Creating a Deployment (Legacy Reference)
What you'll learn: How legacy Google Cloud Deployment Manager configurations provisioned resources like a Compute Engine instance, and what tools to use for infrastructure today.
Important Notice: Google Cloud Deployment Manager reached end of support on March 31, 2026. This lesson is provided strictly as a legacy reference for understanding and maintaining existing environments. For all new infrastructure provisioning, you should direct your work to Google Cloud Infrastructure Manager or Terraform.
The Legacy Deployment Process
In Deployment Manager, creating a deployment involved transforming a YAML blueprint into actual cloud resources. While you should not use this for new projects, understanding this workflow is vital if you are maintaining or migrating older systems. The historical process involved three key steps:
1. Preparing the Configuration
Engineers required a YAML file that defined at least one resource. For a basic deployment, a simple Compute Engine VM configuration was often used to declare the desired state (e.g., machine type, zone, and boot disk).
2. Executing the Create Command
To submit the configuration, engineers used the Deployment Manager CLI. You may still encounter the following command in legacy automation scripts or CI/CD pipelines:
# Legacy reference only - do not use for new deployments
gcloud deployment-manager deployments create my-first-deployment \
--config vm-config.yaml
This command told Google Cloud to read the vm-config.yaml file, validate its syntax, and provision the resources listed inside.
3. Monitoring Creation
Deployment Manager provided real-time progress as it provisioned resources. If a deployment is still active in an old environment, status messages for each resource are logged, and errors halt the process with feedback.
What Happened Behind the Scenes
When the legacy create command was executed, Deployment Manager parsed the YAML, validated the resource properties, checked IAM permissions, and subsequently called the appropriate Google Cloud APIs to provision each resource. It determined the correct order of creation based on the dependencies defined in the template. Modern tools like Infrastructure Manager and Terraform handle API orchestration using a very similar declarative approach.
Key Takeaway: Google Cloud Deployment Manager reached end of support on March 31, 2026, meaning all new deployments must use Infrastructure Manager or Terraform. However, understanding the legacy gcloud deployment-manager deployments create workflow remains necessary for maintaining historical infrastructure as code.