Better Stack GCP Integration

This repository contains setup scripts and Terraform modules for integrating your GCP project or organization with Better Stack.

Repository Structure

Directory	Description
setup.sh	Bash setup script for quick deployment
terraform/	Terraform module for infrastructure-as-code deployment

Getting Started

To get started with Better Stack on GCP, create a new GCP Source in Better Stack and note the ingesting host and source token from the source configuration.

For deployment, choose either the setup script or the Terraform module.

Ingested Data

When you deploy the integration you get:

• Automatic ingestion of all GCP Cloud Logging entries into Better Stack.
• Automatic ingestion of Cloud Monitoring metrics (compute, networking, storage, and more).
• Support for org-level deployment covering all projects, including future ones.
• Remote log filter management — configure which logs are forwarded without redeploying.

Setup Script

The fastest way to deploy. Requires the gcloud CLI.

# Org mode (all projects, including future ones):
./setup.sh \
  --project=my-project \
  --org-id=123456789 \
  --source-token=YOUR_SOURCE_TOKEN \
  --ingesting-host=s4191.g1.betterstackdata.com

# Project mode (single project only):
./setup.sh \
  --project=my-project \
  --source-token=YOUR_SOURCE_TOKEN \
  --ingesting-host=s4191.g1.betterstackdata.com

Optional arguments

Argument	Default	Description
--region	europe-west1	GCP region for the Dataflow job
--batch-count	100	Log entries per batch sent to Better Stack

Teardown

./setup.sh --teardown \
  --project=my-project \
  --ingesting-host=s4191.g1.betterstackdata.com \
  --org-id=123456789

Terraform

See the Terraform module README for full usage and variable reference.

module "betterstack" {
  source         = "github.com/betterstack/gcp-integration//terraform"
  project_id     = "my-project"
  org_id         = "123456789"
  source_token   = var.source_token
  ingesting_host = var.ingesting_host
}

What Gets Created

• Service accounts — betterstack-integration (metrics and log sink management, impersonated by Better Stack via WIF) and betterstack-dataflow (runs the log forwarding pipeline).
• Workload Identity Federation — cross-project authentication using short-lived tokens. No static keys.
• Log sink — captures Cloud Logging entries and routes them to Pub/Sub. Supports org-level or project-level scope.
• Pub/Sub topic + subscription — buffered log delivery between the sink and the forwarding pipeline.
• Dataflow job — batches log entries from Pub/Sub and forwards them to Better Stack.

Prerequisites

The user running the setup needs:

• Project-level: roles/owner or roles/editor on the GCP project
• Org-level (only if deploying with --org-id): roles/resourcemanager.organizationAdmin on the organization

Log Filter Management

After deployment, Better Stack can remotely configure which logs are forwarded by updating the Logging filter from your source's settings page.