CVE-2024-1558: Path Traversal in MLflow

The path traversal vulnerability in MLflow allows an attacker to bypass security checks and access files outside of the intended directory. By crafting a malicious source parameter, an attacker can manipulate the model version creation process and read arbitrary files on the server hosting the MLflow tracking server. This could expose sensitive data such as configuration files, credentials, or even source code. The potential impact extends beyond simple data exposure; an attacker could potentially use this access to further compromise the system or gain a foothold for other attacks. The ability to read arbitrary files represents a significant security risk, particularly in environments where MLflow is used to manage sensitive machine learning models and data.

1. Reserved2024-02-15
2. Published2024-04-16
3. Modified2025-02-04
4. EPSS updated2026-04-02

The primary mitigation for CVE-2024-1558 is to upgrade MLflow to version 2.12.1 or later, which includes the necessary fix. If upgrading immediately is not feasible, consider implementing temporary workarounds. Restrict access to the MLflow tracking server to only authorized users and systems. Implement strict input validation on the source parameter to prevent malicious input. Monitor MLflow logs for suspicious activity, particularly attempts to access files outside of the expected directories. Consider using a Web Application Firewall (WAF) to filter out malicious requests targeting the vulnerability. After upgrading, confirm the fix by attempting to create a model version with a crafted source parameter containing relative path traversal sequences (e.g., ../sensitive_file.txt).