All Posts
No items found.
1
min read

RegreSSHion - CVE-2024-6387

Published on
July 8, 2024
Table of Contents

Introduction

OpenSSH is widely known and used software; therefore, the potential risk associated with the regreSSHion vulnerability is a key security factor to consider. As CISOs and analysts race to ensure that their systems are not affected by this vulnerability for compliance and regulatory purposes while simultaneously applying updates to OpenSSH version 9.8p1, we want to provide some information for our customers and the larger community as they traverse these tasks.

This blog will give an overview of the vulnerability, a breakdown of how it can be exploited, and a detection rule to locate the error message associated with regreSSHion. Anomali wants to help our customers detect this vulnerability as they apply the patch provided in the latest version of OpenSSH. Patches and updates can take time, so Anomali customers can use the detection rule in this blog to search for the vulnerability as they work through patching processes.

Overview

On July 1, 2024, the Qualys Threat Research Unit reported their discovery of a remote unauthenticated code vulnerability, registered as CVE-2024-6387, that affects OpenSSH versions up to 4.4p1 and 8.5p1 to 9.7p1 in glibc-based Linux systems.[1] The vulnerability, dubbed regreSSHion, is a signal handler race condition that resides in OpenSSH’s server (sshd). In addition, regreSSHion is a regression of a previously patched vulnerability, CVE-2006-5051, that was reintroduced in October 2020 with OpenSSH version 8.5p1. If successfully exploited, regreSSHion can result in arbitrary code being remoted executed on the affected system.

The impact could be significant given OpenSSH's integration into numerous systems and devices, though widespread exploitation is unlikely due to technical complexities.

Affected OpenSSH versions:

  • Versions before 4.4p1
  • Versions 8.5p1 to 9.7p1

CVE: CVE-2024-6387
Severity: High
CVSS 3.x Score: 8.1
Affected OpenSSH Versions: before 4.4p1; 8.5p1 to 9.7p1
NIST NVD DOP: July 1, 2024

How to Trigger regreSSHion

During SSH authentication, the user has a default time limit of 120 seconds to complete the process. If this time elapses without authentication, the sshd server calls the SIGALRM handler asynchronously.[2] This subsequently causes system-level memory management functions. Therefore, this execution is unsafe for asynchronous operation, and it may cause a race condition under specific conditions. This can lead to memory boundary violations and potentially allow for arbitrary code execution. An attacker would need to make around 10,000 attempts on average, targeting Linux systems using the GNU C Library (glibc) to attempt successful exploitation.[3]

Researchers have been able to reproduce regreSSHion on a 32-bit Linux/glibc system with address space layout randomization (ASLR), however, the attack required 6-8 hours on average with constant, maximum allowed server connections.[4]

Detecting regreSSHion in Security Analytics with Anomali Query Language 

Security Analytics (SA) users can detect the error message associated with the regreSSHion vulnerability with the following Anomali Query Language (AQL) syntax. 

eventlog
| where sourcetype contains_ci "auth.log"
| where message contains_ci "timeout before authentication"
| timechart count by message

 The AQL above searches through the auth.log for the error message ‘timeout before authentication’ that would be present if threat actors were attempting to exploit CVE-2024-6387. SA users simply need to direct this AQL to the correct log source for OpenSSH information.

FEATURED RESOURCES

November 4, 2026
Operationalized Threat Intelligence
Cyber Threat Intelligence

Crypto Exchange Security: The Financial Cost of Chronic Failures and the Regulatory Reckoning Ahead

The October 2026 Binance incident, in which approximately $6.3 billion in customer assets was affected and roughly $3 billion frozen across multiple mechanisms, is the largest single crypto theft on record. The number is arresting, but what it obscures matters more: the predictable output of a security model that has never been adequate to the assets it protects.
Read More
November 3, 2026
Anomali Cyber Watch

Anomali Cyber Watch: TraderTraitor Heist & Fake AI Malware

Threat intelligence briefing: Analyzing the $6.36B TraderTraitor Binance hack, fake signed OpenAI/Anthropic apps, and Ukrainian emergency app hacks.
Read More
October 28, 2026
Operationalized Threat Intelligence
Cyber Threat Intelligence

APT28 Spent 2 Months Inside a NATO Naval Shipbuilder — and Left Behind a Loader Nobody Has Seen Before

Read More
Explore All