Fint PGerator

FINT pgerator is an operator that creates a PostgreSQL database, user and PGBouncer connection pool when a PGDatabaseAndUser CR is created. Username and password will be generated and stored in secrets.

What does the operator do?

When a PGDatabaseAndUser CR is created:

• The operator will create a PostgreSQL database, connection pool and user.
• Username, password and JDBC uri will be generated and stored in secrets.

When a PGDatabaseAndUser CR is deleted:

• The operator will delete the PostgreSQL user and connection pool.
• The operator will not delete the PostgreSQL database.
• The operator will delete the secrets containing username and password.

How to use the operator:

PGDatabaseAndUser

apiVersion: "fintlabs.no/v1alpha1"
kind: PGDatabaseAndUser
metadata:
    name: <name>

Example of Custom Resource

apiVersion: "fintlabs.no/v1alpha1"
kind: PGDatabaseAndUser
metadata:
    name: test-user

Installation

Below is an introduction on how to set up the project. First for local development in IntelliJ, then to the cluster.

Prerequisites

You will need an authentication token from aiven.

• Log in to https://console.azure.aiven.io/
• Go to User Information (Human icon top right of the site)
• Choose "authentication"
• Choose "Generate token"
• Give your token a name, e.g. Development
• Copy and add the token to 1password so you don't loose track of it.

Local development

1. Clone the repository: git clone https://github.com/FINTLabs/pgerator.git
2. Create Run/Debug configuration "Run -> Edit Configurations..."
3. Add the following three prerequisites (Name - Value):
   • fint.aiven.service - pg-alpha
   • fint.aiven.token - generated in the steps mentioned in the prerequisites
   • fint.data.pool-base-url - jdbc:postgresql://<aiven PgBouncer Host> : <aiven PgBouncer Port>/
4. Cd to the project root and build the Gradle project: ./gradlew clean build
5. Apply the custom-resource: kubectl apply -f build/classes/java/main/META-INF/fabric8/pgdatabaseandusers.fintlabs.no-v1.yml
6. Run or debug the Application in IntelliJ.
7. Check that it worked: kubectl get secrets and kubectl get customresourcedefinitions, you should see the name you specified in the custom .yaml file in the list.
8. Also check that you find the instances created in aiven.
   • Services -> pg-alpha -> Users and Services -> pg-alpha -> Databases