GCP Integration

FaultRay integrates with Google Cloud Platform to scan and model your GCP infrastructure.

Setup

Install the GCP extras:

pip install "faultray[gcp]"

Authenticate with Google Cloud:

gcloud auth application-default login
# or set GOOGLE_APPLICATION_CREDENTIALS
export GOOGLE_APPLICATION_CREDENTIALS=/path/to/service-account.json

Scanning

Full project scan

faultray scan --provider gcp --output gcp-infra.json

Specific project

faultray scan --provider gcp --project my-project-id --output gcp.json

Supported Services

Service	Component Type	Details
Compute Engine	Compute	Instances, instance groups, MIGs
Cloud SQL	Database	Instances, replicas, high availability
Cloud Memorystore	Cache	Redis instances, failover replicas
Cloud Load Balancing	Load Balancer	Backend services, health checks
Cloud Storage	Storage	Buckets, dual-region, turbo replication
Cloud DNS	DNS	Managed zones, routing policies
Cloud CDN	CDN	Backend buckets, cache policies
GKE	Container	Clusters, node pools, regional clusters
Cloud Functions	Serverless	Functions, min instances
Pub/Sub	Queue	Topics, subscriptions, dead letter topics

Example

from faultray.scanners import GCPScanner

scanner = GCPScanner(project="my-project-id")
graph = scanner.scan()

print(f"Discovered {len(graph.nodes)} GCP resources")

Required Roles

The service account needs the following IAM roles:

• roles/compute.viewer
• roles/cloudsql.viewer
• roles/redis.viewer
• roles/storage.objectViewer
• roles/dns.reader
• roles/container.viewer