Mlrun
Monthly
Server-side request forgery in MLRun through 1.11.0 lets an authenticated user who can create or update a run attach a malicious WebhookNotification that makes the API server issue arbitrary outbound HTTP requests once that run transitions to a terminal state. Because the request originates from inside the cluster, it can be aimed at internal services, the in-cluster Kubernetes API, or cloud metadata endpoints such as 169.254.169.254, although the SSRF is blind/semi-blind and practical impact depends on egress filtering and metadata access in the target deployment. Rated 5.3 (CVSS:4.0) with low confidentiality and integrity impact and no availability impact; no public exploit identified at time of analysis.
Server-side request forgery in MLRun through 1.11.0 lets an authenticated user who can create or update a run attach a malicious WebhookNotification that makes the API server issue arbitrary outbound HTTP requests once that run transitions to a terminal state. Because the request originates from inside the cluster, it can be aimed at internal services, the in-cluster Kubernetes API, or cloud metadata endpoints such as 169.254.169.254, although the SSRF is blind/semi-blind and practical impact depends on egress filtering and metadata access in the target deployment. Rated 5.3 (CVSS:4.0) with low confidentiality and integrity impact and no availability impact; no public exploit identified at time of analysis.