Skip to content

Latest commit

 

History

History
363 lines (304 loc) · 16.5 KB

README.md

File metadata and controls

363 lines (304 loc) · 16.5 KB
Raw

Terraform Instaclustr Provider

CircleCI GoLang Version Latest Release Version License

Quality Gate Status

A Terraform provider for managing Instaclustr Platform resources.

It provides a flexible set of resources for provisioning and managing Apache Cassandra, Apache Kafka, Apache Zookeeper, OpenSearch and Redis clusters on the Instaclustr Managed Platform via Terraform.

For further information about Instaclustr, please see the Instaclustr website and Support Pages

For general information about Terraform, visit the official website and GitHub project page.

Key benefits

  • Removes the need to write custom code integration directly with the Instaclustr API
  • Instaclustr based infrastructure as code deployments with minimal effort
  • Ease of integration into existing terraform or automated CI/CD based workflows
  • Ease of customisation and configuration in order to meet operational requirements
  • Use of existing Instaclustr authentication methodologies

Requirements

  • Terraform v0.10.x - .v0.15.x.
  • Go 1.14 or higher

Using The Provider

To install this provider using Terraform 0.13+, copy and paste this code into your Terraform configuration. Then, run terraform init.

terraform {
  required_providers {
    instaclustr = {
      source = "instaclustr/instaclustr"
      version = "1.9.9"
    }
  }
}

provider "instaclustr" {
  # Configuration options
}

For further details on Provider installation, and installation on older versions of terraform please see the Terraform installation guide.

Authentication

This provider requires an API Key in order to provision Instaclustr resources. To create an API key, please log into the Instaclustr Console or signup for an account here if you don't have one. Navigate to Account -> API Keys page, locate the Provisioning role and click Generate Key. This username and API key combination should be placed into the provider configuration:

provider "instaclustr" {
    username = "<Your instaclustr username here>"
    api_key = "<Your provisioning API key here>"
}

If you wish to keep secrets in the ENV instead of stored in your terraform file use the following method:

In console export the desired variable:

export api_key={instaclustrAPIkey}

In your terraform file create a variable:

variable "api_key" {
 type = string
 default = "xxx"
}

In the provider block use the variable:

provider "instaclustr" {
    username= "<Your instaclustr username>"
    api_key = var.api_key
}

When running terraform plan/apply, pipe in the variables as follows:

terraform apply -var= "api_key=$api_key"

Example Usage

It's possible to provision clusters for different cloud providers by changing the variable cluster_provider. The accepted cloud providers are: AWS, GCP, AZURE.

AWS:

resource "instaclustr_cluster" "example" {
    cluster_name = "testcluster"
    node_size = "m5l-250-v2"
    data_centre = "US_WEST_2"
    sla_tier = "NON_PRODUCTION"
    cluster_network = "192.168.0.0/18"
    private_network_cluster = false
    pci_compliant_cluster = false
    cluster_provider = {
        name = "AWS_VPC"
    }
    rack_allocation = {
        number_of_racks = 3
        nodes_per_rack = 1
    }
    bundle {
        bundle = "APACHE_CASSANDRA"
        version = "3.11.8"
        options = {
          auth_n_authz = true
        }
      }
      bundle {
        bundle = "SPARK"
        version = "2.3.2"
      }
}

AZURE:

resource "instaclustr_cluster" "azure_example" {
  cluster_name = "testcluster"
  node_size = "Standard_DS2_v2-256-an"
  data_centre = "CENTRAL_US"
  sla_tier = "NON_PRODUCTION"
  cluster_network = "192.168.0.0/18"
  private_network_cluster = false
  cluster_provider = {
    name = "AZURE_AZ"
  }
  rack_allocation = {
    number_of_racks = 3
    nodes_per_rack = 1
  }

  bundle {
    bundle = "APACHE_CASSANDRA"
    version = "3.11.8"
    options = {
      auth_n_authz = true
    }
  }
}

GCP:

resource "instaclustr_cluster" "gcp_example" {
  cluster_name = "testclustergcp"
  node_size = "n1-standard-2"
  data_centre = "us-east1"
  sla_tier = "NON_PRODUCTION"
  cluster_network = "192.168.0.0/18"
  private_network_cluster = false
  cluster_provider = {
    name = "GCP"
  }
  rack_allocation = {
    number_of_racks = 3
    nodes_per_rack = 1
  }

  bundle {
    bundle = "APACHE_CASSANDRA"
    version = "3.11.8"
    options = {
      auth_n_authz = true
    }
  }
}

Multi Data Centre Provisioning:

For Multi Data Centre provisioning, please specify node_size, rack_allocation, provider and bundles in each data_centres;
For each data_centres, it requires at least one bundles to be the base application, e.g. APACHE_CASSANDRA.

resource "instaclustr_cluster" "multi_DC_example" {
  cluster_name = "testcluster_multiDC"
  sla_tier     = "NON_PRODUCTION"

  data_centres {
    name        = "DC1"
    data_centre = "US_WEST_1"
    network     = "10.1.0.0/18"
    node_size    = "m5l-250-v2"
    rack_allocation = {
      number_of_racks = 2
      nodes_per_rack  = 1
    }
    provider = {
      name = "AWS_VPC"
    }
    bundles {
      bundle = "APACHE_CASSANDRA"
      version = "3.11.8"
      options = {
        auth_n_authz = true
        use_private_broadcast_rpc_address = false
        client_encryption = false
        lucene_enabled = false
        continuous_backup_enabled = true
      }
    }
    bundles {
      bundle = "SPARK"
      version = "2.3.2"
    }
  }

  data_centres {
    name        = "DC2"
    data_centre = "CENTRAL_US"
    network     = "10.0.0.0/18"
    node_size    = "Standard_DS2_v2-256-an"
    rack_allocation = {
      number_of_racks = 2
      nodes_per_rack  = 1
    }
    provider = {
      name = "AZURE_AZ"
    }
    bundles {
      bundle = "APACHE_CASSANDRA"
      version = "3.11.8"
      options = {
        auth_n_authz = true
        use_private_broadcast_rpc_address = false
        client_encryption = false
        lucene_enabled = false
        continuous_backup_enabled = true
      }
    }
  }

  data_centres {
    name = "DC3"
    data_centre = "US_WEST_2"
    network     = "192.168.0.0/18"
    node_size    = "m5l-250-v2"
    rack_allocation = {
      number_of_racks = 2
      nodes_per_rack  = 1
    }
    provider = {
      name = "AWS_VPC"
    }
    bundles {
      bundle = "APACHE_CASSANDRA"
      version = "3.11.8"
      options = {
        auth_n_authz = true
        use_private_broadcast_rpc_address = false
        client_encryption = false
        lucene_enabled = false
        continuous_backup_enabled = true
      }
    }
    bundles {
      bundle = "SPARK"
      version = "2.3.2"
    }
  }
}

Configuration

Configuration documentation can be found at the Instaclustr Terraform Registry

Bundles and Versions

Bundle Versions Compatible With
APACHE_CASSANDRA 2.2.18, 3.0.19, 3.11.8, 4.0 (preview)
SPARK 2.1.3, 2.3.2 APACHE_CASSANDRA
KAFKA 2.7.1, 2.8.1, 3.0.0, 3.1.1
KAFKA_REST_PROXY 5.0.0 KAFKA
KAFKA_SCHEMA_REGISTRY 5.0.0, 5.0.4 KAFKA
KARAPACE_SCHEMA_REGISTRY 3.2.0 KAFKA
Not compatible with:
KAFKA_REST_PROXY, KAFKA_SCHEMA_REGISTRY
KARAPACE_REST_PROXY 3.2.0 KAFKA
Not compatible with:
KAFKA_REST_PROXY, KAFKA_SCHEMA_REGISTRY
OPENSEARCH 1.3.4, 2.0.0
ELASTICSEARCH (For Legacy Support Only) 1.13.3
KAFKA_CONNECT 2.7.1, 2.8.1, 3.0.0, 3.1.1
REDIS 6.2.7, 7.0.4
APACHE_ZOOKEEPER 3.6.3, 3.7.1
POSTGRESQL 13.8, 14.5
PGBOUNCER 1.17.0 POSTGRESQL
CADENCE 0.22.4

Migrating from 0.0.1 → 1.0.0+

A schema change has been made from 0.0.1 which no longer supports the bundles argument and uses bundle blocks instead. This change can cause terraform apply to fail with a message that bundles has been removed and/or updating isn't supported. To resolve this -

  1. Change all usages of the bundles argument → bundle blocks (example under example/main.tf)
  2. In the .tfstate files, replace all keys named bundles with bundle in resources under the Instaclustr provider.

Contributing

Firstly, thanks! We value your time and will do our best to review the PR as soon as possible.

  1. Install golang

  2. Clone repository to: $GOPATH/src/github.com/instaclustr/terraform-provider-instaclustr

  3. Build and install the provider by $ make build install

  4. For local testing of your changes you will need to use the local provider instead of the provider from the registry, so change your provider config to the following

    terraform {
  required_providers {
    instaclustr = {
      source = "terraform.instaclustr.com/instaclustr/instaclustr"
      version = ">= 1.0.0, < 2.0.0"
    }
  }
}

  5. Run the unit tests by $ make test

  6. Create a PR and send it our way :)

Our Circle CI pipeline will automatically run unit tests when a PR is created and new changes committed. It is also capable of running the acceptance tests, however our staff needs to give a manual approval to run the tests. Passing tests are a requirement to merge a PR. Please let us know when your PR is ready for acceptance tests!

Unit tests are within instaclustr folder with _unit_test suffix, and used to test the internal methods.

Acceptance Testing

Acceptance tests are within acc_test folder, and used to run end-to-end testing. We recommend using CircleCI to run your acceptance tests, however you can run them locally. Acceptance tests require end to end interaction with the Instaclustr platform and will create real (paid) infrastructure. If you wish to perform local testing you must set the variables below and run: make testacc

Variable Command Description
TF_ACC $ export TF_ACC=1 Enables online acceptance tests.
IC_USERNAME $ export IC_USERNAME=<your instaclustr username> Authorizes Provisioning API
IC_API_KEY $ export IC_API_KEY=<your provisioning API key> Authorizes Provisioning API
KMS_ARN $ export KMS_ARN=<your KMS ARN> For EBS encryption of nodes. Note: You cannot use an ARN previously added to your account as an encryption key.
IC_PROV_ACC_NAME $ export IC_PROV_ACC_NAME="<your provider name>" Your "Run In Your Own Account" account name.
IC_PROV_VPC_ID $ export IC_PROV_VPC_ID="<your AWS VPC ID>" For provisioning into a custom VPC.
IC_AWS_ACCESS_KEY $ export IC_PROV_VPC_ID="<access key for the AWS S3 bucket>" For Kafka Connect connection information. See bundle options.
IC_AWS_SECRET_KEY $ export IC_PROV_VPC_ID="<secret key for the AWS S3 bucket>" For Kafka Connect connection information. See bundle options.
IC_S3_BUCKET_NAME $ export IC_PROV_VPC_ID="<AWS S3 bucket name>" For Kafka Connect connection information. See bundle options.
IC_AZURE_STORAGE_ACCOUNT_NAME $ export IC_PROV_VPC_ID="<account name for the AZURE container storage>" For Kafka Connect connection information. See bundle options.
IC_AZURE_STORAGE_ACCOUNT_KEY $ export IC_PROV_VPC_ID="<account key for the AZURE container storage>" For Kafka Connect connection information. See bundle options.
IC_AZURE_STORAGE_CONTAINER_NAME $ export IC_PROV_VPC_ID="<the name of the AZURE container storage>" For Kafka Connect connection information. See bundle options.

Running Specific Tests

To run a specific test, use the testtarget makefile goal. TARGET=TestName make testtarget

Further information and documentation

This provider makes use of the Instaclustr API. For further information including the latest updates and value definitions, please see the provisioning API documentation.

Please see https://www.instaclustr.com/support/documentation/announcements/instaclustr-open-source-project-status/ for Instaclustr support status of this project.

License

Apache2 - See the included LICENSE file for more details.