Skip to main content

Your submission was sent successfully! Close

Thank you for signing up for our newsletter!
In these regular emails you will find the latest updates from Canonical and upcoming events where you can meet our team.Close

Thank you for contacting us. A member of our team will be in touch shortly. Close

How to deploy using Terraform

Terraform is an infrastructure automation tool to provision and manage resources in clouds or data centers. To deploy Charmed PostgreSQL using Terraform and Juju, you can use the Juju Terraform Provider.

The easiest way is to start from these examples of terraform modules prepared by Canonical. This page will guide you through a deployment using an example module for PostgreSQL on machines.

For an in-depth introduction to the Juju Terraform Provider, read this Discourse post.

Note: Storage support was added in Juju Terraform Provider version 0.13+.

Summary


Install Terraform tooling

This guide assumes Juju is installed and you have an LXD controller already bootstrapped. For more information, check the Set up the environment tutorial page.

First, install Terraform Provider and example modules:

sudo snap install terraform --classic

Switch to the LXD provider and create a new model:

juju switch lxd
juju add-model my-model

Clone examples and navigate to the PostgreSQL machine module:

git clone https://github.com/canonical/terraform-modules.git
cd terraform-modules/modules/machine/postgresql

Initialise the Juju Terraform Provider:

terraform init

Verify the deployment

Open the main.tf file to see the brief contents of the Terraform module:

resource "juju_application" "machine_postgresql" {
  name  = "postgresql"
  model = "my-model"

  charm {
    name    = "postgresql"
    channel = "14/stable"
  }

  config = {
    plugin_hstore_enable  = true
    plugin_pg_trgm_enable = true
  }

  units = 1
}

Run terraform plan to get a preview of the changes that will be made:

terraform plan -var "juju_model_name=my-model"

Apply the deployment

If everything looks correct, deploy the resources (skip the approval):

terraform apply -auto-approve -var "juju_model_name=my-model"

Check deployment status

Check the deployment status with

juju status --model lxd:my-model --watch 1s

Sample output:

Model         Controller  Cloud/Region         Version  SLA          Timestamp
my-model  lxd         localhost/localhost  3.5.2    unsupported  14:04:26+02:00

App         Version  Status  Scale  Charm       Channel    Rev  Exposed  Message
postgresql  14.11    active      1  postgresql  14/stable  429  no       

Unit           Workload  Agent  Machine  Public address  Ports     Message
postgresql/0*  active    idle   0        10.142.152.90   5432/tcp  Primary

Machine  State    Address        Inst id        Base          AZ  Message
0        started  10.142.152.90  juju-1ea4a4-0  ubuntu@22.04      Running

Continue to operate the charm as usual from here or apply further Terraform changes.

Clean up

To keep the house clean, remove the newly deployed Charmed PostgreSQL by running

terraform destroy -var "juju_model_name=my-model"

Sample output:

juju_application.machine_postgresql: Refreshing state... [id=my-model:postgresql]

Terraform used the selected providers to generate the following execution plan. Resource actions are indicated with the following symbols:
  - destroy

Terraform will perform the following actions:

  # juju_application.machine_postgresql will be destroyed
  - resource "juju_application" "machine_postgresql" {
      - config      = {
          - "plugin_hstore_enable"  = "true"
          - "plugin_pg_trgm_enable" = "true"
        } -> null
      - constraints = "arch=amd64" -> null
      - id          = "my-model:postgresql" -> null
      - model       = "my-model" -> null
      - name        = "postgresql" -> null
      - placement   = "0" -> null
      - storage     = [
          - {
              - count = 1 -> null
              - label = "pgdata" -> null
              - pool  = "rootfs" -> null
              - size  = "99G" -> null
            },
        ] -> null
      - trust       = true -> null
      - units       = 1 -> null

      - charm {
          - base     = "ubuntu@22.04" -> null
          - channel  = "14/stable" -> null
          - name     = "postgresql" -> null
          - revision = 429 -> null
          - series   = "jammy" -> null
        }
    }

Plan: 0 to add, 0 to change, 1 to destroy.

Changes to Outputs:
  - application_name = "postgresql" -> null

Do you really want to destroy all resources?
  Terraform will destroy all your managed infrastructure, as shown above.
  There is no undo. Only 'yes' will be accepted to confirm.

  Enter a value: yes

juju_application.machine_postgresql: Destroying... [id=my-model:postgresql]
juju_application.machine_postgresql: Destruction complete after 1s

Destroy complete! Resources: 1 destroyed.

If you expect having several concurrent connections frequently, it is highly recommended to deploy PgBouncer alongside PostgreSQL. For more information, read our explanation about Connection pooling.

For more examples of Terraform modules for VM, including PostgreSQL HA and PostgreSQL + PgBouncer, see the other directories in the terraform-modules repository.

Feel free to contact us if you have any question and collaborate with us on GitHub!

Last updated a month ago. Help improve this document in the forum.