GPU VulnDB

Database/AI/ML frameworks & serving

MLflow (tracking server): Path traversal (`\..\filename`)

CVE-2023-1177AI/ML frameworks & servingcurated

Impact

Path traversal (\..\filename) → arbitrary file read

Who can reach it

Unauthenticated network to the MLflow tracking server

What to do

Upgrade to 2.2.1+; MLflow has no auth by default

Fleet impact

How widespread

Common - MLflow is the default experiment/model registry alongside GPU training clusters

Cost to remediate

daemon-restart - upgrade the tracking server; the pain is credential rotation, since leaked SSH keys and cloud creds must be assumed compromised

Why it hits the whole fleet

Unauthenticated LFI via the Model Versions API reads any file the server can read - SSH keys, cloud credentials - and one tracking server usually fronts every team's models and artifact buckets

References

This entry is curated: imported from vendor advisories with machine assistance, not yet individually verified. Confirm against your vendor's advisory before acting, and report anything wrong.