MLflow Uncontrolled Resource Consumption vulnerability
Published 20 Mar 2025 · Updated 7 Jul 2026
Is your project on an affected mlflow?
Look after several Python apps? Indie watches up to 15 repos, private ones too, for $5 a month. See Indie
Affected package
- Affected
- <=2.17.2
- Fixed in
- No fix released
Details
In mlflow/mlflow version 2.17.2, the /graphql endpoint is vulnerable to a denial of service attack. An attacker can create large batches of queries that repeatedly request all runs from a given experiment. This can tie up all the workers allocated by MLFlow, rendering the application unable to respond to other requests. This vulnerability is due to uncontrolled resource consumption.
Is GHSA-49m6-vrr9-2cqm in any of your client repos?
Checking one package by hand is fine. Checking it across forty client projects is not. NPMscan reads the lockfiles in every GitHub repo you pick, finds each project with an affected mlflow, and gives you the command that fixes it.
More advisories for mlflow
- HighCVE-2026-116690MLflow vulnerable to arbitrary file read on unauthenticated servers via prompt model version source · 27 Oct 2026
- MediumCVE-2026-116691MLflow CrewAI autologging records LLM provider API keys in logged trace spans · 27 Oct 2026
- CriticalCVE-2026-116692MLflow vulnerable to remote code execution on unauthenticated servers via third-party scorer deserialization importing caller-placed artifact modules · 27 Oct 2026
- MediumCVE-2026-115886MLflow: SSRF on unauthenticated MLflow servers via third-party scorer kwargs in the scorer invoke, register, and GenAI evaluate endpoints · 23 Oct 2026
- HighCVE-2026-53995MLflow FileStore trace IDs allow path traversal and arbitrary file writes · 20 Oct 2026
- HighGHSA-gqvg-gmmx-x4hmMLFLOW_ALLOW_PICKLE_DESERIALIZATION=False safety control bypassed by mlflow.statsmodels flavor — RCE via crafted model artifact · 1 Sept 2026
References
Advisory data from OSV.dev (opens OSV.dev in a new tab), refreshed daily.