A blind SQL injection vulnerability in Cacti, a widely-used network monitoring, performance and fault management framework, could lead to information disclosure and potentially remote code execution.

Cacti is often used in network operation centers of telecoms and web hosting providers, to collect network performance data and store it in RRDtool, a logging and graphing database and system that, through a web interface, creates graphical representations of the collected data.

CVE-2023-51448 is a vulnerability within Cacti’s SNMP Notification Receivers feature that could allow a threat actor to disclose all Cacti database contents or, depending on the database configuration, even trigger remote code execution.

“An authenticated attacker with the ‘Settings/Utilities’ permission can send a crafted HTTP GET request to the endpoint ‘/cacti/managers.php’ with an SQLi payload in the ‘selected graphs array’ HTTP GET parameter,” Cacti noted in the security advisory.

The Cacti maintainers have outlined a possible attack and say that a PoC script for testing it is available “Upon request”.

A year ago, internet-exposed Cacti servers were targeted by attackers wielding an exploit for CVE-2022-46169, a critical command injection flaw that could be exploited remotely by unauthenticated users.