Description
A flaw was found in the Red Hat OpenShift AI (RHOAI) overlay for the training operator. The RHOAI overlay incorrectly aggregates `trainjobs` management permissions into the native Kubernetes `edit ClusterRole`. This allows any user with `edit ClusterRole` permissions in a namespace to create, modify, and delete `TrainJobs`. When combined with a separate vulnerability (TRN-01) that permits arbitrary pod configurations, a remote attacker with namespace editor privileges could exploit this to escalate privileges, potentially leading to arbitrary code execution.
Published: 2026-08-10
Score: 8.8 High
EPSS: < 1% Very Low
KEV: No
Impact: Privilege Escalation via RBAC Misconfiguration
Action: Apply Workaround
AI Analysis

Impact

The vulnerability in Red Hat OpenShift AI arises from an incorrect aggregation of TrainJobs management permissions into the native Kubernetes edit ClusterRole. The flaw allows any user who has edit privileges in a namespace to create, modify, or delete TrainJobs. When combined with another known issue that permits arbitrary pod configurations, an attacker who is a namespace editor could potentially elevate privileges and achieve arbitrary code execution. The weakness is an instance of improper authorization with a CVSS score of 8.8, indicating a high severity.

Affected Systems

Red Hat OpenShift AI (RHOAI) overlay for the training operator is affected. The vulnerability applies to Red Hat OpenShift AI 3.3, 3.4, 3.5 on RHEL 9.

Risk and Exploitability

The CVSS score of 8.8 signals a high impact vulnerability, while the EPSS score of 0.00894 (<1%) indicates a low probability of exploitation. The flaw is not listed in CISA’s KEV catalog. The likely attack vector is within a namespace, exploiting Kubernetes RBAC for users with edit role privileges; a remote attacker with such permissions could exploit the design flaw to manipulate TrainJobs and potentially gain higher privileges.

Generated by OpenCVE AI on September 28, 2026 at 15:42 UTC.

Remediation

Vendor Workaround

Administrators should review and adjust their Kubernetes RBAC configurations within Red Hat OpenShift AI to ensure that `trainjobs` permissions are explicitly managed. This involves removing `trainjobs` from the `aggregate-to-edit` ClusterRole labels or requiring explicit `RoleBinding` for `trainjobs` access. This prevents implicit permission grants to namespace editors and reduces the attack surface. Consult Kubernetes documentation for specific instructions on modifying ClusterRoles and RoleBindings. A restart or reload of affected components may be required for changes to take effect.

OpenCVE Recommended Actions

  • Review Kubernetes RBAC configurations and remove trainjobs from the aggregate‑to‑edit ClusterRole or enforce explicit RoleBinding for trainjobs access.
  • Check Red Hat for an updated RHOAI release that corrects the permission aggregation and apply the vendor patch as soon as it becomes available.
  • After modifying RBAC, restart or reload the affected components to ensure the new permissions take effect.

Generated by OpenCVE AI on September 28, 2026 at 15:42 UTC.

Tracking

Sign in to view the affected projects.

Advisories

No advisories yet.

History

Wed, 30 Sep 2026 16:45:00 +0000

Type Values Removed Values Added
References

Mon, 28 Sep 2026 13:00:00 +0000

Type Values Removed Values Added
Weaknesses CWE-285

Mon, 28 Sep 2026 10:00:00 +0000

Type Values Removed Values Added
Weaknesses CWE-267

Mon, 21 Sep 2026 10:45:00 +0000

Type Values Removed Values Added
CPEs cpe:/a:redhat:openshift_ai:3.5::el9
References

Thu, 27 Aug 2026 16:45:00 +0000

Type Values Removed Values Added
References

Wed, 12 Aug 2026 16:30:00 +0000

Type Values Removed Values Added
Metrics ssvc

{'options': {'Automatable': 'no', 'Exploitation': 'none', 'Technical Impact': 'total'}, 'version': '2.0.3'}

Tue, 11 Aug 2026 19:00:00 +0000

Type Values Removed Values Added
CPEs cpe:/a:redhat:openshift_ai:3.4::el9
References

Tue, 11 Aug 2026 07:45:00 +0000

Type Values Removed Values Added
First Time appeared Redhat openshift Ai 3.3
Vendors & Products Redhat openshift Ai 3.3

Tue, 11 Aug 2026 05:15:00 +0000

Type Values Removed Values Added
CPEs cpe:/a:redhat:openshift_ai cpe:/a:redhat:openshift_ai:3.3::el9
References

Tue, 11 Aug 2026 00:15:00 +0000

Type Values Removed Values Added
References
Metrics threat_severity

None

 threat_severity

Critical

Mon, 10 Aug 2026 23:15:00 +0000

Type Values Removed Values Added
Weaknesses CWE-284
CWE-285

Mon, 10 Aug 2026 21:00:00 +0000

Type Values Removed Values Added
Description A flaw was found in the Red Hat OpenShift AI (RHOAI) overlay for the training operator. The RHOAI overlay incorrectly aggregates `trainjobs` management permissions into the native Kubernetes `edit ClusterRole`. This allows any user with `edit ClusterRole` permissions in a namespace to create, modify, and delete `TrainJobs`. When combined with a separate vulnerability (TRN-01) that permits arbitrary pod configurations, a remote attacker with namespace editor privileges could exploit this to escalate privileges, potentially leading to arbitrary code execution.
Title Odh-training-operator-rhel9: [trainer v2 security] trn-02: rhoai overlay aggregates trainjobs crud into standard edit clusterrole
First Time appeared Redhat
Redhat openshift Ai
CPEs cpe:/a:redhat:openshift_ai
Vendors & Products Redhat
Redhat openshift Ai
References
Metrics cvssV3_1

{'score': 8.8, 'vector': 'CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H'}

Subscriptions

Redhat Openshift Ai Openshift Ai 3.3
cve-icon MITRE

Status: PUBLISHED

Assigner: redhat

Published:

Updated: 2026-09-30T16:21:01.084Z

Reserved: 2026-08-05T13:41:38.511Z

Link: CVE-2026-18951

cve-icon Vulnrichment

Updated: 2026-08-12T15:37:06.267Z

cve-icon NVD

Status : Awaiting Analysis

Published: 2026-08-10T21:17:21.710

Modified: 2026-09-30T17:16:45.010

Link: CVE-2026-18951

cve-icon Redhat

Severity : Critical

Publid Date: 2026-08-10T18:54:28Z

Links: CVE-2026-18951 - Bugzilla

cve-icon OpenCVE Enrichment

Updated: 2026-09-28T15:45:02Z

Weaknesses
  • CWE-267

    Privilege Defined With Unsafe Actions

  • CWE-284

    Improper Access Control