabase.db	%LOCALAPPDATA%\Microsoft\OneDrive\cache\	%LOCALAPPDATA%\Microsoft\OneDrive\logs\
Google Drive	metadata_sqlite_db	%LOCALAPPDATA%\Google\DriveFS{account}\content_cache\	%LOCALAPPDATA%\Google\DriveFS\Logs\
Dropbox	filecache.dbx（加密）	%APPDATA%\Dropbox.dropbox.cache\	%APPDATA%\Dropbox\logs\
Box	sync_db	%LOCALAPPDATA%\Box\Box\cache\	%LOCALAPPDATA%\Box\Box\logs\

参考资料

• SANS 云存储获取: https://www.sans.org/blog/cloud-storage-acquisition-from-endpoint-devices
• Magnet AXIOM Cloud: https://www.magnetforensics.com/blog/how-to-acquire-and-analyze-cloud-data-with-magnet-axiom/
• AWS 云取证框架: https://docs.aws.amazon.com/prescriptive-guidance/latest/security-reference-architecture/cyber-forensics.html
• 云驱动器基于 API 的取证获取: https://arxiv.org/abs/1603.06542