H2O.ai H2O-3 Deserialization Vulnerability in JDBC Connections Allowing Arbitrary Code Execution and File Access

A deserialization vulnerability has been identified in the H2O.ai H2O-3 repository, specifically in the latest master branch version 3.47.0.99999. This vulnerability allows attackers to exploit the deserialization of untrusted data, potentially leading to arbitrary code execution and unauthorized access to system files. The issue arises from the ability to bypass regular expression filters designed to prevent malicious parameter injection in JDBC connections. Attackers can manipulate spaces between parameters to evade detection, exploiting the vulnerability to read sensitive files and execute arbitrary commands on the system.

Impact

Exploitation of this vulnerability allows for arbitrary code execution and unauthorized reading of system files.

Reproduction

The vulnerability can be reproduced by sending a POST request to the '/99/ImportSQLTable' endpoint. The 'connection_url' parameter must be crafted to include malicious JDBC parameters, such as 'autoDeserialize=true' and 'allowLoadLocalInfile=true', while manipulating spaces to bypass regular expression filters. This crafted JDBC connection can then be used to exploit the deserialization vulnerability, leading to arbitrary code execution.

Remediation

Users are advised to update to H2O-3 version 3.46.0.8, where this vulnerability has been fixed.