Improper input validation enables remote code execution in Google Cloud BigQuery Data Transfer Service

Note: This data is for reference and cybersecurity research purposes only.CyStack advises users not to use this information for unlawful purposes.

What is CVE-2026-12717?

CVE-2026-12717 is a vulnerability classified as Improper Neutralization of Special Elements in Output Used by a Downstream Component, affecting BigQuery Data Transfer Service (affected versions: < 2026-05-01). This vulnerability is rated Critical, with a CVSS score of 9.4. Current sources do not report this vulnerability as exploited.

Overview

Original source data

An Improper Input Validation vulnerability in CData JDBC driver integration in Google Cloud BigQuery Data Transfer Service versions prior to 2026-05-01 on Google Cloud Platform allows an authenticated attacker to achieve remote code execution in the connector container and escalate privileges in the tenant project using crafted JDBC connection string parameters. This vulnerability was patched on 1 May 2026, and no customer action is needed.

Affected products and scope

  • Google Cloud BigQuery Data Transfer Service: the default status is unaffected, with the CNA defining the affected range as version date 0 through before 2026-05-01, using versionType date.
  • Google describes the issue in the JDBC driver integration and states that it was patched on 2026-05-01.
  • The record does not provide a separate release identifier or parallel branches that would support a broader open-ended version range; the status of branches not directly listed should not be inferred.

Technical details

The flaw is in the CData JDBC driver integration in BigQuery Data Transfer Service, where attacker-controlled JDBC connection string parameters reach processing inside the connector container. Insufficient input validation allows crafted values to reach a downstream component in a way that can enable code injection and remote code execution. The recorded preconditions include network reachability and an authenticated account; the structured record also describes low initial privileges, low complexity, and no user interaction. The specific parser, execution sink, runtime context, and exact privileges obtained after exploitation are not disclosed.

Exploitability

The vulnerability is remotely reachable over the network by an authenticated attacker with low initial privileges, low attack complexity, and no need for another user's interaction. The supplied record marks public exploit as false, while CISA enrichment records exploitation as none at the time of its assessment. That status does not establish that undisclosed or future exploitation is impossible. No evidence identifies a specific campaign, victim, or breach.

Technical impact

Successful exploitation could result in remote code execution in the connector container and privilege escalation in the tenant project. This could affect the confidentiality, integrity, and availability of project data or operations, but the record does not identify the specific assets or exact privileges obtained. The boundary between the connector container and other tenant resources is not described, so container escape or compromise of the entire project should not be asserted beyond the documented privilege escalation.

Business impact

  • Successful exploitation could provide code execution in the connector container and expanded privileges in the tenant project.
  • Possible operational consequences include access to or alteration of data handled by transfer jobs, changes to service configuration, or disruption of project-dependent operations; these are possible outcomes, not a confirmed breach scenario.
  • Risk is concentrated in tenants with CData JDBC configurations and accounts allowed to create or modify transfer configurations.
  • Google states that the service has been patched and that customers do not need to take action; no specific affected data asset is identified publicly.

Remediation

  1. For the managed Google Cloud service: no customer action is required according to Google because the patch was applied on 2026-05-01.
  2. Verify through internal inventory that projects which used the CData JDBC driver integration are not relying on service behavior from before the patch. This is an internal validation step, not a Google-mandated upgrade.
  3. Review permissions for accounts that can create or modify transfer configurations and reduce them according to least privilege. This is an additional precaution; the record does not specify a separate workaround.
  4. If suspicious configuration or activity is found, investigate tenant-project audit logs and related privilege changes; the record does not provide a recovery procedure or a credential-rotation list.

Detection

  • Inventory projects using BigQuery Data Transfer Service and identify configurations that use the CData JDBC driver integration.
  • Review transfer-job configuration and change history for JDBC connection string parameters created or modified while the service was within the affected scope.
  • Review tenant-project audit logs for unusual transfer-configuration changes, connector execution, or privilege changes following suspicious JDBC activity.
  • The absence of unusual logs or events does not prove safety; these monitoring steps are precautions, not Google-confirmed indicators.
Sources (18)
Learn more

Run an in-depth assessment with complete web risk management

CyStack VulnScan continuously discovers assets, validates vulnerabilities, and helps security teams prioritize remediation across the organization.