{"title":"System Hardening for Infrastructure as a Service (IaaS)","authors":"T. Rose, Xiaobo Zhou","doi":"10.1109/SSS47320.2020.9174202","DOIUrl":null,"url":null,"abstract":"Reduce the attack vector.! Minimize risk.! When discussing the topic of system hardening, the goal is to lock down the system to make it more difficult for a hacker to break in. There exist a number of hardening standards for a variety of applications and products, but perhaps the first stop should be the operating system itself. We examine system hardening in a cloud environment to assess if standard operating system hardening benchmarks are applicable in a cloud virtual machine.Five case studies of large-scale security breaches are analyzed to determine which system hardening benchmarks would have mitigated and possibly prevented the various attacks. These examples serve as a reminder of the importance of system hardening in general, and the IaaS consumer must protect the portion of the environment under their control and responsibility. Two industry standard benchmarks (CIS and STIG) are discussed, and STIG is implemented on an AWS EC2 instance running Red Hat Enterprise Linux 7.After evaluating 207 scripted benchmarks, we recommend and pass 195, leaving only 12 failed items which form the basis of discussion for this paper. These failed items include five false positives, one known bug and two exemptions, subsequently raising our compliance score to 98%. In a real-world audit, the remaining four failed items would be scrutinized for compensating controls or strong justification for a process exemption.","PeriodicalId":210774,"journal":{"name":"2020 IEEE Systems Security Symposium (SSS)","volume":"19 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2020-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2020 IEEE Systems Security Symposium (SSS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/SSS47320.2020.9174202","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
Reduce the attack vector.! Minimize risk.! When discussing the topic of system hardening, the goal is to lock down the system to make it more difficult for a hacker to break in. There exist a number of hardening standards for a variety of applications and products, but perhaps the first stop should be the operating system itself. We examine system hardening in a cloud environment to assess if standard operating system hardening benchmarks are applicable in a cloud virtual machine.Five case studies of large-scale security breaches are analyzed to determine which system hardening benchmarks would have mitigated and possibly prevented the various attacks. These examples serve as a reminder of the importance of system hardening in general, and the IaaS consumer must protect the portion of the environment under their control and responsibility. Two industry standard benchmarks (CIS and STIG) are discussed, and STIG is implemented on an AWS EC2 instance running Red Hat Enterprise Linux 7.After evaluating 207 scripted benchmarks, we recommend and pass 195, leaving only 12 failed items which form the basis of discussion for this paper. These failed items include five false positives, one known bug and two exemptions, subsequently raising our compliance score to 98%. In a real-world audit, the remaining four failed items would be scrutinized for compensating controls or strong justification for a process exemption.
减少攻击向量!最小化风险!在讨论系统加固的主题时,目标是锁定系统,使黑客更难以侵入。针对各种应用程序和产品存在许多强化标准,但也许第一步应该是操作系统本身。我们将检查云环境中的系统加固,以评估标准操作系统加固基准是否适用于云虚拟机。本文分析了五个大规模安全漏洞的案例研究,以确定哪些系统加固基准可以减轻并可能阻止各种攻击。这些例子提醒我们系统加固的重要性,IaaS使用者必须保护他们控制和负责的那部分环境。讨论了两个行业标准基准(CIS和STIG), STIG是在运行Red Hat Enterprise Linux 7的AWS EC2实例上实现的。在评估了207个脚本基准之后,我们推荐并通过了195个,只留下12个不合格的项目,这些项目构成了本文讨论的基础。这些失败的项目包括五个误报,一个已知漏洞和两个豁免,随后将我们的合规得分提高到98%。在真实的审计中,将仔细检查其余四个失败的项目,以确定是否有补偿控制或过程豁免的有力理由。