Horizon Alert
Summary of the vulnerability and why it matters
A security vulnerability has been identified in Kyverno affecting how it validates resource paths, potentially allowing some policies to access information outside their designated namespaces. This could enable unauthorized users with specific permissions to interact with sensitive cluster resources. The primary concern is to determine if your environment utilizes the affected Kyverno versions and configurations.
- Policies can escape their intended scope.
- Allows access to sensitive cluster resources.
- Confirm relevance and exposure in your environment.
Attack Path
How an attacker could exploit the issue
An attacker with the ability to create NamespacedValidatingPolicy in Kyverno could leverage a flaw in how resource arguments are validated. By manipulating path segments within the resource argument, the attacker can bypass namespace restrictions and access arbitrary cluster objects. This allows them to read and write cluster resources, ultimately leading to a complete takeover of the Kubernetes cluster.
- Requires authenticated access to create policies.
- Uses path traversal to escape namespace.
- Enables arbitrary cluster object manipulation.
Live Threat
Current exploitation, exposure, and threat context
Kyverno policies, when configured to allow namespace escape, could enable unauthorized access to and modification of arbitrary cluster objects. This could occur when a tenant with the ability to create specific types of policies inserts directory traversal sequences into resource arguments, allowing them to interact with resources outside their designated namespace using the admission controller's service account.
- Arbitrary cluster objects.
- Malicious policy input.
- Cluster takeover.
Operational Fix
Recommended remediation, mitigation, and detection steps
The Kyverno admission controller's vulnerability requires immediate attention from Kubernetes platform teams and security operations. The first practical step is to identify all Kyverno deployments, assess their exposure to unauthenticated or low-privileged access that could trigger the vulnerability, and determine their criticality to cluster operations. This will enable proper risk-based prioritization and coordinated remediation planning with application owners.
- Platform and Security teams own.
- Verify resource.get()/post() reachability.
- Plan remediation with application owners.