CISA confirms exploitation of Puppet Enterprise flaw

CISA confirms exploitation of Puppet Enterprise flaw

Add CyberWorldOpsto your preferred sources on Google
CISA confirms exploitation of Puppet Enterprise flaw
Vulnerabilities

Illustrative image generated with AI

CVE-2026-115731, a vulnerability that lets attackers run code on Perforce Puppet Server, is being actively exploited, the U.S. Cybersecurity and Infrastructure Security Agency said in its November 17 alert. CISA added the flaw to its Known Exploited Vulnerabilities catalog, raising the urgency of deploying available updates across affected Puppet infrastructure.

An attacker ordinarily needs an agent certificate signed by the Puppet Certificate Authority (CA). However, installations that automatically approve certificate requests without strong verification may let an unauthenticated attacker obtain that certificate over the network and proceed with exploitation.

The exposure also matters to Red Hat Satellite administrators. Red Hat’s advisory says the affected service in Satellite environments accepts client reports over the network by default, allowing systems with authorized agent certificates to reach it.

How report processing becomes code execution

The attack turns a client report into a means of executing commands on the server. A certificate-holding client sends an HTTP PUT request containing a report with a malicious rich-data type. Puppet’s report-processing code deserializes that untrusted content, enabling execution under the Puppet Server service account.

That account’s reach makes the consequences broader than a single report-processing host. Red Hat warns that exploitation can compromise the central CA and all downstream managed systems. Its explanation identifies access to CA keys as a route to further compromise of managed infrastructure.

Red Hat classifies the impact as Critical and assigns a CVSS v3.1 score of 9.9; its page lists the same score for NVD. Security Affairs reports a score of 9.4. These are severity ratings; CISA’s separate finding of active exploitation establishes that attackers are using the flaw.

Security Affairs’ November 18 report relays a different account from Perforce’s latest available information: the vendor said it was unaware of attacks exploiting the vulnerability and had found no evidence of customer data access. The outlet also reported that no particular actor or campaign had been linked to the flaw. Red Hat credits Alain of Neodyme as the original reporter.

Federal obligations and available updates

CISA’s alert invokes Binding Operational Directive 26-04. It requires Federal Civilian Executive Branch agencies to prioritize rapid remediation of KEV-listed flaws on publicly exposed assets where exploitation grants total control, while deferring lower-risk vulnerabilities. The directive also sets expectations for determining whether a system was compromised before patching.

Security Affairs reports a December 1, 2026 deadline for covered federal agencies to remediate this vulnerability. The federal requirements apply to those agencies; CISA encourages other organizations to prioritize KEV entries through risk-based vulnerability management.

Security Affairs says Perforce issued updates on November 3, 2026, identifying these fixed releases:

  • Puppet Enterprise: 2026.0.1, 2025.12.1 and 2023.8.13.
  • Puppet Core: 9.1.3 and 8.22.2.

A supplementary mitigation section dated November 4 on Red Hat’s page also names OpenVox Server 9.2.0 and 8.18.0 and Puppet Core agent 9.1.3 and 8.22.2. Red Hat labels that section as AI-generated guidance provided pending an official patch and asks customers to evaluate and test it independently.

The supplement describes limiting report-event deserialization in FromDataConverter to approved types, blocking untrusted __ptype instantiation and dynamic class_eval execution. It identifies Puppet Server and OpenVox Server instances that process agent reports as exposed to the issue.

Its temporary measures include restricting network access to PUT /puppet/v3/report and validating report schemas and event structures against strict allowlists before deserialization. For detection, it proposes EDR monitoring of unusual child processes and post-exploitation activity under the puppetserver service user, auditd logging of unauthorized system calls and command execution, and Falco or eBPF monitoring for unexpected shells or sensitive credential reads. These controls do not repair the underlying defect.

Administrators should promptly deploy the relevant fixed release. Until patching is complete, follow Red Hat’s main mitigation: require strong verification before automatically signing certificate requests, preventing untrusted clients from acquiring the agent certificate needed to exploit the server.

Read next

Sources

Back to home

Latest Cybersecurity News

All cybersecurity news →