Openshift api admission checks does not enforce "custom-host" permissions
Description
A flaw was found in OpenShift API, as admission checks do not enforce "custom-host" permissions. This issue could allow an attacker to violate the boundaries, as permissions will not be applied.
AI Insight
LLM-synthesized narrative grounded in this CVE's description and references.
OpenShift API admission checks fail to enforce custom-host permissions on ingress objects, allowing unauthorized custom host names.
Vulnerability
The OpenShift API admission checks do not enforce the "custom-host" permission for ingress objects under certain circumstances [1], [2]. This flaw occurs when a cluster administrator revokes the "custom-host" permission from project administrators, but the admission checks still allow those administrators to set custom host names on ingress objects. The issue affects OpenShift Container Platform versions before the fix was applied.
Exploitation
An attacker who is a project administrator with the "custom-host" permission revoked by the cluster administrator can bypass the admission check by creating or modifying an ingress object with a custom host name. The attacker does not require any special network position or additional authentication beyond being a project administrator.
Impact
Successful exploitation allows the attacker to set custom host names on ingress objects despite lacking the required permission, violating the intended security boundaries. This could lead to host name collisions, unauthorized service exposure, or other impacts depending on the cluster configuration.
Mitigation
The specific fixed version is not disclosed in the available references [1], [2]. Cluster administrators should apply the latest OpenShift Container Platform updates as recommended by Red Hat. As a workaround, ensure that the "custom-host" permission is managed appropriately, though this does not fully mitigate the vulnerability due to the admission check bypass.
AI Insight generated on May 26, 2026. Synthesized from this CVE's description and the cited reference URLs; citations are validated against the source bundle.
Affected products
6- cpe:/a:redhat:acm:2
- cpe:/a:redhat:ansible_automation_platform
- cpe:/a:redhat:ansible_tower:3
cpe:/a:redhat:openshift:3.11+ 1 more
- cpe:/a:redhat:openshift:3.11
- cpe:/a:redhat:openshift:4
Patches
0No patches discovered yet.
Vulnerability mechanics
AI mechanics synthesis has not run for this CVE yet.
References
2- access.redhat.com/security/cve/CVE-2022-3248mitrevdb-entryx_refsource_REDHAT
- bugzilla.redhat.com/show_bug.cgimitreissue-trackingx_refsource_REDHAT
News mentions
0No linked articles in our index yet.