Creating a Cloud SQL instance in GCP - Visual Walkthrough

Process Flow - Creating a Cloud SQL instance

Start: Define instance config

↓

Call API to create instance

↓

Cloud SQL service processes request

↓

Instance provisioning begins

↓

Instance status: PENDING

↓

Instance status: CREATING

↓

Instance status: RUNNABLE

↓

Instance ready for connections

This flow shows how defining a Cloud SQL instance configuration leads to API call, provisioning, and finally a ready-to-use database instance.

Execution Sample

GCP

gcloud sql instances create my-instance \
  --database-version=POSTGRES_15 \
  --tier=db-f1-micro \
  --region=us-central1

This command creates a small PostgreSQL Cloud SQL instance in the us-central1 region.

Process Table

Step	Action	API Request	Instance Status	Result
1	Define instance parameters	N/A	N/A	Parameters set: name=my-instance, version=POSTGRES_15, tier=db-f1-micro, region=us-central1
2	Send create instance command	CreateInstance API called	PENDING	Request accepted, provisioning starts
3	Cloud SQL service provisions resources	N/A	CREATING	Compute and storage allocated
4	Initialize database engine	N/A	CREATING	Database software installed and configured
5	Instance becomes runnable	N/A	RUNNABLE	Instance ready to accept connections
6	Connect to instance	N/A	RUNNABLE	Connection successful, instance operational

💡 Instance status RUNNABLE means creation finished and instance is ready

Status Tracker

Variable	Start	After Step 2	After Step 3	After Step 5	Final
instance_status	N/A	PENDING	CREATING	RUNNABLE	RUNNABLE
api_request	None	CreateInstance API called	N/A	N/A	N/A
connection_ready	False	False	False	True	True

Key Moments - 2 Insights

Why does the instance status change from PENDING to CREATING before it becomes RUNNABLE?

The execution_table shows that after the API call (PENDING), Cloud SQL allocates resources and installs software (CREATING) before the instance is ready (RUNNABLE). This is normal provisioning progress.

Can you connect to the instance immediately after sending the create command?

No. According to the execution_table, connection_ready is False until the instance status is RUNNABLE (Step 5). Trying earlier will fail.

Visual Quiz - 3 Questions

Look at the execution_table, what is the instance_status at Step 3?

A CREATING

B PENDING

C RUNNABLE

D TERMINATED

💡 Hint

Check the 'Instance Status' column at Step 3 in the execution_table

At which step does connection_ready become True according to variable_tracker?

A After Step 2

B After Step 3

C After Step 5

D After Step 1

💡 Hint

Look at the 'connection_ready' row in variable_tracker and see when it changes to True

If the API request failed at Step 2, what would be the instance_status in the execution_table?

A RUNNABLE

B No status or error

C CREATING

D PENDING

💡 Hint

Consider what happens if the CreateInstance API call does not succeed at Step 2

Concept Snapshot

Creating a Cloud SQL instance:
- Use 'gcloud sql instances create [NAME]'
- Specify database version, tier, and region
- Instance status moves: PENDING -> CREATING -> RUNNABLE
- Only connect when status is RUNNABLE
- Provisioning includes resource allocation and software setup

Full Transcript

Creating a Cloud SQL instance starts by defining the instance parameters like name, database version, machine tier, and region. Then, a create instance API call is made. The Cloud SQL service processes this request and begins provisioning resources. The instance status changes from PENDING to CREATING as compute and storage are allocated and the database engine is installed. When provisioning finishes, the status becomes RUNNABLE, meaning the instance is ready to accept connections. Only after this point can you connect to the database instance successfully.