Formal Verification for Preventing Misconfigured Access Policies in Kubernetes Clusters

Kubernetes clusters now underpin the bulk of modern production workloads, recent 2024 Cloud Native Computing Foundation surveys report >96% enterprise adoption, stretching from 5G edge nodes and AI/ML pipelines to heavily-regulated fintech and healthcare back-ends. Every action in those e...

Full description

Saved in:
Bibliographic Details
Main Authors: Aditya Sissodiya, Eric Chiquito, Ulf Bodin, Johan Kristiansson
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11122676/
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Kubernetes clusters now underpin the bulk of modern production workloads, recent 2024 Cloud Native Computing Foundation surveys report >96% enterprise adoption, stretching from 5G edge nodes and AI/ML pipelines to heavily-regulated fintech and healthcare back-ends. Every action in those environments funnels through the API server, so a single access-control slip can jeopardise an entire fleet. Yet most deployments still rely on a patchwork of Role-Based Access Control (RBAC) rules and policy-as-code admission controllers such as OPA Gatekeeper or Kyverno. In practice these controls are brittle: minor syntactic oversights, wildcard privileges, or conflicting rules can silently create privilege-escalation paths that elude linters and manual review. This paper presents a framework that models both RBAC and admission policies as first-order logic and uses an SMT solver to exhaustively search for counter-examples to stated security invariants before policies reach the cluster. The approach detects policy conflicts, unreachable denies, and unintended permissions. Three real-world case studies are presented to illustrate how the framework reveals latent misconfigurations and validates the soundness of the corrected rules. These case studies include a supply-chain image bypass, an RBAC “shadow-admi” escalation, and a multi-tenant namespace breach. To aid replication and further study, we release a fully scripted GitHub testbed: a Minikube cluster, AuthzForce PDP, admission-webhook adapter, and Z3-backed CLI that recreates each scenario and verifies policies end-to-end. While the framework does not address runtime threats, it closes a critical verification gap and substantially raises the bar for attackers targeting the most widely deployed orchestration platform.
ISSN:2169-3536