top of page

Plague PAM Backdoor Malware Targets Linux Systems with Undetectable SSH Persistence

August 6th, 2025

Critical

Our Cyber Threat Intelligence Unit is monitoring a new Linux-focused malware strain, identified as “Plague.” This malware leverages a stealthy Pluggable Authentication module (PAM) backdoor to establish persistent SSH access and silently harvest credentials. The backdoor directly targets authentication mechanisms within Linux systems, modifying PAM components to bypass security controls and avoid detection. Plague’s modular design enables covert access to high-value Linux environments, including cloud-hosted assets and publicly exposed infrastructure, without leaving typical forensic traces. This campaign and its accompanying techniques mark a significant escalation in the abuse of authentication frameworks for persistent access.

Technical Details

  • Attack Type: Persistent Backdoor (Credential Theft & SSH Access).

  • Severity: Critical.

  • Technique: PAM Module Insertion, Anti-Forensic Obfuscation.

  • Malware Name: Plague.


Plague is designed for long-term stealth and persistence, embedding itself within Linux authentication workflows by modifying or replacing legitimate PAM modules. Once deployed, the malware intercepts login attempts to capture user credentials, which are exfiltrated to the attacker-controlled infrastructure.


To maintain uninterrupted SSH access, Plague installs a covert backdoor that allows privileged logins without appearing in standard authentication logs. The malware uses multiple layers of evasion, including anti-debugging routines, binary obfuscation, and forensic interference techniques. Notably, several samples have avoided detection by antivirus solutions for over a year, highlighting the sophistication of its anti-analysis capabilities.

Image by ThisisEngineering

Impact

Plague poses a critical threat to enterprise Linux environments. Key risks include:

  • Stealthy Persistence: Allows remote access via SSH without triggering standard authentication logs.

  • Credential Compromise: Silently captures usernames and passwords during login events.

  • Privilege Escalation & Lateral Movement: Facilitates complete system takeover in shared or high-privilege environments.

  • Defense Evasion: Bypasses antivirus and EDR detection via direct PAM integration and multi-layer obfuscation.

  • Reconnaissance & Exfiltration: Supports long-term credential harvesting and data theft, with emphasis on cloud and critical infrastructure targets.

Given its persistence, Plague can remain undetected for extended periods, significantly increasing incident response complexity and organizational risk.

Detection Method

Detecting Plague infections involves combining file integrity monitoring, behavioral analysis, and network telemetry. Recommended approaches include:

  • Baseline Comparison: Verify PAM module files and configurations against trusted baselines.

  • Hash Verification: Use cryptographic hashes to verify module integrity.

  • Authentication Anomalies: Identify and analyze SSH logins that bypass PAM logging or rely on static, backdoor credentials.

  • EDR Visibility: Employ EDR tools capable of detecting credential interception and unauthorized PAM hooks.

  • Binary Analysis: Flag PAM modules with unusual compile timestamps, obfuscated strings, or packed binaries

  • Outbound Communication: Monitor for suspicious network activity, particularly encrypted traffic to unrecognized external servers indicative of credential exfiltration.

Indicators of Compromise

Type 

Indicator 

SHA256 Hash 

85c66835657e3ee6a478a2e0b1fd3d87119bebadc43a16814c30eb94c53766bb 

SHA256 Hash 

7c3ada3f63a32f4727c62067d13e40bcb9aa9cbec8fb7e99a319931fc5a9332e 

SHA256 Hash 

9445da674e59ef27624cd5c8ffa0bd6c837de0d90dd2857cf28b16a08fd7dba6 

SHA256 Hash 

5e6041374f5b1e6c05393ea28468a91c41c38dc6b5a5230795a61c2b60ed14bc 

SHA256 Hash 

6d2d30d5295ad99018146c8e67ea12f4aaa2ca1a170ad287a579876bf03c2950 

SHA256 Hash 

e594bca43ade76bbaab2592e9eabeb8dca8a72ed27afd5e26d857659ec173261

SHA256 Hash 

14b0c90a2eff6b94b9c5160875fcf29aff15dcfdfd3402d953441d9b0dca8b39 

 

mix of red, purple, orange, blue bubble shape waves horizontal for cybersecurity and netwo

Recommendations

Organizations should implement the following measures to detect, contain, and prevent Plague infections:

  • PAM Integrity Validation: Verify all PAM module files and configurations against established baselines.

  • Isolation and Triage: Immediately isolate affected systems for forensic analysis upon detection.

  • File Monitoring: Continuously monitor PAM files for unauthorized modifications or unexpected binaries.

  • Credential Hardening: Enforce multi-factor authentication (MFA) for all privileged and administrative accounts.

  • Behavioral Detection: Deploy EDR solutions that can inspect in-memory modules and identify credential interception techniques.

  • Access Control: Limit SSH access by using strict allow-lists, segmentation, and restricted user roles.

  • Lateral Movement Mitigation: Restrict user permissions and segregate high-privilege systems to minimize post-compromise movement.

  • Threat Intelligence Integration: Keep up to date on emerging IOCs and tradecraft related to PAM exploitation.

Conclusion

The “Plague” Pluggable Access Module (PAM) backdoor campaign demonstrates how threat actors are evolving their methods to target trust in core system components, highlighting the importance of adopting a zero-trust approach to authentication systems. By integrating itself within the PAM authentication framework, Plague bypasses conventional security layers, allowing for the long-term compromise of Linux environments with minimal forensic evidence. We urge organizations to enforce a zero-trust approach to authentication and system-level module integrity, especially in critical infrastructure environments. Continuous PAM integrity validation, behavioral detection, and strict access control policies are essential for identifying and preventing such threats from gaining footholds within enterprise networks.

bottom of page