CVE-2026-2393

[CVE catalog]

A flaw was found in MLflow. An authenticated attacker can exploit a Server-Side Request Forgery (SSRF) vulnerability by providing an unvalidated URL parameter to the `_create_webhook()` function. This allows the MLflow backend to be forced into sending HTTP requests to internal services, cloud metadata endpoints, or arbitrary external servers. The lack of input validation on the webhook URL enables this exploitation, potentially leading to cloud credential theft, internal network access, and data exfiltration. mlflow: MLflow: Server-Side Request Forgery (SSRF) allows internal network access and data exfiltration. This Important Server-Side Request Forgery (SSRF) vulnerability in MLflow does not affect Red Hat OpenShift AI (RHOAI) products as the vulnerable code is not present in the shipped components. The CVSS score(s) listed for this vulnerability do not reflect the associated product's status, and are included for informational purposes to better understand the severity of this vulnerability.

Severity

High · CVSS 8.5CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:L/A:N

Published

May 11, 2026, 4:30:04 PM UTC

Source

[NVD] · [CVE record]

EPSS

0.29% · Oct 27, 2026

{
  "package_releases": [],
  "fixed_versions": [],
  "distribution_packages": [
    {
      "name": "mlflow",
      "fixed_version": "3.9.0",
      "applicability_unknown": true
    }
  ]
}