首页> 外文会议>IEEE Conference on Application, Information and Network Security >Detecting Abuse of Domain Administrator Privilege Using Windows Event Log
【24h】

Detecting Abuse of Domain Administrator Privilege Using Windows Event Log

机译:使用Windows事件日志检测域管理员权限的滥用

获取原文

摘要

In an Advanced persistent threat(APT) attack, attackers that infiltrate into an organization network tend to stay inside the network until they are able to accomplish their final goal that may include exploiting sensitive information. When Active Directory is in place, attackers try to obtain a Domain Administrator account which has a privilege to control all users and files in the AD environment. There are several methods for attackers to abuse a legitimate Domain Administrator account. One is to exploit vulnerabilities on Active Directory such as CVE-2014-0317. The other is to steal credentials with password dump tools such as mimikatz. Moreover, attackers are likely to create a backdoor that disguises itself as a legitimate Domain Administrator account called a “Golden Ticket”, in order to obtain long-term administrative privilege. If an attacker abuses a legitimate Domain Administrator account, it is not easy to differentiate a legitimate access and an malicious access. In order to overcome this difficulty, several methods have already been proposed for detecting attacks against AD by analyzing Windows event logs. Each detection method is useful under specific conditions, however none of them cover the entire scope of multiple attacking methods. In this research, we clarify and evaluate the effectiveness of existing methods using a dataset, and propose a new detection algorithm with improved detection rate.
机译:在高级持久威胁(APT)攻击中,渗透到组织网络的攻击者往往留在网络内,直到它们能够完成可能包括利用敏感信息的最终目标。当Active Directory到位时,攻击者会尝试获取域管理员帐户,该帐户具有控制广告环境中的所有用户和文件的权限。攻击者有几种方法可以滥用合法的域名管理员帐户。一个是在CVE-2014-0317等活动目录上利用漏洞。另一个是用密码转储工具(如Mimikatz)窃取凭据。此外,攻击者可能会创建一个后门,以伪装为称为“金牌票”的合法域管理员帐户,以获得长期管理权限。如果攻击者滥用合法的域管理员帐户,则不易区分合法访问和恶意访问。为了克服这种困难,已经提出了几种方法通过分析Windows事件日志来检测针对AD的攻击。每个检测方法在特定条件下都很有用,但是它们都不涵盖多次攻击方法的整个范围。在这项研究中,我们阐明了使用数据集的现有方法的有效性,并提出了一种具有改进的检测率的新检测算法。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号