GHSA-M23H-6MWM-39M8
Vulnerability from github – Published: 2026-05-19 19:30 – Updated: 2026-05-19 19:30Summary
A vulnerability in the Kong Ingress Controller (KIC) allows for the unauthorized exfiltration of TLS certificates and private keys across Kubernetes namespace boundaries. In "managed" mode (where the GatewayClass lacks an unmanaged annotation), the Gateway TLS translator skips critical status checks. This bypass allows the translator to fetch Secrets from any namespace KIC watches, even when a ReferenceGrant explicitly denies access or is missing.
An actor with RBAC permissions to create or modify Gateways in a low-privileged namespace can reference a Secret in a high-privileged namespace, causing KIC to "leak" that Secret's sensitive private key material into the Kong dataplane configuration.
Am I affected?
You are affected if all of these hold:
1. You are using Kong Ingress Controller with the Gateway API.
2. Your GatewayClass is operating in managed mode (default behavior, no unmanaged annotation).
3. KIC is configured to watch multiple namespaces (multi-tenant environment).
4. Users have RBAC permissions to create or update gateways.gateway.networking.k8s.io in their own namespaces.
You are not affected if any of this:
- You only use KIC for Ingress resources (not Gateway API).
- Your GatewayClass uses the konghq.com/gateway-unmanaged annotation.
- KIC is restricted via RBAC or configuration to only watch a single namespace.
- You have strictly limited Gateway creation/modification permissions to trusted cluster administrators only.
Mitigation
- Add unmanaged gateway annotation: add the
konghq.com/gateway-unmanagedannotation to yourGatewayClass
Additional best practicies
- Restrict Gateway RBAC: Limit the ability to create or modify Gateway resources to high-trust administrative users until a patch is applied.
- Namespace Isolation: If possible, limit the namespaces KIC is permitted to watch using the
WATCH_NAMESPACEenvironment variable or specific RBAC RoleBindings.
Fix
The fix mandates ReferenceGrant validation for all cross-namespace certificate references. By requiring a Programmed: True listener status, the translator now strictly authorizes external Secret access while maintaining default access for same-namespace certificates, effectively closing the exfiltration vector.
Fixed in #7920, with backports to supported release branches in #7921 and #7922.
Upgrade to one of the following patched versions (or later): - 3.4.14 - 3.5.7
CVSS
CVSS:4.0/AV:N/AC:L/AT:P/PR:H/UI:N/VC:H/VI:N/VA:N/SC:H/SI:N/SA:N/E:P = 5.6 Medium
{
"affected": [
{
"database_specific": {
"last_known_affected_version_range": "\u003c= 3.5.6"
},
"package": {
"ecosystem": "Go",
"name": "github.com/kong/kubernetes-ingress-controller/v3"
},
"ranges": [
{
"events": [
{
"introduced": "3.5.0"
},
{
"fixed": "3.5.7"
}
],
"type": "ECOSYSTEM"
}
]
},
{
"database_specific": {
"last_known_affected_version_range": "\u003c= 3.4.13"
},
"package": {
"ecosystem": "Go",
"name": "github.com/kong/kubernetes-ingress-controller/v3"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "3.4.14"
}
],
"type": "ECOSYSTEM"
}
]
},
{
"package": {
"ecosystem": "Go",
"name": "github.com/kong/kubernetes-ingress-controller/v2"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"last_affected": "2.12.8"
}
],
"type": "ECOSYSTEM"
}
]
},
{
"package": {
"ecosystem": "Go",
"name": "github.com/kong/kubernetes-ingress-controller"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"last_affected": "1.3.4"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [],
"database_specific": {
"cwe_ids": [
"CWE-200"
],
"github_reviewed": true,
"github_reviewed_at": "2026-05-19T19:30:23Z",
"nvd_published_at": null,
"severity": "MODERATE"
},
"details": "## Summary\n\nA vulnerability in the Kong Ingress Controller (KIC) allows for the unauthorized exfiltration of TLS certificates and private keys across Kubernetes namespace boundaries. In \"managed\" mode (where the `GatewayClass` lacks an unmanaged annotation), the Gateway TLS translator skips critical status checks. This bypass allows the translator to fetch Secrets from any namespace KIC watches, even when a `ReferenceGrant` explicitly denies access or is missing.\n\nAn actor with RBAC permissions to create or modify Gateways in a low-privileged namespace can reference a Secret in a high-privileged namespace, causing KIC to \"leak\" that Secret\u0027s sensitive private key material into the Kong dataplane configuration.\n\n## Am I affected?\n\nYou are affected if all of these hold:\n1. You are using Kong Ingress Controller with the **Gateway API**.\n2. Your `GatewayClass` is operating in **managed mode** (default behavior, no unmanaged annotation).\n3. KIC is configured to **watch multiple namespaces** (multi-tenant environment).\n4. Users have RBAC permissions to `create` or `update` `gateways.gateway.networking.k8s.io` in their own namespaces.\n\nYou are not affected if any of this:\n- You only use KIC for `Ingress` resources (not Gateway API).\n- Your `GatewayClass` uses the `konghq.com/gateway-unmanaged` annotation.\n- KIC is restricted via RBAC or configuration to only watch a single namespace.\n- You have strictly limited Gateway creation/modification permissions to trusted cluster administrators only.\n\n## Mitigation\n\n1. **Add unmanaged gateway annotation**: add the `konghq.com/gateway-unmanaged` annotation to your `GatewayClass`\n\n### Additional best practicies\n\n1. **Restrict Gateway RBAC**: Limit the ability to create or modify Gateway resources to high-trust administrative users until a patch is applied.\n2. **Namespace Isolation**: If possible, limit the namespaces KIC is permitted to watch using the `WATCH_NAMESPACE` environment variable or specific RBAC RoleBindings.\n\n## Fix\n\nThe fix mandates `ReferenceGrant` validation for all cross-namespace certificate references. By requiring a `Programmed: True` listener status, the translator now strictly authorizes external Secret access while maintaining default access for same-namespace certificates, effectively closing the exfiltration vector.\n\nFixed in [#7920](https://github.com/Kong/kubernetes-ingress-controller/pull/7920), with backports to supported release branches in [#7921](https://github.com/Kong/kubernetes-ingress-controller/pull/7921) and [#7922](https://github.com/Kong/kubernetes-ingress-controller/pull/7922).\n\nUpgrade to one of the following patched versions (or later):\n- **3.4.14**\n- **3.5.7**\n\n## CVSS\n\n`CVSS:4.0/AV:N/AC:L/AT:P/PR:H/UI:N/VC:H/VI:N/VA:N/SC:H/SI:N/SA:N/E:P` = **5.6 Medium**",
"id": "GHSA-m23h-6mwm-39m8",
"modified": "2026-05-19T19:30:23Z",
"published": "2026-05-19T19:30:23Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/Kong/kubernetes-ingress-controller/security/advisories/GHSA-m23h-6mwm-39m8"
},
{
"type": "WEB",
"url": "https://github.com/Kong/kubernetes-ingress-controller/pull/7920"
},
{
"type": "WEB",
"url": "https://github.com/Kong/kubernetes-ingress-controller/pull/7921"
},
{
"type": "WEB",
"url": "https://github.com/Kong/kubernetes-ingress-controller/pull/7922"
},
{
"type": "PACKAGE",
"url": "https://github.com/Kong/kubernetes-ingress-controller"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:4.0/AV:N/AC:L/AT:P/PR:H/UI:N/VC:H/VI:N/VA:N/SC:H/SI:N/SA:N/E:P",
"type": "CVSS_V4"
}
],
"summary": "Kong Ingress Controller for Kubernetes (KIC): Cross-namespace TLS Secret Exfiltration in Gateways with GatewayClass missing `konghq.com/gatewayclass-unmanaged: \u0027true\u0027` annotation"
}
Drag nodes to arrange them. Double-click a related vulnerability to open it, or a source to visit its link. The Physics button on the left tunes the layout and the simulation; the legend at the bottom left hides or shows nodes by kind and by sighting type.
Sightings
| Author | Source | Type | Date | Other |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or observed by the user.
- Confirmed: The vulnerability has been validated from an analyst's perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
- Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
- Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
- Not confirmed: The user expressed doubt about the validity of the vulnerability.
- Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.
Vulnerability timeline
Sightings and recorded dates, grouped by day · UTC
How durations are calculated
Following the VTEM framework (Black Hat USA 2015), time to patch starts when the vendor or a coordinator was informed, total time to patch at discovery, and disclosure to patch at public disclosure; vendor response time runs from the report to its acknowledgement. Time to exploit and time unpatched start at the earliest discovery or report, falling back to public disclosure. Unpatched time ends at the first recorded patch, or the calculation time when no patch is recorded. Exploitation to patch shows the interval between those two events; a published proof of concept is a public exploit, not exploitation.
Each result shows its source dates. A negative interval means the end event came first; when date-only inputs leave the order open, the result says so. Date-only inputs are approximate. The earliest evidence wins: KEV report dates and sightings are labelled as such. Records describing another vulnerability are listed but never calculated from, and a PoC alone does not establish exploitation. Missing patch evidence does not prove that no patch exists. Filters below change the event list only.
Loading timeline…
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.
Related by attack behaviour
Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.