CVE-2007-3279 in PostgreSQL
Summary
by MITRE
PostgreSQL 8.1 and probably later versions, when the PL/pgSQL (plpgsql) language has been created, grants certain plpgsql privileges to the PUBLIC domain, which allows remote attackers to create and execute functions, as demonstrated by functions that perform local brute-force password guessing attacks, which may evade intrusion detection.
If you want to get the best quality for vulnerability data then you always have to consider VulDB.
Analysis
by VulDB Data Team • 07/26/2019
PostgreSQL version 8.1 and subsequent releases contained a significant privilege escalation vulnerability within their PL/pgSQL language implementation that exposed the system to remote exploitation. This flaw occurred when the plpgsql language was created within the database environment, automatically granting extensive privileges to the PUBLIC role, which represents all database users. The vulnerability stemmed from improper access control mechanisms that failed to properly isolate the privileges associated with procedural language execution from the broader user base. Attackers could exploit this weakness by leveraging the elevated privileges granted to PUBLIC to create malicious functions that would execute with elevated permissions. The technical implementation of this vulnerability involved the creation of PL/pgSQL functions that could perform operations typically restricted to database administrators, including local brute-force password guessing attacks that would otherwise be prevented by proper access controls.
The operational impact of this vulnerability was particularly severe as it allowed remote attackers to bypass traditional intrusion detection systems that would normally flag suspicious activities. The privilege escalation occurred at the database level, enabling attackers to execute functions with elevated privileges without requiring direct database administrator credentials. This created a stealthy attack vector where malicious activities could be performed under the guise of legitimate database operations, making detection significantly more challenging. The vulnerability specifically enabled the execution of local brute-force password guessing attacks, which could potentially compromise user accounts and escalate privileges to administrative levels. This type of attack could be particularly devastating in environments where database users had access to sensitive data and where the database server was accessible from external networks.
Security professionals should understand that this vulnerability represents a classic case of improper privilege management within database systems, aligning with CWE-276, which addresses incorrect permissions for creating resources. The attack pattern follows techniques described in the MITRE ATT&CK framework under privilege escalation and defense evasion tactics, where adversaries leverage legitimate system access to perform malicious activities while avoiding detection. Organizations should implement immediate mitigations including restricting PL/pgSQL language creation permissions, reviewing and revoking unnecessary privileges from the PUBLIC role, and implementing network segmentation to limit access to database servers. The vulnerability also highlights the importance of proper least privilege principles in database administration and demonstrates how default configurations can create security risks that persist across multiple versions of database software. Database administrators should conduct thorough privilege audits and ensure that language extensions are only created with explicit administrative authorization to prevent unauthorized function creation and execution.