A critical vulnerability in the Apache Airflow Kafka provider enables authenticated users to gain arbitrary code execution within the control plane, posing a significant risk to system integrity.

This flaw involves unsafe reflection (CWE-470), where the provider resolves dotted-path strings in the Kafka connection extra field into Python callables without an allowlist. An authenticated user with permissions to modify connections can trigger arbitrary code execution on the scheduler if specific event producer features are enabled.

The ability for an authenticated user to execute arbitrary code in the control plane constitutes a complete compromise of the Airflow scheduler. With a CVSS score of 8.8, this vulnerability represents a high-severity risk that could lead to unauthorized data access, lateral movement, or full system takeover. Organizations relying on Airflow for automated workflows face significant operational and security exposure until the fix is applied.

Immediate Action: Upgrade to apache-airflow-providers-apache-kafka version 2.0.0 or later, which introduces a mandatory allowlist for connection-string callbacks.

Proactive Monitoring: Review Airflow connection configurations for suspicious or unauthorized entries in the extra field and monitor scheduler process logs for unexpected Python execution patterns.

Compensating Controls: Ensure that the features dag_run_events_enabled and task_instance_events_enabled remain disabled if they are not required, as these serve as the primary attack surface for this vulnerability.

Public Exploit Available: Unknown

Given the potential for arbitrary code execution in the control plane, this update should be prioritized immediately. Administrators must verify their current provider version and apply the 2.0.0 patch across all affected environments to mitigate the risk of unauthorized command execution by authenticated users.