digitalocean_database_cluster

Provides a DigitalOcean database cluster resource.

Example Usage

Create a new PostgreSQL database cluster

resource "digitalocean_database_cluster" "postgres-example" {
  name       = "example-postgres-cluster"
  engine     = "pg"
  version    = "15"
  size       = "db-s-1vcpu-1gb"
  region     = "nyc1"
  node_count = 1
}

Create a new MySQL database cluster

resource "digitalocean_database_cluster" "mysql-example" {
  name       = "example-mysql-cluster"
  engine     = "mysql"
  version    = "8"
  size       = "db-s-1vcpu-1gb"
  region     = "nyc1"
  node_count = 1
}

Create a new Redis database cluster

resource "digitalocean_database_cluster" "redis-example" {
  name       = "example-redis-cluster"
  engine     = "redis"
  version    = "7"
  size       = "db-s-1vcpu-1gb"
  region     = "nyc1"
  node_count = 1
}

Create a new Kafka database cluster

resource "digitalocean_database_cluster" "kafka-example" {
  name       = "example-kafka-cluster"
  engine     = "kafka"
  version    = "3.5"
  size       = "db-s-2vcpu-2gb"
  region     = "nyc1"
  node_count = 3
}

Create a new MongoDB database cluster

resource "digitalocean_database_cluster" "mongodb-example" {
  name       = "example-mongo-cluster"
  engine     = "mongodb"
  version    = "6"
  size       = "db-s-1vcpu-1gb"
  region     = "nyc3"
  node_count = 1
}

Create a new database cluster based on a backup of an existing cluster.

resource "digitalocean_database_cluster" "doby" {
  name       = "dobydb"
  engine     = "pg"
  version    = "15"
  size       = "db-s-1vcpu-2gb"
  region     = "nyc1"
  node_count = 1
  tags       = ["production"]
}

resource "digitalocean_database_cluster" "doby_backup" {
  name       = "dobydupe"
  engine     = "pg"
  version    = "15"
  size       = "db-s-1vcpu-2gb"
  region     = "nyc1"
  node_count = 1
  tags       = ["production"]

  backup_restore {
    database_name = "dobydb"
  }

  depends_on = [
    digitalocean_database_cluster.doby
  ]
}

Argument Reference

The following arguments are supported:

• name - (Required) The name of the database cluster.
• engine - (Required) Database engine used by the cluster (ex. pg for PostreSQL, mysql for MySQL, redis for Redis, mongodb for MongoDB, or kafka for Kafka).
• size - (Required) Database Droplet size associated with the cluster (ex. db-s-1vcpu-1gb). See here for a list of valid size slugs.
• region - (Required) DigitalOcean region where the cluster will reside.
• node_count - (Required) Number of nodes that will be included in the cluster. For kafka clusters, this must be 3.
• version - (Required) Engine version used by the cluster (ex. 14 for PostgreSQL 14). When this value is changed, a call to the Upgrade major Version for a Database API operation is made with the new version.
• tags - (Optional) A list of tag names to be applied to the database cluster.
• private_network_uuid - (Optional) The ID of the VPC where the database cluster will be located.
• project_id - (Optional) The ID of the project that the database cluster is assigned to. If excluded when creating a new database cluster, it will be assigned to your default project.
• eviction_policy - (Optional) A string specifying the eviction policy for a Redis cluster. Valid values are: noeviction, allkeys_lru, allkeys_random, volatile_lru, volatile_random, or volatile_ttl.
• sql_mode - (Optional) A comma separated string specifying the SQL modes for a MySQL cluster.
• maintenance_window - (Optional) Defines when the automatic maintenance should be performed for the database cluster.
• storage_size_mib - (Optional) Defines the disk size, in MiB, allocated to the cluster. This can be adjusted on MySQL and PostreSQL clusters based on predefined ranges for each slug/droplet size.

maintenance_window supports the following:

• day - (Required) The day of the week on which to apply maintenance updates.

• hour - (Required) The hour in UTC at which maintenance updates will be applied in 24 hour format.

• backup_restore - (Optional) Create a new database cluster based on a backup of an existing cluster.

backup_restore supports the following:

• database_name - (Required) The name of an existing database cluster from which the backup will be restored.
• backup_created_at - (Optional) The timestamp of an existing database cluster backup in ISO8601 combined date and time format. The most recent backup will be used if excluded.

This resource supports customized create timeouts. The default timeout is 30 minutes.

Attributes Reference

In addition to the above arguments, the following attributes are exported:

• id - The ID of the database cluster.
• urn - The uniform resource name of the database cluster.
• host - Database cluster’s hostname.
• private_host - Same as host, but only accessible from resources within the account and in the same region.
• port - Network port that the database cluster is listening on.
• uri - The full URI for connecting to the database cluster.
• private_uri - Same as uri, but only accessible from resources within the account and in the same region.
• database - Name of the cluster’s default database.
• user - Username for the cluster’s default user.
• password - Password for the cluster’s default user.

Import

Database clusters can be imported using the id returned from DigitalOcean, e.g.

terraform import digitalocean_database_cluster.mycluster 245bcfd0-7f31-4ce6-a2bc-475a116cca97