Phase 1: GCP onboarding wizard

1.1 Start the GCP onboarding wizard

1.1.1 Access the GCP onboarding wizard in Cortex Cloud:

1. In Cortex Cloud, select Settings → Data Sources & Integrations.
2. On the Data Sources & Integrations page, click + Add New.
3. On the Add Data Sources or Integrations page, search for Google Cloud Platform (GCP), then hover over it and click Add.

1.1.2 Select the GCP environment

In the GCP onboarding wizard, select the type of GCP environment:

• Government: GCP GovCloud environments for compatibility with FedRAMP-certified tenants.
• Commercial: (Default) Standard cloud deployment typically used for private and public sector organizations that do not require isolated government-specific infrastructure.

1.1.3 Select the scope

Select the scope for this cloud instance:

• Organization: (Selected for this guide) A collection of GCP projects that are managed centrally.
• Folder: A GCP folder can contain projects, folders, or a combination of both projects and folders.
• Project: A specific GCP project.

1.1.4 Choose the scan mode

Specify the scanning infrastructure for your cloud instance by selecting one of the following scan modes:

• Cloud Scan: (Recommended) Security scanning is performed in the Cortex Cloud cloud environment.
• Scan with Outpost: Security scanning is performed on infrastructure deployed to a cloud account owned by you. If you select this option, choose the outpost account to use for this instance. Scanning with an outpost may require additional GCP permissions and may incur additional CSP costs.

1.1.5 Configure advanced settings

Click Show advanced settings to define the following advanced settings:

• Instance Name: Enter a unique instance name or leave it empty to be automatically populated. The automatic naming convention is GCP-<projectID>. Cortex does not prevent you from reusing instance names, but it is best practice to use a unique name for every cloud instance.
• Deployment Method: Select whether you want to onboard with a Cortex-generated Terraform authentication template or to perform a manual deployment.

  • Terraform: (Recommended). Proceed to How to onboard Google Cloud Platform.
  • Manual: Select this option to manually configure service accounts and permissions, so your cloud setup stays aligned with internal governance requirements.
• Scope Modifications: Use these settings to fine-tune your GCP scope. You can modify the scope by including or excluding specific regions. Additionally, if you selected an organization or folder as the scope, you can modify the scope by including or excluding specific projects. For more details, see Apply region or account filters.
• Additional Security Capabilities: Choose which security capabilities you want to benefit from. Some security capabilities are enabled by default and can be modified. Adding security capability typically requires additional cloud provider permissions. For detailed information on the permissions required, see Cloud service provider permissions.

  • Data security posture management: An agentless data security scanner that discovers, classifies, protects, and governs sensitive data.
  • Registry scanning: A container registry scanner that scans registry images for vulnerabilities, malware, and secrets. For more details, see Configure registry scanning for cloud accounts.
  • Serverless functions scanning: Implement serverless scanning to detect and remediate vulnerabilities within serverless functions during the development lifecycle. Seamless integration into CI/CD pipelines enables automated security scans for a continuously secure pre-production environment.
  • Automation: Use automation to pre-configure a list of integrations and associated commands to automate security issue responses. Commands can be utilized individually or as part of custom playbooks for issue remediation.

    • Log Level: (Optional - for Automation only) Configure the automation integration logging level. Possible values are:

      • Off (Default)
      • Debug
      • Verbose
  • Agentless disk scanning: (Recommended) Implement agentless disk scanning to remotely detect and remediate vulnerabilities during the development lifecycle.
• Cloud Tags: Define tags and tag values to be added to any new resource created by Cortex Cloud in GCP. Note: The managed_by paloaltonetworks tag is automatically added to all resources. This tag is mandatory. You cannot edit or remove this tag.
• Log Collection Configuration: To maximize security coverage, include the collection of audit logs (GCP Pub/Sub). This may require additional cloud service provider permissions. For detailed information on the permissions required, see Cloud service provider permissions.
• Connect to GCP Workspace: Gain a comprehensive view of your Google Workspace identities and security. This provides you with detailed information on your users, groups, and organizational units, and collects security event logs to help you detect threats, improve your security posture, and meet compliance requirements. Note: If you want to connect to your GCP Workspace, you must first complete onboarding with the option disabled. Once the GCP cloud instance is created, perform the steps detailed in Connect Google Workspace with your GCP cloud instance.
• Upload unknown files to WildFire: Use this option to upload unknown files scanned during registry image scans to WildFire for detonation analysis. This option expands malware detection by allowing WildFire to analyze new samples found in your registry images. When a detonation result returns a malicious verdict, the system re-evaluates the relevant registry image and creates a malware finding.

Upload unknown files to WildFire

Send samples to detonation in WildFire. Supported in: Registry Scanning.

This feature requires data egress, which incurs standard data transfer fees.

Notes

• The file types sent for WildFire analysis depend on the platform type. WildFire accepts files up to 300 MB in size.
• This setting applies only to registry image scans and is enabled by default for new GCP instances. For existing instances, this setting is disabled by default to preserve the current behavior. You can enable it at any time by editing the instance configuration.
• Your cloud provider may charge standard outbound data transfer (egress) fees when scanning with an Outpost.

1.2 Download the customized identifiers file

The Identifiers JSON file is a customized configuration document generated based on the specific security capabilities selected during the initial onboarding wizard. It acts as a workbook to track service accounts and organization-specific metadata across the different phases of deployment.

The file is divided into two primary sections, each serving a distinct purpose in your manual onboarding process:

• The "identifiers" section: This section is pre-populated by the GCP onboarding wizard. It contains the specific Google Service Account (GSA) emails generated for your environment. You will reference these emails in Phase 2 to grant the necessary IAM roles within your GCP Console.
• The "manual_details" section: These fields are initially empty. During the manual setup in Phase 2, you will retrieve specific values from your GCP environment, such as your Organization ID and Audit Pub/Sub Subscription ID, and input them directly into this section of the file.

In the GCP onboarding wizard, in the Manual Connection Setup step, click Identifiers to download your customized identifiers file.