Executive brief
Jupyter Enterprise Gateway is a tool used to manage and launch computing tasks across large clusters like Kubernetes and Apache Spark. A security flaw allows users to bypass safety restrictions and run these tasks with administrative (root) privileges. This could allow an attacker to escape their isolated environment, take control of the underlying server hardware, and potentially compromise the entire computing cluster and its data.
Technical details
An improper input validation vulnerability exists in the `ContainerProcessProxy._enforce_prohibited_ids` method of Jupyter Enterprise Gateway. The gateway attempts to block the launch of kernels with UID or GID 0 (root) by comparing user-supplied strings against a prohibited list; however, this check can be bypassed by adding trailing whitespace (e.g., '0 ') to the `KERNEL_UID` or `KERNEL_GID` environment variables. Because the downstream Jinja2 templates for Kubernetes manifests cast these values to integers, the whitespace is ignored during pod creation, resulting in a container running as root. An attacker can leverage these elevated privileges alongside hostPath volume mounts to achieve a container escape, leading to remote code execution on the worker node and potential lateral movement across the Kubernetes cluster. The issue is fixed in version 3.3.0.
Affected products
- jupyter-server enterprise_gateway >= 2.0.0rc1, < 3.3.0
Timeline
- 2026-06-03: advisory: GitHub Security Advisory GHSA-chq7-94j8-cj28 published
- 2026-07-16: disclosed: CVE-2026-44180 published to NVD
- 2022-06-01: patched: Version 3.3.0 released with fix