CVE-2023-32196

Rancher's External RoleTemplates can lead to privilege escalation

#!/bin/bash
set -euo pipefail

# get all RoleTemplates with .context == "project" or .context == "" that don't have externalRules.
rts=$(kubectl get roletemplates -o json | jq -r  '.items[] | select((.context == "project" or .context == "") and .external == true and .externalRules == null and .builtin == false) | .metadata.name')
found_invalid_rt=false

for rt in $rts; do
  if ! kubectl get clusterrole "$rt" > /dev/null 2>&1; then
     echo "$rt" # prints RoleTemplate names that don't have a backing ClusterRole
     found_invalid_rt=true
  fi
done

if ! $found_invalid_rt ; then
    echo 'This cluster is not affected by CVE-2023-32197: no RoleTemplate objects found'
fi

A vulnerability has been identified whereby privilege escalation checks are not properly enforced for RoleTemplateobjects when external=true, which in specific scenarios can lead to privilege escalation.

The bug in the webhook rule resolver ignores rules from a ClusterRole for external RoleTemplates when its context is set to either project or is left empty. The fix introduces a new field to the RoleTemplate CRD named ExternalRules. The new field will be used to resolve rules directly from the RoleTemplate. Additionally, rules from the backing ClusterRole will be used if ExternalRules is not provided. The new field will always take precedence when it is set, and serve as the source of truth for rules used when creating Rancher resources on the local cluster.

Please note that this is a breaking change for external RoleTemplates, when context is set to project or empty and the backing ClusterRole does not exist, as this was not previously required.

Important: The fix is automatically applied when upgrading to the release lines 2.8and above. For users still on the 2.7 release line, after the upgrade to a patched version, users are required to opt-in to the fix by enabling the external-rules feature flag.

Please consult the associated MITRE ATT&CK - Technique - Exploitation for Privilege Escalation for further information about this category of attack.

The application assigns, modifies, tracks, or checks privileges incorrectly, allowing a user to gain elevated access. Typical impact: privilege escalation beyond the intended level.

CVE-2023-32196 has a CVSS score of 6.6 (High). The vector is network-reachable, high privileges required, and no user interaction. A CVSS score reflects the worst-case severity of the vulnerability, not your specific exposure. Whether this affects your application depends on whether the vulnerable code is present and reachable in your environment. A fixed version is available (2.7.14, 2.8.5); upgrading removes the vulnerable code path.

Kodem intelligence

Already deployed Kodem?

Patched versions include releases 2.7.14 and 2.8.5.

1. What is CVE-2023-32196? CVE-2023-32196 is a high-severity improper privilege management vulnerability in github.com/rancher/rancher (go), affecting versions >= 2.7.0, < 2.7.14. It is fixed in 2.7.14, 2.8.5. The application assigns, modifies, tracks, or checks privileges incorrectly, allowing a user to gain elevated access.
2. How severe is CVE-2023-32196? CVE-2023-32196 has a CVSS score of 6.6 (High). This score reflects the worst-case severity of the vulnerability, not your specific exposure. Whether it represents real risk in your environment depends on whether the vulnerable code is present and reachable.
3. Which versions of github.com/rancher/rancher are affected by CVE-2023-32196? github.com/rancher/rancher (go) versions >= 2.7.0, < 2.7.14 is affected.
4. Is there a fix for CVE-2023-32196? Yes. CVE-2023-32196 is fixed in 2.7.14, 2.8.5. Upgrade to this version or later.
5. Is CVE-2023-32196 exploitable, and should I be worried? Whether CVE-2023-32196 is exploitable in your environment depends on whether the vulnerable code is present and reachable. A CVSS score is a worst-case rating; it does not account for your specific deployment, configuration, or usage patterns. Kodem, an Intelligent Application Security platform, uses runtime intelligence to show which vulnerabilities actually execute in production, so you can focus on the ones that represent real risk. Get a demo
6. What actually determines whether CVE-2023-32196 is exploitable, and how bad it is? Exploitability and impact are not fixed properties of a CVE. They depend on runtime truth: whether the vulnerable code is present, reachable, and actually executes in your application. A high CVSS score on a dependency that never runs is not the same as real risk. Kodem, an Intelligent Application Security platform, uses runtime intelligence to reveal which vulnerabilities actually execute in production, so teams prioritize the ones that genuinely matter.
7. How do I fix CVE-2023-32196?
   • Upgrade github.com/rancher/rancher to 2.7.14 or later
   • Upgrade github.com/rancher/rancher to 2.8.5 or later

Other vulnerabilities in github.com/rancher/rancher