CVE-2026-44935: Multi-tenant Isolation Flaw in SUSE Rancher Fleet

Affected Versions and Fixed Releases

The affected version ranges are specifically documented in the CVE record. Vulnerable releases are:

If you run SUSE Rancher Fleet in any of those ranges, you should treat the instance as exposed to cross-tenant credential access until upgraded. This is especially urgent for hosted internal platforms, shared clusters, managed service environments, or any deployment where different teams or customers have separate tenant ownership.

The fixed versions are also clearly stated: 0.12.15, 0.13.11, 0.14.6, and 0.15.2. If your Fleet deployment is on one of those releases or newer within the same maintained line, the specific issue described by CVE-2026-44935 should be addressed. If you do not know your Fleet version, validate that immediately before making assumptions about exposure.

Technical Notes

You can inventory Fleet components in Kubernetes with standard commands. The exact namespace and release naming may vary by deployment, so adjust selectors to match your environment.

kubectl get deployments -A | grep -i fleet
kubectl get pods -A | grep -i fleet
helm list -A | grep -i fleet

To inspect image tags or chart versions in-cluster:

kubectl get deployment -A -o json | jq -r '
  .items[]
  | select(.metadata.name|test("fleet"; "i"))
  | [.metadata.namespace, .metadata.name, (.spec.template.spec.containers[]?.image // "")]
  | @tsv'

If your deployment is managed through Helm, confirm the chart/app version before and after maintenance:

helm status <release-name> -n <namespace>
helm history <release-name> -n <namespace>

Why This Matters Operationally

A CVSS 9.9 score puts this in the highest urgency bucket, but the operational impact matters more than the raw score. This flaw breaks tenant isolation in a product explicitly used to manage deployment state across Kubernetes environments. Once a tenant boundary fails in a control-plane-adjacent component, the blast radius can extend beyond simple read access.

The explicit impact in the advisory language is access to Fleet credentials of other tenants. Depending on how Fleet credentials are scoped in your environment, that may expose repository access, downstream cluster interaction, or deployment-related secrets. Even where the credential scope is narrow, any cross-tenant secret access should be treated as a serious incident class because it undermines platform trust and segregation guarantees.

Technical Notes

The exact CVSS vector string was not available in the retrieved NVD data. That means defenders should avoid guessing whether the issue is considered network, adjacent, or otherwise by the scoring authority. In practice, you should assume exposure exists anywhere a tenant owner can interact with Fleet-managed Helm deployment workflows that use valuesFrom references.

A useful scoping question for defenders is: do tenant owners have the ability to define or influence Helm values sources through GitOps repos, Fleet bundles, or namespace-scoped objects? If yes, prioritize those environments first for upgrade and review.

Exploitation Status and Threat Intelligence

As of 2026-07-02, there is no verified evidence from the primary sources used here that exploitation is confirmed in the wild. The CVE is not listed in CISA’s Known Exploited Vulnerabilities catalog based on the current check noted in the research material. That means there is no CISA-confirmed active exploitation status at this time.

There is also no verified public proof-of-concept established from the referenced primary materials. The NVD record and GitHub advisory were available, but no confirmed exploit repository or PoC code was identified in the supplied research. That does not reduce the importance of patching. For multi-tenant flaws, exploitability often depends more on product knowledge and access to a valid tenant account than on mass-exploitation tooling.

Technical Notes

In the absence of a public PoC, defenders should assume capable internal users or attackers with compromised tenant-owner access may be able to reproduce the issue once patch details are studied. That is a common pattern for authorization and validation flaws in management platforms.

Because exploitation is not confirmed, response priority should be based on architecture rather than headlines. Single-tenant or lab environments may have lower practical exposure, while shared production platforms should treat this as a fast-turn remediation item.

What Defenders Should Do Next

First, determine whether you run any affected Fleet version: 0.12.0 through 0.12.14, 0.13.0 through 0.13.10, 0.14.0 through 0.14.5, or 0.15.0 through 0.15.1. If yes, prioritize upgrades in environments with multiple tenants, delegated application teams, or any user population that can influence Helm value sources.

Second, conduct a focused review of recent tenant-driven deployment changes. Look for unusual valuesFrom references, secret access patterns spanning namespaces, or credentials that may have been exposed to unauthorized tenants. Because no confirmed exploitation data is currently available, defenders should not assume safety; they should assume limited visibility and act accordingly.

For further information on related vulnerabilities, check out our articles on CVE-2026-11450 and CVE-2026-31431.