GHSA-6x34-89p7-95wg
CRITICALRancher has downstream cluster privilege escalation through cluster and project role template binding (CRTB/PRTB)
EPSS Exploitation Probability
EPSS (Exploit Prediction Scoring System) is a daily probability model maintained by FIRST.org. It estimates the likelihood a CVE will be exploited in production environments within the next 30 days, derived from real-world threat intelligence signals.
Blast Radius
github.com/rancher/rancher🐹github.com/rancher/rancherReal-time download stats are indexed for npm and PyPI packages. This vulnerability affects Go packages — download data is not available via public APIs for these ecosystems.
Description
Impact
An issue was discovered in Rancher versions up to and including 2.5.15 and 2.6.6 where a flaw with authorization logic allows privilege escalation through cluster role template binding (CRTB) and project role template binding (PRTB). This issue does not affect the local cluster, it affects only downstream clusters.
The vulnerability can be exploited by any user who has permissions to create/edit cluster role template bindings or project role template bindings (such as cluster-owner, manage cluster members, project-owner and manage project members) to gain owner permission in another project in the same cluster or in another project on a different downstream cluster.
-
The user must have
kubectlaccess in the local cluster to exploit this scenario. -
This can only be abused to gain
cluster-ownerpermission on a different downstream cluster if the user is alreadycluster-owneron at least one downstream cluster. -
Example of a modified CRTB (note: the
clusterNamepoints to the cluster ID of the cluster that the privileges will be escalated andnamespacepoints to the current cluster ID that the user has permissions):
kubectl edit clusterroletemplatebindings crtb-<crtb-ID> -n c-<cluster-ID>
---
apiVersion: management.cattle.io/v3
clusterName: <ID-of-the-cluster-to-escalate>
kind: ClusterRoleTemplateBinding
metadata:
annotations:
<omitted>
finalizers:
- <omitted>
generateName: crtb-
labels:
<omitted>
cattle.io/creator: norman
name: crtb-<crtb-ID>
namespace: c-<current-cluster-ID>
roleTemplateName: cluster-owner
userName: u-<user-ID>
userPrincipalName: local://u-<user-ID>
An artifact to flag the exploitation of this issue is that the namespace of the CRTB/PRTB will not match the cluster name (clusterName) of the CRTB/PRTB. For example, every CRTB in the c-123xyz namespace should have a cluster name of c-123xyz. If instead, the cluster name is c-abc567, for example, this is likely a result of a user exploiting this flaw.
For more information about cluster and project roles, please consult Rancher's documentation.
Patches
Patched versions include releases 2.5.16, 2.6.7 and later versions.
Workarounds
Limit access in Rancher to trusted users. There is not a direct mitigation besides upgrading to the patched Rancher versions.
Important:
- It is highly advised to check the local and downstream clusters for potential unrecognized CRTBs (
kubectl get clusterroletemplatebindings -A) and PRTBs (kubectl get projectroletemplatebindings -A) assignments. - The ability to add other users to projects and clusters is a highly-privileged permission which may result in users being able to operate beyond their explicitly specified RBAC. It is recommended that this permission be granted selectively.
The following script can be used as a helper to detect possible deviations of CRTBs and PRTBs that do not match the expected value. Further investigation is required to determine if the flagged objects were maliciously modified or not. The script requires kubectl access to the local cluster and the jq command.
#!/usr/bin/env bash
echo "CRTBs that don't match cluster:"
kubectl get clusterroletemplatebindings -A -o=jsonpath="{range .items[?(@[email protected])]}{.metadata.name}{'\n'}{end}"
echo "PRTBs that don't match project:"
kubectl get projectroletemplatebindings -A -ojson | jq -r '.items[]|.metadata as $m|select(.projectName|endswith($m.namespace)|not)|.metadata.name'
For more information
If you have any questions or comments about this advisory:
- Reach out to SUSE Rancher Security team for security related inquiries.
- Open an issue in Rancher repository.
- Verify our support matrix and product support lifecycle.
Affected Packages
| Ecosystem | Package | Vulnerable range | Fix |
|---|---|---|---|
| 🐹Go | github.com/rancher/rancher | ≥ 2.5.0&&< 2.5.16 | 2.5.16 |
| 🐹Go | github.com/rancher/rancher | ≥ 2.6.0&&< 2.6.7 | 2.6.7 |
Research use only. For defensive security, authorized penetration testing, and academic research only. Never execute exploit code against systems without explicit written authorization.
Detection & mitigation playbook
Open-source dependencyDetect
Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for github.com/rancher/rancher. O3's reachability analysis confirms whether the vulnerable code path is actually invoked in your application, so you act on real exposure instead of every transitive match.
Fix
Update github.com/rancher/rancher to 2.5.16 or later, then make sure no transitive (indirect) dependency still pins the vulnerable range — O3 confirms GHSA-6x34-89p7-95wg is resolved across your whole dependency graph.
Workarounds
If you can't upgrade right away: gate or disable the affected feature, validate untrusted input at the boundary, and avoid passing attacker-controlled data into the vulnerable path. O3's runtime protection blocks exploitation in production as an interim safeguard until the upgrade lands.
How O3 protects you
O3 pinpoints whether GHSA-6x34-89p7-95wg is reachable in your code and exactly where to fix it, then blocks exploitation in production at runtime until the patched version is deployed.
Tailored to GHSA-6x34-89p7-95wg. Runtime protection reduces exposure until a permanent patch is applied and verified — it complements patching, it doesn't replace it.
Frequently Asked Questions
Is GHSA-6x34-89p7-95wg in your dependencies?
O3 detects GHSA-6x34-89p7-95wg across Go dependencies and uses function-level reachability to confirm whether the vulnerable code path is actually reachable — not just present. No false positives.