Tanzu Data Management Console (TDMC) developer tools provide a way to perform operations programmatically, which can help developers by enabling self-service database provisioning. It includes

Terraform Provider
TDMC CLI
Swagger API

In this blog, I will showcase how a developer/db admin can leverage these automation tools to perform self-service database provisioning

Pre-requistie

Copy the configs for this blog from the following repo

https://github.com/munishpalmakhija/tdmc

TDMC Terraform Provider

Step 1 – Create a Terraform file

Navigate to the terraform folder of the repo and copy the file https://github.com/munishpalmakhija/tdmc/blob/main/terraform/main.tf

Step 2 – Configure TDMC Terraform Provider

host = “TDMC_HOST_URL”
type = “user_creds”
username = “TDMC_USERNAME”
password = “TDMCPASSWORD”

Step 3 – Update Dataplane ID

data_plane_id = “DP_ID” #

Step 4 – Specify Postgres Config

Specify the following config related to the PostgreSQL database

Version
Storage Policy Name
PG Cluster Name
PG Cluster Metadata such as DB Name, Username, and Password (I will be leaving it default)

Note – Terraform will create a new network policy named tf-pg-nw-policy, allowing access from anywhere.

Step 5 – Initialize Terraform

terraform init

Step 6 – Terraform Plan

terraform plan

Step 7 – Terraform Apply

terraform apply –auto-approve

Step 8 – Verify PostgreSQL in TDMC UI

TDMC CLI

Step 1 – Install TDMC CLI

Click on the Downloads Tab to download the required binary by clicking on the respective OS tab. In my case ,I will be downloading amd64 binary for macOS

Move the downloaded binary to your working directory & set the execute permissions

In case you run into the following error on Mac. You will need to allow the application under Privacy & Security in System Settings

Step 2 – Create Profile

To use the tdmc CLI, you need to create a CLI profile by executing the following command

tdmc profile create

Note – Org is for the mm-demo org, which we created in my previous blog

Step 3 – Configue CLI

tdmc configure

Step 4 – Export Profile Password

Store your profile password in your local environment:

export TDMC_PASSWORD='<PROFILE-PASSWORD>’

Step 5 – Verify

You can now verify if you can execute cli commands against the environment.

Below I listing my existing Postgres Cluster, which I deployed in my previous blog

tdmc postgres-cluster list

Step 6 – Create PostgreSQL Cluster using yaml file

Copy the same payload from the repo shared above and update the following values https://github.com/munishpalmakhija/tdmc/blob/main/cli/tdmc_cli_postgre_payload.json

Sample
Provider –
Dataplane ID
NetworkPolicies
Version

tdmc postgres-cluster create -p <provider> -f <yaml_file_path>

Step 7 – Verify PostgreSQL Cluster Creation Status

tdmc task get –id xxx

You can also check the UI for the current status

TDMC API

TDMC has Swagger-based APIs, which can also be leveraged for automation.

Step 1 – Login

curl -k –location ‘https://<tdmc_host>/api/authservice/auth/login’ \
–header ‘accept: text/plain; charset=utf-8’ \
–header ‘Content-Type: application/json’ \
–data-raw ‘{“username”:”<tdmc_org_admin_username>”,”password”:”<tdmc_org_admin_pass>”}’ \
-c cookie.txt

Step 2 – Create Postgres Cluster via TDMC API

Execute the following API to create a PostgreSQL Cluster

Use the Request body from the repo shared above and update the following values https://github.com/munishpalmakhija/tdmc/blob/main/api/tdmc_api_postgres_request_body.json

Sample
Provider –
Dataplane ID
NetworkPolicies
Version

curl -k –location ‘https://<tdmc_host>/api/controller/mdsclusters’ \
–header ‘accept: application/json’ \
-b cookie.txt \
–header ‘Content-Type: application/json’ \
–data ‘{
“name”: “<postgres_cluster_name>”,
“serviceType”: “POSTGRES”,
“tags”: [],
“instanceSize”: “XX-SMALL”,
“version”:”postgres-15″,
“provider”: “tkgs”,
“dataPlaneId”: “<dataplaneid>”,
“region”: “tanzu”,
“dedicated”: false,
“shared”: true,
“networkPolicyIds”:[“<network_policy_id>”],
“clusterMetadata”: {
“username”: “<postgres_db_user>”,
“password”: “<postgres_db_pass>”,
“database”: “<postgres_db_name>”,
“objectStoreId”: null,
“restoreFrom”:null,
“extensions”:[],
“parameters”:{}
},
“storagePolicyName”: “vsan-default-storage-policy”
}’

Once executed, you should see a response with taskId. This is similar to the ID we got using the CLI method.

Conclusion

TDMC allows customers to rapidly deploy customized databases to increase the velocity, and by providing industry-standard automation tools, it can be integrated into the existing pipeline framework that customers are already using it.

Provisioning PostgreSQL Database with Tanzu Data Management Console

Self-Service PostgreSQL Provisioning with TDMC Developer Tools

Pre-requistie