Delivery you GCE instance to instance group managed

This script is designed to automate the delivery and deployment of your GCE instance to a regional managed instances group.

The creation of resources in GCP:

• Image of disk ( global )
• Instance Template ( region )
• Managed Instance Group ( region )
• Autoscaler
• Internal TCP Load Balancer ( region )
  • backend service ( region )
  • address ( region )
  • forwarding rule ( region )

How script works:

• Loads parameters from the metadata file and performs the following actions through a GCP API call:
  • creating image from you GCE instance disks
  • creating Instance Template
  • creating Instance Group Manager
    • awaiting group stabilization
  • creating Autoscaler
  • creating Backend Service
  • creating Address in subnetwork
  • creating Forwarding Rule
• Performs accessibility of service through the load balancer

To work you will need:

• Create a service account in GCP with the following roles:

    "roles/compute.instanceAdmin"
    "roles/iam.serviceAccountUser"
    "roles/container.clusterViewer"
    "roles/container.viewer"
    "roles/monitoring.metricWriter"
  

• Create json key of service account
• Create GCE instance and deploying you app
• Define which instance you want to deploy as a managed group
• On what subnet
• With what parameters
• Fill out the metadata file

Script run arguments:

• --metadata
• --gcp-token
• --service
• --version
• --operation
• --help

Metadata file example in metadata.example.yaml

The script can be used in the CI/CD pipeline: example: