Facebook Google Plus Twitter LinkedIn YouTube RSS Menu Search Resource - BlogResource - WebinarResource - ReportResource - Eventicons_066 icons_067icons_068icons_069icons_070

Tenable 博客

订阅

CVE-2020-8597: Buffer Overflow Vulnerability in Point-to-Point Protocol Daemon (pppd)

Multiple widely used Linux distributions are impacted by a critical flaw that has existed in pppd for 17 years.

背景

On March 4, researchers at the CERT Coordination Center (CERT/CC) published vulnerability note #782301 for a critical vulnerability in the Point-to-Point Protocol Daemon (pppd) versions 2.4.2 through 2.4.8, with disclosure credited to Ilja van Sprundel of IOActive.

The Point-to-Point Protocol (PPP) is a full-duplex protocol that enables the encapsulation and transmission of basic data across Layer 2 or data-link services ranging from dial-up connections to DSL broadband to virtual private networks (VPNs) implementing SSL encryption. PPP is also used to implement IP and TCP over two directly connected nodes, as these protocols do not support point-to-point connections. pppd is a daemon on Unix-like operating systems used to manage PPP session establishment and session termination between two nodes.

分析

CVE-2020-8597 is a buffer overflow vulnerability in pppd due to a logic flaw in the packet processor of the Extensible Authentication Protocol (EAP). An unauthenticated, remote attacker who sends a specially crafted EAP packet to a vulnerable PPP client or server could cause a denial-of-service condition or gain arbitrary code execution. As pppd works in conjunction with kernel drivers and often runs with high privileges such as system or even root, any code execution could also be run with these same privileges.

According to CERT/CC’s vulnerability note, the logic flaw exists in several EAP functions.

In the eap_request and eap_response functions, a pointer and length are received as input using the first byte as a type. If this type is EAPT_MD5CHAP(4), it looks at an embedded 1-byte length field. This check was implemented to ensure the embedded length is smaller than that of the entire packet length. And if the check passes successfully, then the hostname located after the embedded length is copied into a local stack buffer. The vulnerability is in the logic of how these functions parse the code. If the bounds check is incorrect and proceeds to copy memory with an arbitrary length of data, a stack buffer overflow is possible.

The eap_input function contains an additional flaw in its code that fails to validate if EAP was negotiated during the Link Control Protocol (LCP) phase within PPP. Essentially, regardless of whether the failure to validate was the result of an incorrect pre-shared passphrase during the LCP phase or due to a lack of support for EAP, an unauthenticated attacker could send an EAP packet that would be processed. The processing of this unverified EAP packet can result in a stack buffer overflow.

概念验证

At the time this blog post was published, there was no working proof-of-concept (PoC) for this vulnerability.

However, multiple GitHub repositories have been published that may soon host a working PoC. One appears to be a work-in-progress, while another claims that a PoC will be released for this vulnerability “in a week or two when things die down.”

解决方案

The vulnerability was patched in eap.c on February 2. Various Linux distributions have since released updates to address the vulnerability in PPP and additional patches may be released in the coming days. The following is a list of known distribution releases that address this vulnerability:

Distribution Version Advisory
Debian Jessie LTS 2.4.6-3.1 DLA-2097-1
Debian Stretch, Debian Buster 2.4.7-1+4 2.4.7-2+4.1 DSA-4632-1
Ubuntu 19.10, 18.04 LTS, 16.04 LTS USN-4288-1
Ubuntu 14.04 ESM 12.04 ESM USN-4288-2
Red Hat Enterprise Linux 6 RHSA-2020:0631
Red Hat Enterprise Linux 7 RHSA-2020:0630
Red Hat Enterprise Linux 8 RHSA-2020:0634
SUSE SLED12 SUSE-SU-2020:0490-1
SUSE SLED15 SUSE-SU-2020:0489-1
CentOS 6 CESA-2020:0631
CentOS 7 CESA-2020:0630
Oracle Linux 6 ELSA-2020-0631
Oracle Linux 7 ELSA-2020-0630
Oracle Linux 8 ELSA-2020-0633

Additionally, Cisco has assigned CSCvs95534 as the bug ID associated with this vulnerability as it reviews the potential impact it may have on its products.

This is not an exhaustive list, and we anticipate more vendors will publish advisories as they determine the impact of this vulnerability on their products.

识别受影响的系统

A list of Tenable plugins to identify this vulnerability can be found here.

获取更多信息

加入 Tenable Community 中的 Tenable 安全响应团队

了解有关 Tenable 这款首创 Cyber Exposure 平台的更多信息,全面管理现代攻击面。

获取 30 天免费试用版 Tenable.io Vulnerability Management

相关文章

您是否易受最新漏洞利用的攻击?

输入您的电子邮件以在收件箱中接收最新的 Cyber Exposure 警报。

选择 Tenable.io

免费试用 30 天

可全面访问基于云的现代化漏洞管理平台,从而以无可比拟的精确度发现并追踪所有资产。 立即注册。

立即购买 Tenable.io

可全面访问基于云的现代化漏洞管理平台,从而以无可比拟的精确度发现并追踪所有资产。 立即购买年度订阅。

65 项资产

选择您的订阅选项:

立即购买

免费试用 Nessus Professional

免费试用 7 天

Nessus® 是当今市场上功能最全面的漏洞扫描器。Nessus Professional 可帮助自动化漏洞扫描流程、节省合规周期的时间,并让您调动起 IT 团队的积极性。

购买 Nessus Professional

Nessus® 是当今市场上功能最全面的漏洞扫描器。Nessus Professional 可帮助自动化漏洞扫描流程、节省合规周期的时间,并让您调动起 IT 团队的积极性。

购买多年期许可,即享优惠价格添加高级支持功能,获取一年 365 天、一天 24 小时的电话、社区和聊天支持。

选择您的许可证

购买多年期许可,即享优惠价格

添加支持和培训

试用 Tenable.io Web Application Scanning

免费试用 30 天

完整享有专为现代化应用程序而设、属于 Tenable.io 平台组成部分的最新 Web 应用程序扫描功能。可安全扫描全部在线资产组合的漏洞,具有高度准确性,而且无需繁重的手动操作或中断关键的 Web 应用程序。 立即注册。

购买 Tenable.io Web Application Scanning

可全面访问基于云的现代化漏洞管理平台,从而以无可比拟的精确度发现并追踪所有资产。 立即购买年度订阅。

5 个 FQDN

$3,578

立即购买

试用 Tenable.io Container Security

免费试用 30 天

完整获得已集成至漏洞管理平台之唯一容器安全产品的功能。监控容器映像中的漏洞、恶意软件和策略违规。与持续集成和持续部署 (CI/CD) 系统进行整合,以支持 DevOps 实践、增强安全性并支持企业政策合规。

购买 Tenable.io Container Security

Tenable.io Container Security 经由与构建流程的集成,可供全面了解容器映像的安全性,包括漏洞、恶意软件和策略违规,借以无缝且安全地启用 DevOps 流程。

获取 Tenable.sc 演示

请将您的联系方式填写在下方表格中,我们的销售代表很快与您联系安排演示。您也可以写下简短评论(不得超过 255 个字符)。请注意,带星号 (*) 的字段为必填项。

试用 Tenable Lumin

免费试用 30 天

通过 Tenable Lumin 直观呈现及探索 Cyber Exposure,长期追踪风险降低状况,并比照同行业者进行基准度量。

购买 Tenable Lumin

联系销售代表,了解 Lumin 如何帮助获取整个企业的洞见并管理网络安全风险。

申请演示 Tenable.ot

获得符合需求的运营技术安全性。
降低避之不及的风险。

Tenable.ad

持续检测并响应 Active Directory 攻击事件。无需代理。无需权限账号。本地私有化或云端部署。