Resource definitions for GCP - Commands & Configuration

Introduction

When you want to create and manage cloud resources on Google Cloud Platform, you define them using configuration files. These files describe what resources you want, like virtual machines or storage buckets, so GCP can set them up for you automatically.

When you want to create a virtual machine instance with specific settings like CPU and memory.

When you need to set up a storage bucket to save files and data.

When you want to define a network with firewall rules to control traffic.

When you want to automate resource creation to avoid manual setup errors.

When you want to keep your infrastructure setup as code for easy updates and sharing.

Config File - compute_instance.yaml

compute_instance.yaml

resources:
- name: example-instance
  type: compute.v1.instance
  properties:
    zone: us-central1-a
    machineType: zones/us-central1-a/machineTypes/n1-standard-1
    disks:
    - deviceName: boot
      type: PERSISTENT
      boot: true
      autoDelete: true
      initializeParams:
        sourceImage: projects/debian-cloud/global/images/family/debian-11
    networkInterfaces:
    - network: global/networks/default
      accessConfigs:
      - name: External NAT
        type: ONE_TO_ONE_NAT

This YAML file defines a virtual machine instance named example-instance.

resources: The list of resources to create.

name: The unique name for the instance.

type: The kind of resource, here a Compute Engine VM.

properties: Settings for the VM like zone, machine type, disk, and network.

This file tells GCP to create a VM in us-central1-a zone with a Debian 11 image and default network access.

Commands

This command tells Google Cloud to create a deployment named 'example-deployment' using the resource definitions in the 'compute_instance.yaml' file. It sets up the VM as described in the file.

Terminal

gcloud deployment-manager deployments create example-deployment --config compute_instance.yaml

Expected Output

Create operation operation-1234567890abcdef started...

Waiting for operation to complete...

Create operation operation-1234567890abcdef completed successfully.

→

--config - Specifies the configuration file with resource definitions.

This command checks if the VM named 'example-instance' was created successfully by listing instances with that name.

Terminal

gcloud compute instances list --filter="name=example-instance"

Expected Output

NAME              ZONE           MACHINE_TYPE    PREEMPTIBLE  INTERNAL_IP  EXTERNAL_IP     STATUS

example-instance   us-central1-a  n1-standard-1                 10.128.0.2   34.68.123.45    RUNNING

→

--filter - Filters the list to show only the instance with the specified name.

This command deletes the deployment named 'example-deployment' and all resources it created, cleaning up your cloud environment.

Terminal

gcloud deployment-manager deployments delete example-deployment

Expected Output

Delete operation operation-abcdef1234567890 started...

Waiting for operation to complete...

Delete operation operation-abcdef1234567890 completed successfully.

Key Concept

If you remember nothing else from this pattern, remember: resource definitions describe what you want in GCP, and deployment commands make it real.

Common Mistakes

Using incorrect indentation or syntax in the YAML file.

YAML is sensitive to spaces and indentation; errors cause deployment failures.

Use a YAML validator or editor to ensure correct formatting before deploying.

Not specifying the correct zone or machine type in the resource properties.

GCP will reject or create resources in unexpected locations or with wrong specs.

Double-check zone and machineType values match available GCP options.

Trying to create a deployment with a name that already exists without deleting it first.

Deployment Manager will fail because deployment names must be unique.

Delete or update existing deployments before creating new ones with the same name.

Summary

Write resource definitions in a YAML file to describe your GCP resources.

Use 'gcloud deployment-manager deployments create' with the config file to create resources.

Verify resource creation with 'gcloud compute instances list' filtered by name.

Clean up resources by deleting the deployment with 'gcloud deployment-manager deployments delete'.

Practice

1. What is the main purpose of a resource definition in Google Cloud Platform (GCP)?

easy

A. To manually start and stop virtual machines

B. To write application code that runs on GCP

C. To monitor the usage of cloud resources

D. To tell GCP what cloud parts to create and how to configure them

Solution

1. Step 1: Understand the role of resource definitions

   Resource definitions specify what cloud resources to create and their settings.

2. Step 2: Compare options to the definition role

   Only To tell GCP what cloud parts to create and how to configure them describes creating and configuring cloud parts, which matches resource definitions.

3. Final Answer:

   To tell GCP what cloud parts to create and how to configure them -> Option D

4. Quick Check:

   Resource definitions = create and configure cloud parts [OK]

Hint: Resource definitions describe cloud parts to create [OK]

Common Mistakes:

• Confusing resource definitions with application code
• Thinking resource definitions monitor usage
• Believing resource definitions manually control VMs

2. Which of the following is the correct minimal structure for a GCP resource definition in YAML?

easy

A. resources:\n - name: my-vm\n type: compute.instance\n properties:\n location: us-central1-a

B. resources:\n - id: my-vm\n kind: compute.v1.instance\n settings:\n zone: us-central1-a

C. resources:\n - name: my-vm\n type: compute.v1.instance\n properties:\n zone: us-central1-a

D. resource_list:\n - name: my-vm\n type: compute.v1.instance\n properties:\n zone: us-central1-a

Solution

1. Step 1: Check the required keys in GCP resource definitions

   GCP expects 'resources' list with each item having 'name', 'type', and 'properties'.

2. Step 2: Validate each option

   resources:\n - name: my-vm\n type: compute.v1.instance\n properties:\n zone: us-central1-a uses correct keys and valid type 'compute.v1.instance'. Others use wrong keys like 'id', 'kind', or wrong root key 'resource_list'. resources:\n - name: my-vm\n type: compute.instance\n properties:\n location: us-central1-a uses incorrect type 'compute.instance'.

3. Final Answer:

   resources:\n - name: my-vm\n type: compute.v1.instance\n properties:\n zone: us-central1-a -> Option C

4. Quick Check:

   Correct keys and type = resources:\n - name: my-vm\n type: compute.v1.instance\n properties:\n zone: us-central1-a [OK]

Hint: Look for 'resources', 'name', 'type', 'properties' keys [OK]

Common Mistakes:

• Using 'id' instead of 'name'
• Using 'kind' instead of 'type'
• Wrong root key like 'resource_list'
• Incorrect resource type format

3. Given this resource definition snippet, what will be the zone of the created VM?

resources:
  - name: test-vm
    type: compute.v1.instance
    properties:
      zone: europe-west1-b
      machineType: n1-standard-1
      disks:
        - deviceName: boot
          boot: true
          autoDelete: true
          initializeParams:
            sourceImage: projects/debian-cloud/global/images/family/debian-11

medium

A. europe-west1-b

B. us-central1-a

C. asia-east1-a

D. No zone specified, deployment will fail

Solution

1. Step 1: Locate the zone property in the resource definition

   The 'zone' property is set to 'europe-west1-b' under 'properties'.

2. Step 2: Understand zone usage in VM creation

   The VM will be created in the specified zone 'europe-west1-b'.

3. Final Answer:

   europe-west1-b -> Option A

4. Quick Check:

   Zone property value = europe-west1-b [OK]

Hint: Check 'zone' under 'properties' for VM location [OK]

Common Mistakes:

• Assuming default zone if specified
• Confusing machineType with zone
• Thinking deployment fails without explicit zone

4. Identify the error in this resource definition snippet:

resources:
  - name: my-bucket
    type: storage.v1.bucket
    properties:
      location: us-east1
      storageClass: STANDARD
  - name: my-bucket
    type: storage.v1.bucket
    properties:
      location: us-west1
      storageClass: NEARLINE

medium

A. Invalid storageClass values used

B. Duplicate resource names cause deployment failure

C. Missing 'type' field in second resource

D. Location values are not allowed in bucket properties

Solution

1. Step 1: Check resource names for uniqueness

   Both resources have the same name 'my-bucket', which is not allowed.

2. Step 2: Verify other fields

   Both have 'type' and valid 'storageClass' and 'location'. Only duplicate names cause error.

3. Final Answer:

   Duplicate resource names cause deployment failure -> Option B

4. Quick Check:

   Unique resource names required = Duplicate resource names cause deployment failure [OK]

Hint: Resource names must be unique in definitions [OK]

Common Mistakes:

• Ignoring duplicate names
• Thinking storageClass values are invalid
• Believing location is disallowed

5. You want to define a GCP resource that creates a Compute Engine VM with a specific network interface and a startup script. Which of the following resource definitions correctly includes both properties?

hard

A. resources:
- name: vm-with-script
type: compute.v1.instance
properties:
zone: us-central1-a
machineType: n1-standard-1
networkInterfaces:
- network: global/networks/default
metadata:
items:
- key: startup-script
value: |-
#!/bin/bash
echo Hello World > /var/log/startup.log

B. resources:
- name: vm-with-script
type: compute.v1.instance
properties:
zone: us-central1-a
machineType: n1-standard-1
network: default
startupScript: |
#!/bin/bash
echo Hello World > /var/log/startup.log

C. resources:
- name: vm-with-script
type: compute.v1.instance
properties:
zone: us-central1-a
machineType: n1-standard-1
networkInterfaces:
- network: default
metadata:
startup-script: |
#!/bin/bash
echo Hello World > /var/log/startup.log

D. resources:
- name: vm-with-script
type: compute.v1.instance
properties:
zone: us-central1-a
machineType: n1-standard-1
networkInterfaces:
- network: default
metadata:
items:
- key: startup
value: echo Hello World

Solution

1. Step 1: Check network interface property format

   Correct format uses 'networkInterfaces' list with 'network' as full resource path or relative path like 'global/networks/default'.

2. Step 2: Check metadata startup script format

   Metadata must have 'items' list with 'key' and 'value' pairs. The key for startup script is 'startup-script'.

3. Step 3: Validate each option

   resources: - name: vm-with-script type: compute.v1.instance properties: zone: us-central1-a machineType: n1-standard-1 networkInterfaces: - network: global/networks/default metadata: items: - key: startup-script value: |- #!/bin/bash echo Hello World > /var/log/startup.log correctly uses 'networkInterfaces' with full network path and metadata with 'items' list and proper startup script. Others have incorrect keys or missing 'items' list.

4. Final Answer:

   Option A correctly defines network interface and startup script -> Option A

5. Quick Check:

   Correct networkInterfaces and metadata items = resources: - name: vm-with-script type: compute.v1.instance properties: zone: us-central1-a machineType: n1-standard-1 networkInterfaces: - network: global/networks/default metadata: items: - key: startup-script value: |- #!/bin/bash echo Hello World > /var/log/startup.log [OK]

Hint: Use 'networkInterfaces' list and metadata 'items' with key/value [OK]

Common Mistakes:

• Using 'network' instead of 'networkInterfaces'
• Missing 'items' list in metadata
• Wrong key name for startup script
• Incorrect network path format