Amazon GuardDuty 的新增功能 – Amazon EBS 卷的恶意软件检测

{"value":"借助 [Amazon GuardDuty](https://aws.amazon.com/guardduty/)，您可以监控自己的 AWS 账户和工作负载以检测恶意活动。我们于近日在 GuardDuty 中增加了检测恶意软件的功能。恶意软件用于危害工作负载、重新调整资源用途或未经授权访问数据。启用 **GuardDuty 恶意软件保护**之后，当 [GuardDuty 检测到您在 EC2 上运行的 EC2 实例或容器工作负载之一正在执行可疑操作](https://docs.aws.amazon.com/guardduty/latest/ug/gd-findings-initiate-malware-protection-scan.html)时，就会启动恶意软件扫描。例如，当 EC2 实例与已知恶意的命令和控制服务器进行通信时，或者正在对其他 EC2 实例执行拒绝服务 (DoS) 或蛮力破解攻击时，就会触发恶意软件扫描。\n\nGuardDuty [支持许多文件系统类型](https://docs.aws.amazon.com/guardduty/latest/ug/malware-protection-supported-file-formats.html)，该工具会扫描已知用于传播或包含恶意软件的文件格式，包括 Windows 和 Linux 可执行文件、PDF 文件、存档、二进制文件、脚本、安装程序、电子邮件数据库和普通电子邮件。\n\n识别出潜在的恶意软件后，将生成可操作的安全调查结果，其中包含威胁和文件名、文件路径、EC2 实例 ID、资源标签以及（如果是容器）容器 ID 和使用的容器映像等信息。GuardDuty 支持在 EC2 上运行的容器工作负载，包括客户管理的 [Kubernetes](https://kubernetes.io/) 集群或个别 [Docker 容器](https://www.docker.com/resources/what-container/)。如果容器由 [Amazon Elastic Kubernetes Service (EKS) ](https://aws.amazon.com/eks/)or [Amazon Elastic Container Service (Amazon ECS)](https://aws.amazon.com/ecs/) 管理，则调查结果还包括集群名称和[任务](https://docs.aws.amazon.com/AmazonECS/latest/developerguide/task_definitions.html)或 [pod](https://kubernetes.io/docs/concepts/workloads/pods/) ID，因此应用程序安全团队可以快速找到受影响的容器资源。\n\n与所有其他 GuardDuty 调查结果一样，恶意软件检测将发送到 GuardDuty 控制台，通过 [Amazon EventBridge](https://aws.amazon.com/eventbridge) 推送，路由到 [AWS Security Hub](https://aws.amazon.com/security-hub/)，然后在 [Amazon Detective](https://aws.amazon.com/detective) 中提供以开展事件调查。\n\n#### **++GuardDuty 恶意软件保护的工作原理++**\n启用恶意软件保护后，您将设置一个 [Amazon Identity and Access Management (IAM) 服务相关角色，该角色授予 GuardDuty 执行恶意软件扫描的权限](https://aws.amazon.com/iam/)。对 EC2 实例启动恶意软件扫描时，GuardDuty 恶意软件保护使用这些权限拍摄小于 1 TB 的附加 [Amazon Elastic Block Store (EBS)](https://aws.amazon.com/ebs/) 卷快照，然后在同一 [Amazon 区域](https://aws.amazon.com/about-aws/global-infrastructure/regions_az/#Regions)内的 Amazon 服务账户中恢复 EBS 卷，以对它们进行恶意软件扫描。您可以使用标记，从这些权限和扫描中包括或排除 EC2 实例。这样，您无需部署安全软件或代理来监控恶意软件，扫描卷的操作也不会影响正在运行的工作负载。扫描完成后，将删除服务账户中的 EBS 卷和您账户中的快照。或者，您可以在检测到恶意软件时保留快照。\n\n服务相关角色授予 GuardDuty 访问用于加密 EBS 卷的 [Amazon Key Management Service (Amazon KMS) ](https://aws.amazon.com/kms/)密钥的权限。如果对附加到可能遭到入侵的 EC2 实例的 EBS 卷使用客户托管式密钥进行加密，则 GuardDuty 恶意软件保护也会使用相同的密钥来加密副本 EBS 卷。如果这些卷未加密，GuardDuty 将使用自己的密钥来加密副本 EBS 卷并确保隐私。不支持使用 EBS 托管式密钥加密的卷。\n\n云中的安全是您与 Amazon [共同承担的责任](https://aws.amazon.com/compliance/shared-responsibility-model/)。作为防护机制，GuardDuty 恶意软件保护使用的服务相关角色如果具有 ```GuardDutyExcluded\n```标签，则无法对您的资源（例如 EBS 快照和卷、EC2 实例和 KMS 密钥）执行任何操作。在将 ```\nGuardDutyExcluded\n```设置为 ```true\n```的情况下标记快照后，GuardDuty 服务将无法访问这些快照。```GuardDutyExclud``` 标签将取代任何包含标记。权限还会限制 GuardDuty 修改快照的方式，以便在与 GuardDuty 服务账户共享时无法将其公开。\n\n由 GuardDuty 创建的 EBS 卷始终是加密的。GuardDuty 只能在具有 GuardDuty 扫描 ID 标签的 EBS 快照上使用 KMS 密钥。在获得 EC2 检测结果后创建快照时，GuardDuty 会添加扫描 ID 标签。无法从除 Amazon EBS 服务之外的任何其他[上下文](https://docs.aws.amazon.com/kms/latest/developerguide/concepts.html#encrypt_context)调用与 GuardDuty 服务账户共享的 KMS 密钥。扫描成功完成后，将撤销 KMS 密钥授予并删除 GuardDuty 服务账户中的卷副本，从而确保 GuardDuty 服务在完成扫描操作后无法访问您的数据。\n\n#### **++为 Amazon 账户启用恶意软件保护++**\n如果您尚未使用 GuardDuty，则在为自己的账户激活 GuardDuty 时，默认情况下会启用恶意软件保护。我已经在使用 GuardDuty，因此需要从控制台启用恶意软件保护。如果您正在使用 [Amazon Organizations](https://aws.amazon.com/organizations/)，您的[代理管理员账户](https://docs.aws.amazon.com/guardduty/latest/ug/guardduty_organizations.html)可以为现有成员账户启用此功能，并配置是否应自动注册组织中的新 Amazon 账户。\n\n在 GuardDuty 控制台中，我选择导航窗格中 **Settings**（设置）下的 **Malware Protection**（恶意软件保护）。在其中，我选择 **Enable**（启用），然后选择 **Enable Malware Protection**（启用恶意软件保护）。\n\n\n\n屏幕截图在扫描后会自动删除。在 **General settings**（常规设置）中，我可以选择将检测到恶意软件的屏幕截图保留在我的 Amazon 账户中，并将其用于进一步分析。\n\n\n\n在 **Scan options**（扫描选项）中，我可以配置包含标签的列表，以便只扫描带有这些标签的 EC2 实例；或者排除标签，以便跳过列表中包含标签的 EC2 实例。\n\n\n\n#### **++测试恶意软件保护 GuardDuty 检测结果++**\n为了生成几个 Amazon GuardDuty 检测结果，包括新的恶意软件保护检测结果，我克隆了 Amazon GuardDuty Tester 存储库：\n\n```\n$ git clone https://github.com/awslabs/amazon-guardduty-tester\n```\n\n首先，我使用 ```guardduty-tester.template``` 文件创建一个 [Amazon CloudFormation](https://aws.amazon.com/cloudformation/) 堆栈。堆栈准备就绪后，我会按照说明将 SSH 客户端配置为通过堡垒主机登录到测试器实例。然后，我连接到测试器实例：\n\n```\n$ ssh tester\n```\n\n从测试器实例中，我启动```guardduty_tester.sh``` 脚本来生成检测结果：\n\n```\n$ ./guardduty_tester.sh \n\n*********************************************************************\n* Test #1 - Internal port scanning *\n* This simulates internal reconaissance by an internal actor or an *\n* external actor after an initial compromise.This is considered a *\n* low priority finding for GuardDuty because its not a clear indicator*\n* of malicious intent on its own. *\n*********************************************************************\n\n\nStarting Nmap 6.40 ( http://nmap.org ) at 2022-05-19 09:36 UTC\nNmap scan report for ip-172-16-0-20.us-west-2.compute.internal (172.16.0.20)\nHost is up (0.00032s latency).\nNot shown: 997 filtered ports\nPORT STATE SERVICE\n22/tcp open ssh\n80/tcp closed http\n5050/tcp closed mmcc\nMAC Address: 06:25:CB:F4:E0:51 (Unknown)\n\nNmap done: 1 IP address (1 host up) scanned in 4.96 seconds\n\n-----------------------------------------------------------------------\n\n*********************************************************************\n* Test #2 - SSH Brute Force with Compromised Keys *\n* This simulates an SSH brute force attack on an SSH port that we *\n* can access from this instance.It uses (phony) compromised keys in *\n* many subsequent attempts to see if one works.This is a common *\n* techique where the bad actors will harvest keys from the web in *\n* places like source code repositories where people accidentally leave*\n* keys and credentials (This attempt will not actually succeed in *\n* obtaining access to the target linux instance in this subnet) *\n*********************************************************************\n\n2022-05-19 09:36:29 START\n2022-05-19 09:36:29 Crowbar v0.4.3-dev\n2022-05-19 09:36:29 Trying 172.16.0.20:22\n2022-05-19 09:36:33 STOP\n2022-05-19 09:36:33 No results found...\n2022-05-19 09:36:33 START\n2022-05-19 09:36:33 Crowbar v0.4.3-dev\n2022-05-19 09:36:33 Trying 172.16.0.20:22\n2022-05-19 09:36:37 STOP\n2022-05-19 09:36:37 No results found...\n2022-05-19 09:36:37 START\n2022-05-19 09:36:37 Crowbar v0.4.3-dev\n2022-05-19 09:36:37 Trying 172.16.0.20:22\n2022-05-19 09:36:41 STOP\n2022-05-19 09:36:41 No results found...\n2022-05-19 09:36:41 START\n2022-05-19 09:36:41 Crowbar v0.4.3-dev\n2022-05-19 09:36:41 Trying 172.16.0.20:22\n2022-05-19 09:36:45 STOP\n2022-05-19 09:36:45 No results found...\n2022-05-19 09:36:45 START\n2022-05-19 09:36:45 Crowbar v0.4.3-dev\n2022-05-19 09:36:45 Trying 172.16.0.20:22\n2022-05-19 09:36:48 STOP\n2022-05-19 09:36:48 No results found...\n2022-05-19 09:36:49 START\n2022-05-19 09:36:49 Crowbar v0.4.3-dev\n2022-05-19 09:36:49 Trying 172.16.0.20:22\n2022-05-19 09:36:52 STOP\n2022-05-19 09:36:52 No results found...\n2022-05-19 09:36:52 START\n2022-05-19 09:36:52 Crowbar v0.4.3-dev\n2022-05-19 09:36:52 Trying 172.16.0.20:22\n2022-05-19 09:36:56 STOP\n2022-05-19 09:36:56 No results found...\n2022-05-19 09:36:56 START\n2022-05-19 09:36:56 Crowbar v0.4.3-dev\n2022-05-19 09:36:56 Trying 172.16.0.20:22\n2022-05-19 09:37:00 STOP\n2022-05-19 09:37:00 No results found...\n2022-05-19 09:37:00 START\n2022-05-19 09:37:00 Crowbar v0.4.3-dev\n2022-05-19 09:37:00 Trying 172.16.0.20:22\n2022-05-19 09:37:04 STOP\n2022-05-19 09:37:04 No results found...\n2022-05-19 09:37:04 START\n2022-05-19 09:37:04 Crowbar v0.4.3-dev\n2022-05-19 09:37:04 Trying 172.16.0.20:22\n2022-05-19 09:37:08 STOP\n2022-05-19 09:37:08 No results found...\n2022-05-19 09:37:08 START\n2022-05-19 09:37:08 Crowbar v0.4.3-dev\n2022-05-19 09:37:08 Trying 172.16.0.20:22\n2022-05-19 09:37:12 STOP\n2022-05-19 09:37:12 No results found...\n2022-05-19 09:37:12 START\n2022-05-19 09:37:12 Crowbar v0.4.3-dev\n2022-05-19 09:37:12 Trying 172.16.0.20:22\n2022-05-19 09:37:16 STOP\n2022-05-19 09:37:16 No results found...\n2022-05-19 09:37:16 START\n2022-05-19 09:37:16 Crowbar v0.4.3-dev\n2022-05-19 09:37:16 Trying 172.16.0.20:22\n2022-05-19 09:37:20 STOP\n2022-05-19 09:37:20 No results found...\n2022-05-19 09:37:20 START\n2022-05-19 09:37:20 Crowbar v0.4.3-dev\n2022-05-19 09:37:20 Trying 172.16.0.20:22\n2022-05-19 09:37:23 STOP\n2022-05-19 09:37:23 No results found...\n2022-05-19 09:37:23 START\n2022-05-19 09:37:23 Crowbar v0.4.3-dev\n2022-05-19 09:37:23 Trying 172.16.0.20:22\n2022-05-19 09:37:27 STOP\n2022-05-19 09:37:27 No results found...\n2022-05-19 09:37:27 START\n2022-05-19 09:37:27 Crowbar v0.4.3-dev\n2022-05-19 09:37:27 Trying 172.16.0.20:22\n2022-05-19 09:37:31 STOP\n2022-05-19 09:37:31 No results found...\n2022-05-19 09:37:31 START\n2022-05-19 09:37:31 Crowbar v0.4.3-dev\n2022-05-19 09:37:31 Trying 172.16.0.20:22\n2022-05-19 09:37:34 STOP\n2022-05-19 09:37:34 No results found...\n2022-05-19 09:37:35 START\n2022-05-19 09:37:35 Crowbar v0.4.3-dev\n2022-05-19 09:37:35 Trying 172.16.0.20:22\n2022-05-19 09:37:38 STOP\n2022-05-19 09:37:38 No results found...\n2022-05-19 09:37:38 START\n2022-05-19 09:37:38 Crowbar v0.4.3-dev\n2022-05-19 09:37:38 Trying 172.16.0.20:22\n2022-05-19 09:37:42 STOP\n2022-05-19 09:37:42 No results found...\n2022-05-19 09:37:42 START\n2022-05-19 09:37:42 Crowbar v0.4.3-dev\n2022-05-19 09:37:42 Trying 172.16.0.20:22\n2022-05-19 09:37:46 STOP\n2022-05-19 09:37:46 No results found...\n\n-----------------------------------------------------------------------\n\n*********************************************************************\n* Test #3 - RDP Brute Force with Password List *\n* This simulates an RDP brute force attack on the internal RDP port *\n* of the windows server that we installed in the environment. It uses*\n* a list of common passwords that can be found on the web.This test *\n* will trigger a detection, but will fail to get into the target *\n* windows instance. *\n*********************************************************************\n\nSending 250 password attempts at the windows server...\nHydra v9.4-dev (c) 2022 by van Hauser/THC & David Maciejak - Please do not use in military or secret service organizations, or for illegal purposes (this is non-binding, these *** ignore laws and ethics anyway).\n\nHydra (https://github.com/vanhauser-thc/thc-hydra) starting at 2022-05-19 09:37:46\n[WARNING] rdp servers often don't like many connections, use -t 1 or -t 4 to reduce the number of parallel connections and -W 1 or -W 3 to wait between connection to allow the server to recover\n[INFO] Reduced number of tasks to 4 (rdp does not like many parallel connections)\n[WARNING] the rdp module is experimental.Please test, report - and if possible, fix.\n[DATA] max 4 tasks per 1 server, overall 4 tasks, 1792 login tries (l:7/p:256), ~448 tries per task\n[DATA] attacking rdp://172.16.0.24:3389/\n[STATUS] 1099.00 tries/min, 1099 tries in 00:01h, 693 to do in 00:01h, 4 active\n1 of 1 target completed, 0 valid password found\nHydra (https://github.com/vanhauser-thc/thc-hydra) finished at 2022-05-19 09:39:23\n\n-----------------------------------------------------------------------\n\n*********************************************************************\n* Test #4 - CryptoCurrency Mining Activity *\n* This simulates interaction with a cryptocurrency mining pool which *\n* can be an indication of an instance compromise.In this case, we are*\n* only interacting with the URL of the pool, but not downloading *\n* any files.This will trigger a threat intel based detection. *\n*********************************************************************\n\nCalling bitcoin wallets to download mining toolkits\n\n-----------------------------------------------------------------------\n\n*********************************************************************\n* Test #5 - DNS Exfiltration *\n* A common exfiltration technique is to tunnel data out over DNS *\n* to a fake domain. Its an effective technique because most hosts *\n* have outbound DNS ports open. This test wont exfiltrate any data, *\n* but it will generate enough unusual DNS activity to trigger the *\n* detection. *\n*********************************************************************\n\nCalling large numbers of large domains to simulate tunneling via DNS\n\n*********************************************************************\n* Test #6 - Fake domain to prove that GuardDuty is working *\n* This is a permanent fake domain that customers can use to prove that*\n* GuardDuty is working. Calling this domain will always generate the *\n* Backdoor:EC2/C&CActivity.B!DNS finding type *\n*********************************************************************\n\nCalling a well known fake domain that is used to generate a known finding\n\n; <<>> DiG 9.11.4-P2-RedHat-9.11.4-26.P2.amzn2.5.2 <<>> GuardDutyC2ActivityB.com any\n;; global options: +cmd\n;; Got answer:\n;; ->>HEADER<<- opcode: QUERY, status: NOERROR, id: 11495\n;; flags: qr rd ra; QUERY: 1, ANSWER: 8, AUTHORITY: 0, ADDITIONAL: 1\n\n;; OPT PSEUDOSECTION:\n; EDNS: version: 0, flags:; udp: 4096\n;; QUESTION SECTION:\n;GuardDutyC2ActivityB.com.\tIN\tANY\n\n;; ANSWER SECTION:\nGuardDutyC2ActivityB.com.6943\tIN\tSOA\tns1.markmonitor.com. hostmaster.markmonitor.com.2018091906 86400 3600 2592000 172800\nGuardDutyC2ActivityB.com.6943\tIN\tNS\tns3.markmonitor.com.\nGuardDutyC2ActivityB.com.6943\tIN\tNS\tns5.markmonitor.com.\nGuardDutyC2ActivityB.com.6943\tIN\tNS\tns7.markmonitor.com.\nGuardDutyC2ActivityB.com.6943\tIN\tNS\tns2.markmonitor.com.\nGuardDutyC2ActivityB.com.6943\tIN\tNS\tns4.markmonitor.com.\nGuardDutyC2ActivityB.com.6943\tIN\tNS\tns6.markmonitor.com.\nGuardDutyC2ActivityB.com.6943\tIN\tNS\tns1.markmonitor.com.\n\n;; Query time: 27 msec\n;; SERVER: 172.16.0.2#53(172.16.0.2)\n;; WHEN: Thu May 19 09:39:23 UTC 2022\n;; MSG SIZE rcvd: 238\n\n\n*****************************************************************************************************\n预期的 GuardDuty 检测结果\n\n测试 1：内部端口扫描\n预期的检测结果：EC2 实例 i-011e73af27562827b 正在对远程主机执行出站端口扫描。172.16.0.20\n检测结果类型：Recon:EC2/Portscan\n\n测试 2：使用泄露密钥的 SSH 蛮力攻击\n预期有两个检测结果 - 一个用于出站检测，另一个用于入站检测\n出站：i-011e73af27562827b 正在对 172.16.0.20 进行 SSH 蛮力攻击\n入站：172.16.0.25 正在对 i-0bada13e0aa12d383 进行 SSH 蛮力攻击\n检测结果类型：UnauthorizedAccess:EC2/SSHBruteForce\n\n测试 3：使用密码列表的 RDP 蛮力攻击\n预期有两个检测结果 - 一个用于出站检测，另一个用于入站检测\n出站：i-011e73af27562827b 正在对 172.16.0.24 进行 RDP 蛮力攻击\n入站：172.16.0.25 正在对 i-0191573dec3b66924 进行 RDP 蛮力攻击\n检测结果类型：UnauthorizedAccess:EC2/RDPBruteForce\n\n测试 4：加密货币活动\n预期的检测结果：EC2 实例 i-011e73af27562827b 正在查询与比特币活动相关的域名\n检测结果类型：CryptoCurrency:EC2/BitcoinTool.B!DNS\n\n测试 5：DNS 渗透\n预期的检测结果：EC2 实例 i-011e73af27562827b 正在尝试查询与泄露数据相似的域名\n检测结果类型：Trojan:EC2/DNSDataExfiltration\n\n测试 6：C&C 活动\n预期检测结果：EC2 实例 i-011e73af27562827b 正在查询与已知命令与控制服务器关联的域名。\n检测结果类型：Backdoor:EC2/C&CActivity.B!DNS\n```\n\n几分钟后，检测结果将显示在 GuardDuty 控制台中。在顶部，我看到了新的恶意软件保护功能发现的恶意文件。其中一个检测结果与 EC2 实例相关，另一个与 ECS 集群相关。\n\n\n\n首先，我选择与 EC2 实例相关的检测结果。在面板中，我看到有关实例和恶意文件的信息，例如文件名和路径。在 Malware scan details（恶意软件扫描详细信息）部分中，Trigger finding ID（触发检测结果 ID）指向触发恶意软件扫描的原始 GuardDuty 检测结果。就我而言，最初的检测结果是此 EC2 实例正在对另一个 EC2 实例执行 RDP 蛮力攻击。\n\n\n\n在此处，我选择 **Investigate with Detective**（使用 Detective 调查），然后直接从 GuardDuty 控制台转到 [Detective 控制台](https://console.aws.amazon.com/detective)，从中可视化 EC2 实例、Amazon 账户和受检测结果影响的 IP 地址的 [Amazon CloudTrail](https://aws.amazon.com/cloudtrail/) 和 [Amazon Virtual Private Cloud (Amazon VPC)](https://aws.amazon.com/vpc/) 流数据。使用 Detective，我可以分析、调查和确定 GuardDuty 发现的可疑活动的根本原因。\n\n\n\n选择与 ECS 群集相关的检测结果时，我将获得有关受影响资源的更多信息，例如 ECS 集群、任务、容器和容器映像的详细信息。\n\n\n\n使用 GuardDuty 测试器脚本可以更轻松地测试 GuardDuty 与您使用的其他安全框架的整体集成，以便在检测到真正的威胁时做好准备。\n\n#### **++将 GuardDuty 恶意软件防护保护与 Amazon Inspector 进行比较++**\n此时，您可能会问自己 GuardDuty 恶意软件保护与 [Amazon Inspector](https://aws.amazon.com/inspector/) 有何关系，Amazon Inspector 是一项扫描 AWS 工作负载的软件漏洞和意外网络暴露的服务。这两种服务相辅相成，提供不同的保护层：\n\n- Amazon Inspector 通过识别和修复已知软件和应用程序漏洞提供主动防护，这些漏洞是攻击者入侵资源和安装恶意软件的入口点。\n- GuardDuty 恶意软件保护可检测到活跃运行的工作负载中存在的恶意软件。此时，系统已经受损，但是 GuardDuty 可以限制感染的时间，并在系统受损导致影响业务的事件之前采取措施。\n#### **++可用性和定价++**\n[Amazon GuardDuty](https://aws.amazon.com/guardduty/) 恶意软件保护现已在[所有提供 GuardDuty 的 Amazon 区域](https://aws.amazon.com/about-aws/global-infrastructure/regional-product-services/)推出，不包括 AWS 中国（北京）、Amazon 中国（宁夏）、Amazon GovCloud（美国东部）和 Amazon GovCloud（美国西部）区域。\n\n在发布时，GuardDuty 恶意软件保护已与以下合作伙伴产品集成：\n\n- [BitDefender](https://www.bitdefender.com/)\n- [CloudHesive](https://www.cloudhesive.com/)\n- [Crowdstrike](http://www.crowdstrike.com/)\n- [Fortinet](https://www.fortinet.com/)\n- [Palo Alto Networks](https://www.paloaltonetworks.com/)\n- [Rapid7](https://www.rapid7.com/)\n- [Sophos](https://www.sophos.com/)\n- [Sysdig](https://sysdig.com/)\n- [Trellix](https://www.trellix.com/)\n\n使用 GuardDuty，您无需部署安全软件或代理来监控恶意软件。您只需为文件系统中扫描的 GB 量（而不是 EBS 卷的大小）和 EBS 快照保存在您账户中的时间付费。除非您在发现恶意软件时启用快照保留功能，否则 GuardDuty 创建的所有 EBS 快照都会在扫描后自动删除。有关更多信息，请参阅 [GuardDuty 定价](https://aws.amazon.com/guardduty/pricing/)和 [EBS 定价](https://aws.amazon.com/ebs/pricing/)。请注意，GuardDuty 仅扫描小于 1 TB 的 EBS 卷。为了帮助您控制成本并避免重复警报，同一卷的扫描频率不会超过每 24 小时一次。\n\n#### **[使用 Amazon GuardDuty 检测恶意活动并保护您的应用程序免受恶意软件侵害。](https://aws.amazon.com/guardduty/faqs/#GuardDuty_Malware_Protection)**\n\n— [Danilo](https://twitter.com/danilop)","render":"<p>借助 <a href=\"https://aws.amazon.com/guardduty/\" target=\"_blank\">Amazon GuardDuty</a>，您可以监控自己的 AWS 账户和工作负载以检测恶意活动。我们于近日在 GuardDuty 中增加了检测恶意软件的功能。恶意软件用于危害工作负载、重新调整资源用途或未经授权访问数据。启用 <strong>GuardDuty 恶意软件保护</strong>之后，当 <a href=\"https://docs.aws.amazon.com/guardduty/latest/ug/gd-findings-initiate-malware-protection-scan.html\" target=\"_blank\">GuardDuty 检测到您在 EC2 上运行的 EC2 实例或容器工作负载之一正在执行可疑操作</a>时，就会启动恶意软件扫描。例如，当 EC2 实例与已知恶意的命令和控制服务器进行通信时，或者正在对其他 EC2 实例执行拒绝服务 (DoS) 或蛮力破解攻击时，就会触发恶意软件扫描。</p>\n<p>GuardDuty <a href=\"https://docs.aws.amazon.com/guardduty/latest/ug/malware-protection-supported-file-formats.html\" target=\"_blank\">支持许多文件系统类型</a>，该工具会扫描已知用于传播或包含恶意软件的文件格式，包括 Windows 和 Linux 可执行文件、PDF 文件、存档、二进制文件、脚本、安装程序、电子邮件数据库和普通电子邮件。</p>\n<p>识别出潜在的恶意软件后，将生成可操作的安全调查结果，其中包含威胁和文件名、文件路径、EC2 实例 ID、资源标签以及（如果是容器）容器 ID 和使用的容器映像等信息。GuardDuty 支持在 EC2 上运行的容器工作负载，包括客户管理的 <a href=\"https://kubernetes.io/\" target=\"_blank\">Kubernetes</a> 集群或个别 <a href=\"https://www.docker.com/resources/what-container/\" target=\"_blank\">Docker 容器</a>。如果容器由 <a href=\"https://aws.amazon.com/eks/\" target=\"_blank\">Amazon Elastic Kubernetes Service (EKS) </a>or <a href=\"https://aws.amazon.com/ecs/\" target=\"_blank\">Amazon Elastic Container Service (Amazon ECS)</a> 管理，则调查结果还包括集群名称和<a href=\"https://docs.aws.amazon.com/AmazonECS/latest/developerguide/task_definitions.html\" target=\"_blank\">任务</a>或 <a href=\"https://kubernetes.io/docs/concepts/workloads/pods/\" target=\"_blank\">pod</a> ID，因此应用程序安全团队可以快速找到受影响的容器资源。</p>\n<p>与所有其他 GuardDuty 调查结果一样，恶意软件检测将发送到 GuardDuty 控制台，通过 <a href=\"https://aws.amazon.com/eventbridge\" target=\"_blank\">Amazon EventBridge</a> 推送，路由到 <a href=\"https://aws.amazon.com/security-hub/\" target=\"_blank\">AWS Security Hub</a>，然后在 <a href=\"https://aws.amazon.com/detective\" target=\"_blank\">Amazon Detective</a> 中提供以开展事件调查。</p>\n<h4><a id=\"GuardDuty__8\"></a><strong><ins>GuardDuty 恶意软件保护的工作原理</ins></strong></h4>\n<p>启用恶意软件保护后，您将设置一个 <a href=\"https://aws.amazon.com/iam/\" target=\"_blank\">Amazon Identity and Access Management (IAM) 服务相关角色，该角色授予 GuardDuty 执行恶意软件扫描的权限</a>。对 EC2 实例启动恶意软件扫描时，GuardDuty 恶意软件保护使用这些权限拍摄小于 1 TB 的附加 <a href=\"https://aws.amazon.com/ebs/\" target=\"_blank\">Amazon Elastic Block Store (EBS)</a> 卷快照，然后在同一 <a href=\"https://aws.amazon.com/about-aws/global-infrastructure/regions_az/#Regions\" target=\"_blank\">Amazon 区域</a>内的 Amazon 服务账户中恢复 EBS 卷，以对它们进行恶意软件扫描。您可以使用标记，从这些权限和扫描中包括或排除 EC2 实例。这样，您无需部署安全软件或代理来监控恶意软件，扫描卷的操作也不会影响正在运行的工作负载。扫描完成后，将删除服务账户中的 EBS 卷和您账户中的快照。或者，您可以在检测到恶意软件时保留快照。</p>\n<p>服务相关角色授予 GuardDuty 访问用于加密 EBS 卷的 <a href=\"https://aws.amazon.com/kms/\" target=\"_blank\">Amazon Key Management Service (Amazon KMS) </a>密钥的权限。如果对附加到可能遭到入侵的 EC2 实例的 EBS 卷使用客户托管式密钥进行加密，则 GuardDuty 恶意软件保护也会使用相同的密钥来加密副本 EBS 卷。如果这些卷未加密，GuardDuty 将使用自己的密钥来加密副本 EBS 卷并确保隐私。不支持使用 EBS 托管式密钥加密的卷。</p>\n<p>云中的安全是您与 Amazon <a href=\"https://aws.amazon.com/compliance/shared-responsibility-model/\" target=\"_blank\">共同承担的责任</a>。作为防护机制，GuardDuty 恶意软件保护使用的服务相关角色如果具有 <code>GuardDutyExcluded </code>标签，则无法对您的资源（例如 EBS 快照和卷、EC2 实例和 KMS 密钥）执行任何操作。在将 <code>GuardDutyExcluded</code>设置为 <code>true </code>的情况下标记快照后，GuardDuty 服务将无法访问这些快照。<code>GuardDutyExclud</code> 标签将取代任何包含标记。权限还会限制 GuardDuty 修改快照的方式，以便在与 GuardDuty 服务账户共享时无法将其公开。</p>\n<p>由 GuardDuty 创建的 EBS 卷始终是加密的。GuardDuty 只能在具有 GuardDuty 扫描 ID 标签的 EBS 快照上使用 KMS 密钥。在获得 EC2 检测结果后创建快照时，GuardDuty 会添加扫描 ID 标签。无法从除 Amazon EBS 服务之外的任何其他<a href=\"https://docs.aws.amazon.com/kms/latest/developerguide/concepts.html#encrypt_context\" target=\"_blank\">上下文</a>调用与 GuardDuty 服务账户共享的 KMS 密钥。扫描成功完成后，将撤销 KMS 密钥授予并删除 GuardDuty 服务账户中的卷副本，从而确保 GuardDuty 服务在完成扫描操作后无法访问您的数据。</p>\n<h4><a id=\"_Amazon__21\"></a><strong><ins>为 Amazon 账户启用恶意软件保护</ins></strong></h4>\n<p>如果您尚未使用 GuardDuty，则在为自己的账户激活 GuardDuty 时，默认情况下会启用恶意软件保护。我已经在使用 GuardDuty，因此需要从控制台启用恶意软件保护。如果您正在使用 <a href=\"https://aws.amazon.com/organizations/\" target=\"_blank\">Amazon Organizations</a>，您的<a href=\"https://docs.aws.amazon.com/guardduty/latest/ug/guardduty_organizations.html\" target=\"_blank\">代理管理员账户</a>可以为现有成员账户启用此功能，并配置是否应自动注册组织中的新 Amazon 账户。</p>\n<p>在 GuardDuty 控制台中，我选择导航窗格中 <strong>Settings</strong>（设置）下的 <strong>Malware Protection</strong>（恶意软件保护）。在其中，我选择 <strong>Enable</strong>（启用），然后选择 <strong>Enable Malware Protection</strong>（启用恶意软件保护）。</p>\n<p><img src=\"https://dev-media.amazoncloud.cn/a96814621aa9466a8d758e5e0c51a844_image.png\" alt=\"image.png\" /></p>\n<p>屏幕截图在扫描后会自动删除。在 <strong>General settings</strong>（常规设置）中，我可以选择将检测到恶意软件的屏幕截图保留在我的 Amazon 账户中，并将其用于进一步分析。</p>\n<p><img src=\"https://dev-media.amazoncloud.cn/583297e9ec7a4afa9a6a3c54d6712e57_image.png\" alt=\"image.png\" /></p>\n<p>在 <strong>Scan options</strong>（扫描选项）中，我可以配置包含标签的列表，以便只扫描带有这些标签的 EC2 实例；或者排除标签，以便跳过列表中包含标签的 EC2 实例。</p>\n<p><img src=\"https://dev-media.amazoncloud.cn/467bb293b9d040e984773c3cffe5e225_image.png\" alt=\"image.png\" /></p>\n<h4><a id=\"_GuardDuty__36\"></a><strong><ins>测试恶意软件保护 GuardDuty 检测结果</ins></strong></h4>\n<p>为了生成几个 Amazon GuardDuty 检测结果，包括新的恶意软件保护检测结果，我克隆了 Amazon GuardDuty Tester 存储库：</p>\n<pre><code class=\"lang-\">$ git clone https://github.com/awslabs/amazon-guardduty-tester\n</code></pre>\n<p>首先，我使用 <code>guardduty-tester.template</code> 文件创建一个 <a href=\"https://aws.amazon.com/cloudformation/\" target=\"_blank\">Amazon CloudFormation</a> 堆栈。堆栈准备就绪后，我会按照说明将 SSH 客户端配置为通过堡垒主机登录到测试器实例。然后，我连接到测试器实例：</p>\n<pre><code class=\"lang-\">$ ssh tester\n</code></pre>\n<p>从测试器实例中，我启动<code>guardduty_tester.sh</code> 脚本来生成检测结果：</p>\n<pre><code class=\"lang-\">$ ./guardduty_tester.sh \n\n*********************************************************************\n* Test #1 - Internal port scanning *\n* This simulates internal reconaissance by an internal actor or an *\n* external actor after an initial compromise.This is considered a *\n* low priority finding for GuardDuty because its not a clear indicator*\n* of malicious intent on its own. *\n*********************************************************************\n\n\nStarting Nmap 6.40 ( http://nmap.org ) at 2022-05-19 09:36 UTC\nNmap scan report for ip-172-16-0-20.us-west-2.compute.internal (172.16.0.20)\nHost is up (0.00032s latency).\nNot shown: 997 filtered ports\nPORT STATE SERVICE\n22/tcp open ssh\n80/tcp closed http\n5050/tcp closed mmcc\nMAC Address: 06:25:CB:F4:E0:51 (Unknown)\n\nNmap done: 1 IP address (1 host up) scanned in 4.96 seconds\n\n-----------------------------------------------------------------------\n\n*********************************************************************\n* Test #2 - SSH Brute Force with Compromised Keys *\n* This simulates an SSH brute force attack on an SSH port that we *\n* can access from this instance.It uses (phony) compromised keys in *\n* many subsequent attempts to see if one works.This is a common *\n* techique where the bad actors will harvest keys from the web in *\n* places like source code repositories where people accidentally leave*\n* keys and credentials (This attempt will not actually succeed in *\n* obtaining access to the target linux instance in this subnet) *\n*********************************************************************\n\n2022-05-19 09:36:29 START\n2022-05-19 09:36:29 Crowbar v0.4.3-dev\n2022-05-19 09:36:29 Trying 172.16.0.20:22\n2022-05-19 09:36:33 STOP\n2022-05-19 09:36:33 No results found...\n2022-05-19 09:36:33 START\n2022-05-19 09:36:33 Crowbar v0.4.3-dev\n2022-05-19 09:36:33 Trying 172.16.0.20:22\n2022-05-19 09:36:37 STOP\n2022-05-19 09:36:37 No results found...\n2022-05-19 09:36:37 START\n2022-05-19 09:36:37 Crowbar v0.4.3-dev\n2022-05-19 09:36:37 Trying 172.16.0.20:22\n2022-05-19 09:36:41 STOP\n2022-05-19 09:36:41 No results found...\n2022-05-19 09:36:41 START\n2022-05-19 09:36:41 Crowbar v0.4.3-dev\n2022-05-19 09:36:41 Trying 172.16.0.20:22\n2022-05-19 09:36:45 STOP\n2022-05-19 09:36:45 No results found...\n2022-05-19 09:36:45 START\n2022-05-19 09:36:45 Crowbar v0.4.3-dev\n2022-05-19 09:36:45 Trying 172.16.0.20:22\n2022-05-19 09:36:48 STOP\n2022-05-19 09:36:48 No results found...\n2022-05-19 09:36:49 START\n2022-05-19 09:36:49 Crowbar v0.4.3-dev\n2022-05-19 09:36:49 Trying 172.16.0.20:22\n2022-05-19 09:36:52 STOP\n2022-05-19 09:36:52 No results found...\n2022-05-19 09:36:52 START\n2022-05-19 09:36:52 Crowbar v0.4.3-dev\n2022-05-19 09:36:52 Trying 172.16.0.20:22\n2022-05-19 09:36:56 STOP\n2022-05-19 09:36:56 No results found...\n2022-05-19 09:36:56 START\n2022-05-19 09:36:56 Crowbar v0.4.3-dev\n2022-05-19 09:36:56 Trying 172.16.0.20:22\n2022-05-19 09:37:00 STOP\n2022-05-19 09:37:00 No results found...\n2022-05-19 09:37:00 START\n2022-05-19 09:37:00 Crowbar v0.4.3-dev\n2022-05-19 09:37:00 Trying 172.16.0.20:22\n2022-05-19 09:37:04 STOP\n2022-05-19 09:37:04 No results found...\n2022-05-19 09:37:04 START\n2022-05-19 09:37:04 Crowbar v0.4.3-dev\n2022-05-19 09:37:04 Trying 172.16.0.20:22\n2022-05-19 09:37:08 STOP\n2022-05-19 09:37:08 No results found...\n2022-05-19 09:37:08 START\n2022-05-19 09:37:08 Crowbar v0.4.3-dev\n2022-05-19 09:37:08 Trying 172.16.0.20:22\n2022-05-19 09:37:12 STOP\n2022-05-19 09:37:12 No results found...\n2022-05-19 09:37:12 START\n2022-05-19 09:37:12 Crowbar v0.4.3-dev\n2022-05-19 09:37:12 Trying 172.16.0.20:22\n2022-05-19 09:37:16 STOP\n2022-05-19 09:37:16 No results found...\n2022-05-19 09:37:16 START\n2022-05-19 09:37:16 Crowbar v0.4.3-dev\n2022-05-19 09:37:16 Trying 172.16.0.20:22\n2022-05-19 09:37:20 STOP\n2022-05-19 09:37:20 No results found...\n2022-05-19 09:37:20 START\n2022-05-19 09:37:20 Crowbar v0.4.3-dev\n2022-05-19 09:37:20 Trying 172.16.0.20:22\n2022-05-19 09:37:23 STOP\n2022-05-19 09:37:23 No results found...\n2022-05-19 09:37:23 START\n2022-05-19 09:37:23 Crowbar v0.4.3-dev\n2022-05-19 09:37:23 Trying 172.16.0.20:22\n2022-05-19 09:37:27 STOP\n2022-05-19 09:37:27 No results found...\n2022-05-19 09:37:27 START\n2022-05-19 09:37:27 Crowbar v0.4.3-dev\n2022-05-19 09:37:27 Trying 172.16.0.20:22\n2022-05-19 09:37:31 STOP\n2022-05-19 09:37:31 No results found...\n2022-05-19 09:37:31 START\n2022-05-19 09:37:31 Crowbar v0.4.3-dev\n2022-05-19 09:37:31 Trying 172.16.0.20:22\n2022-05-19 09:37:34 STOP\n2022-05-19 09:37:34 No results found...\n2022-05-19 09:37:35 START\n2022-05-19 09:37:35 Crowbar v0.4.3-dev\n2022-05-19 09:37:35 Trying 172.16.0.20:22\n2022-05-19 09:37:38 STOP\n2022-05-19 09:37:38 No results found...\n2022-05-19 09:37:38 START\n2022-05-19 09:37:38 Crowbar v0.4.3-dev\n2022-05-19 09:37:38 Trying 172.16.0.20:22\n2022-05-19 09:37:42 STOP\n2022-05-19 09:37:42 No results found...\n2022-05-19 09:37:42 START\n2022-05-19 09:37:42 Crowbar v0.4.3-dev\n2022-05-19 09:37:42 Trying 172.16.0.20:22\n2022-05-19 09:37:46 STOP\n2022-05-19 09:37:46 No results found...\n\n-----------------------------------------------------------------------\n\n*********************************************************************\n* Test #3 - RDP Brute Force with Password List *\n* This simulates an RDP brute force attack on the internal RDP port *\n* of the windows server that we installed in the environment. It uses*\n* a list of common passwords that can be found on the web.This test *\n* will trigger a detection, but will fail to get into the target *\n* windows instance. *\n*********************************************************************\n\nSending 250 password attempts at the windows server...\nHydra v9.4-dev (c) 2022 by van Hauser/THC &amp; David Maciejak - Please do not use in military or secret service organizations, or for illegal purposes (this is non-binding, these *** ignore laws and ethics anyway).\n\nHydra (https://github.com/vanhauser-thc/thc-hydra) starting at 2022-05-19 09:37:46\n[WARNING] rdp servers often don't like many connections, use -t 1 or -t 4 to reduce the number of parallel connections and -W 1 or -W 3 to wait between connection to allow the server to recover\n[INFO] Reduced number of tasks to 4 (rdp does not like many parallel connections)\n[WARNING] the rdp module is experimental.Please test, report - and if possible, fix.\n[DATA] max 4 tasks per 1 server, overall 4 tasks, 1792 login tries (l:7/p:256), ~448 tries per task\n[DATA] attacking rdp://172.16.0.24:3389/\n[STATUS] 1099.00 tries/min, 1099 tries in 00:01h, 693 to do in 00:01h, 4 active\n1 of 1 target completed, 0 valid password found\nHydra (https://github.com/vanhauser-thc/thc-hydra) finished at 2022-05-19 09:39:23\n\n-----------------------------------------------------------------------\n\n*********************************************************************\n* Test #4 - CryptoCurrency Mining Activity *\n* This simulates interaction with a cryptocurrency mining pool which *\n* can be an indication of an instance compromise.In this case, we are*\n* only interacting with the URL of the pool, but not downloading *\n* any files.This will trigger a threat intel based detection. *\n*********************************************************************\n\nCalling bitcoin wallets to download mining toolkits\n\n-----------------------------------------------------------------------\n\n*********************************************************************\n* Test #5 - DNS Exfiltration *\n* A common exfiltration technique is to tunnel data out over DNS *\n* to a fake domain. Its an effective technique because most hosts *\n* have outbound DNS ports open. This test wont exfiltrate any data, *\n* but it will generate enough unusual DNS activity to trigger the *\n* detection. *\n*********************************************************************\n\nCalling large numbers of large domains to simulate tunneling via DNS\n\n*********************************************************************\n* Test #6 - Fake domain to prove that GuardDuty is working *\n* This is a permanent fake domain that customers can use to prove that*\n* GuardDuty is working. Calling this domain will always generate the *\n* Backdoor:EC2/C&amp;CActivity.B!DNS finding type *\n*********************************************************************\n\nCalling a well known fake domain that is used to generate a known finding\n\n; &lt;&lt;&gt;&gt; DiG 9.11.4-P2-RedHat-9.11.4-26.P2.amzn2.5.2 &lt;&lt;&gt;&gt; GuardDutyC2ActivityB.com any\n;; global options: +cmd\n;; Got answer:\n;; -&gt;&gt;HEADER&lt;&lt;- opcode: QUERY, status: NOERROR, id: 11495\n;; flags: qr rd ra; QUERY: 1, ANSWER: 8, AUTHORITY: 0, ADDITIONAL: 1\n\n;; OPT PSEUDOSECTION:\n; EDNS: version: 0, flags:; udp: 4096\n;; QUESTION SECTION:\n;GuardDutyC2ActivityB.com.\tIN\tANY\n\n;; ANSWER SECTION:\nGuardDutyC2ActivityB.com.6943\tIN\tSOA\tns1.markmonitor.com. hostmaster.markmonitor.com.2018091906 86400 3600 2592000 172800\nGuardDutyC2ActivityB.com.6943\tIN\tNS\tns3.markmonitor.com.\nGuardDutyC2ActivityB.com.6943\tIN\tNS\tns5.markmonitor.com.\nGuardDutyC2ActivityB.com.6943\tIN\tNS\tns7.markmonitor.com.\nGuardDutyC2ActivityB.com.6943\tIN\tNS\tns2.markmonitor.com.\nGuardDutyC2ActivityB.com.6943\tIN\tNS\tns4.markmonitor.com.\nGuardDutyC2ActivityB.com.6943\tIN\tNS\tns6.markmonitor.com.\nGuardDutyC2ActivityB.com.6943\tIN\tNS\tns1.markmonitor.com.\n\n;; Query time: 27 msec\n;; SERVER: 172.16.0.2#53(172.16.0.2)\n;; WHEN: Thu May 19 09:39:23 UTC 2022\n;; MSG SIZE rcvd: 238\n\n\n*****************************************************************************************************\n预期的 GuardDuty 检测结果\n\n测试 1：内部端口扫描\n预期的检测结果：EC2 实例 i-011e73af27562827b 正在对远程主机执行出站端口扫描。172.16.0.20\n检测结果类型：Recon:EC2/Portscan\n\n测试 2：使用泄露密钥的 SSH 蛮力攻击\n预期有两个检测结果 - 一个用于出站检测，另一个用于入站检测\n出站：i-011e73af27562827b 正在对 172.16.0.20 进行 SSH 蛮力攻击\n入站：172.16.0.25 正在对 i-0bada13e0aa12d383 进行 SSH 蛮力攻击\n检测结果类型：UnauthorizedAccess:EC2/SSHBruteForce\n\n测试 3：使用密码列表的 RDP 蛮力攻击\n预期有两个检测结果 - 一个用于出站检测，另一个用于入站检测\n出站：i-011e73af27562827b 正在对 172.16.0.24 进行 RDP 蛮力攻击\n入站：172.16.0.25 正在对 i-0191573dec3b66924 进行 RDP 蛮力攻击\n检测结果类型：UnauthorizedAccess:EC2/RDPBruteForce\n\n测试 4：加密货币活动\n预期的检测结果：EC2 实例 i-011e73af27562827b 正在查询与比特币活动相关的域名\n检测结果类型：CryptoCurrency:EC2/BitcoinTool.B!DNS\n\n测试 5：DNS 渗透\n预期的检测结果：EC2 实例 i-011e73af27562827b 正在尝试查询与泄露数据相似的域名\n检测结果类型：Trojan:EC2/DNSDataExfiltration\n\n测试 6：C&amp;C 活动\n预期检测结果：EC2 实例 i-011e73af27562827b 正在查询与已知命令与控制服务器关联的域名。\n检测结果类型：Backdoor:EC2/C&amp;CActivity.B!DNS\n</code></pre>\n<p>几分钟后，检测结果将显示在 GuardDuty 控制台中。在顶部，我看到了新的恶意软件保护功能发现的恶意文件。其中一个检测结果与 EC2 实例相关，另一个与 ECS 集群相关。</p>\n<p><img src=\"https://dev-media.amazoncloud.cn/011b2459766242b082512373132a8e5e_image.png\" alt=\"image.png\" /></p>\n<p>首先，我选择与 EC2 实例相关的检测结果。在面板中，我看到有关实例和恶意文件的信息，例如文件名和路径。在 Malware scan details（恶意软件扫描详细信息）部分中，Trigger finding ID（触发检测结果 ID）指向触发恶意软件扫描的原始 GuardDuty 检测结果。就我而言，最初的检测结果是此 EC2 实例正在对另一个 EC2 实例执行 RDP 蛮力攻击。</p>\n<p><img src=\"https://dev-media.amazoncloud.cn/11848354ea9c42949b7903e289fc67fc_image.png\" alt=\"image.png\" /></p>\n<p>在此处，我选择 <strong>Investigate with Detective</strong>（使用 Detective 调查），然后直接从 GuardDuty 控制台转到 <a href=\"https://console.aws.amazon.com/detective\" target=\"_blank\">Detective 控制台</a>，从中可视化 EC2 实例、Amazon 账户和受检测结果影响的 IP 地址的 <a href=\"https://aws.amazon.com/cloudtrail/\" target=\"_blank\">Amazon CloudTrail</a> 和 <a href=\"https://aws.amazon.com/vpc/\" target=\"_blank\">Amazon Virtual Private Cloud (Amazon VPC)</a> 流数据。使用 Detective，我可以分析、调查和确定 GuardDuty 发现的可疑活动的根本原因。</p>\n<p><img src=\"https://dev-media.amazoncloud.cn/b1806092ac054173a9a66d7cc11c7875_image.png\" alt=\"image.png\" /></p>\n<p>选择与 ECS 群集相关的检测结果时，我将获得有关受影响资源的更多信息，例如 ECS 集群、任务、容器和容器映像的详细信息。</p>\n<p><img src=\"https://dev-media.amazoncloud.cn/fbe69d5f07be4ee896817493be048a0d_image.png\" alt=\"image.png\" /></p>\n<p>使用 GuardDuty 测试器脚本可以更轻松地测试 GuardDuty 与您使用的其他安全框架的整体集成，以便在检测到真正的威胁时做好准备。</p>\n<h4><a id=\"_GuardDuty__Amazon_Inspector__324\"></a><strong><ins>将 GuardDuty 恶意软件防护保护与 Amazon Inspector 进行比较</ins></strong></h4>\n<p>此时，您可能会问自己 GuardDuty 恶意软件保护与 <a href=\"https://aws.amazon.com/inspector/\" target=\"_blank\">Amazon Inspector</a> 有何关系，Amazon Inspector 是一项扫描 AWS 工作负载的软件漏洞和意外网络暴露的服务。这两种服务相辅相成，提供不同的保护层：</p>\n<ul>\n<li>Amazon Inspector 通过识别和修复已知软件和应用程序漏洞提供主动防护，这些漏洞是攻击者入侵资源和安装恶意软件的入口点。</li>\n<li>GuardDuty 恶意软件保护可检测到活跃运行的工作负载中存在的恶意软件。此时，系统已经受损，但是 GuardDuty 可以限制感染的时间，并在系统受损导致影响业务的事件之前采取措施。</li>\n</ul>\n<h4><a id=\"_329\"></a><strong><ins>可用性和定价</ins></strong></h4>\n<p><a href=\"https://aws.amazon.com/guardduty/\" target=\"_blank\">Amazon GuardDuty</a> 恶意软件保护现已在<a href=\"https://aws.amazon.com/about-aws/global-infrastructure/regional-product-services/\" target=\"_blank\">所有提供 GuardDuty 的 Amazon 区域</a>推出，不包括 AWS 中国（北京）、Amazon 中国（宁夏）、Amazon GovCloud（美国东部）和 Amazon GovCloud（美国西部）区域。</p>\n<p>在发布时，GuardDuty 恶意软件保护已与以下合作伙伴产品集成：</p>\n<ul>\n<li><a href=\"https://www.bitdefender.com/\" target=\"_blank\">BitDefender</a></li>\n<li><a href=\"https://www.cloudhesive.com/\" target=\"_blank\">CloudHesive</a></li>\n<li><a href=\"http://www.crowdstrike.com/\" target=\"_blank\">Crowdstrike</a></li>\n<li><a href=\"https://www.fortinet.com/\" target=\"_blank\">Fortinet</a></li>\n<li><a href=\"https://www.paloaltonetworks.com/\" target=\"_blank\">Palo Alto Networks</a></li>\n<li><a href=\"https://www.rapid7.com/\" target=\"_blank\">Rapid7</a></li>\n<li><a href=\"https://www.sophos.com/\" target=\"_blank\">Sophos</a></li>\n<li><a href=\"https://sysdig.com/\" target=\"_blank\">Sysdig</a></li>\n<li><a href=\"https://www.trellix.com/\" target=\"_blank\">Trellix</a></li>\n</ul>\n<p>使用 GuardDuty，您无需部署安全软件或代理来监控恶意软件。您只需为文件系统中扫描的 GB 量（而不是 EBS 卷的大小）和 EBS 快照保存在您账户中的时间付费。除非您在发现恶意软件时启用快照保留功能，否则 GuardDuty 创建的所有 EBS 快照都会在扫描后自动删除。有关更多信息，请参阅 <a href=\"https://aws.amazon.com/guardduty/pricing/\" target=\"_blank\">GuardDuty 定价</a>和 <a href=\"https://aws.amazon.com/ebs/pricing/\" target=\"_blank\">EBS 定价</a>。请注意，GuardDuty 仅扫描小于 1 TB 的 EBS 卷。为了帮助您控制成本并避免重复警报，同一卷的扫描频率不会超过每 24 小时一次。</p>\n<h4><a id=\"_Amazon_GuardDuty_httpsawsamazoncomguarddutyfaqsGuardDuty_Malware_Protection_346\"></a><strong><a href=\"https://aws.amazon.com/guardduty/faqs/#GuardDuty_Malware_Protection\" target=\"_blank\">使用 Amazon GuardDuty 检测恶意活动并保护您的应用程序免受恶意软件侵害。</a></strong></h4>\n<p>— <a href=\"https://twitter.com/danilop\" target=\"_blank\">Danilo</a></p>\n"}