terraform-provider-confluent

README

Terraform Confluent Platform Provider

This is provider for Confluent Platform, not Confluent Cloud. If you're seeking the Confluent Cloud provider, please visit This project.

But if you're seeking the terraform provider to implement IaC for your hosted Confluent Platform, this project is exactly what you need.

Getting started

1. Installation

• Clone this project

git clone git@github.com:wayarmy/gonfluent.git && cd gonfluent

• Build the binary:

make prepare
make compile

• Find your own binary in the bin folder. For example, if you're running terraform in the linux X86-64 machine, you can copy linux_amd64 binary into terraform plugins directory:

cp bin/terraform-provider-confluent-kafka_v0.1.0 ${HOME}/.terraform.d/plugins/registry.terraform.io/hashicorp/confluent-kafka/0.1.0/linux_amd64

2. Start your first terraform plan

Create main.tf file:

terraform {
  required_providers {
    confluent-kafka = {
      version = "0.1.0"
      source = "confluent-kafka"
    }
  }
}

Init

terraform init

Plan

terraform plan

Apply

terraform apply

3. Resources supported

• Define the provider

provider "confluent-kafka" {
  alias = "confluent" # Alias of the provider
  bootstrap_servers = ["localhost:9093"] # List of Kafka bootstrap servers
  ca_cert           = "certs/ca.pem" # CA cert to connect to Kafka cluster
  client_cert       = "certs/cert.pem" # Client certs to connect to Kafka cluster
  client_key        = "certs/key.pem" # Client private key to connect to Kafka cluster
  skip_tls_verify   = true # Skip TlS verification
  username = "xxxx" # LDAP or SASL username to connect to Confluent or MDS
  password = "yyyy" # LDAP or SASL password to connect to Confluent or MDS
}

3.1 Topics

• Topic Example

resource "kafka_topic" "example_topic" {
  cluster_id = "cluster-1" # Optional: If not, terraform will use first cluster in the cluster list
  name = "test-terraform-confluent-provider" # Topic name
  replication_factor = 3 # Replication factor
  partitions = 5 # The number of partition

  config = {
    "segment.ms" = "20000"
    "cleanup.policy" = "compact"
  }
  provider = confluent-kafka.confluent
}

3.2 Cluster role binding

• Will describe and bind the cluster role to principal (User or scope)

• Example:

resource "cluster_role_binding" "example_role_binding" {
  cluster_id = "cluster-1" # Optional: If not, terraform will use first cluster in the cluster list
  role = "UserAdmin" # Allow roles: "AuditAdmin", "ClusterAdmin", "DeveloperManage", "DeveloperRead", "DeveloperWrite", "Operator", "ResourceOwner", "SecurityAdmin", "SystemAdmin", "UserAdmin",
  principal = "User:username" # Allow convention: User:<user_name> or Group:<group_name>
  cluster_type = "Kafka" # Support 4 types of clusters: Kafka, SchemaRegistry, KSQL, Connect
  provider = confluent-kafka.confluent
}

3.3 Kafka topic RBAC

• Will describe and bind the resource role (Not Cluster role) to principal

• Example:

resource "kafka_topic_rbac" "example_topic_rbac" {
  principal = "User:username" # Allow convention: User:<user_name>, Group:<group_name>, User:CN=<domain>
  role = "ResourceOwner" # Allow roles: "DeveloperRead", "DeveloperWrite", "Operator", "ResourceOwner"
  resource_type = "Topic" # Allow only: Topic
  pattern_type = "PREFIXED" # Allow: PREFIXED and LITERAL
  name = "test-" # The pattern contains in topic name
  cluster_id = "cluster-1" # Optional: If not, terraform will use first cluster in the cluster list
  provider = confluent-kafka.confluent
}