fix: Datahub upgrade to version 0.12.0

analytics/terraform/datahub-on-eks/datahub-addon/main.tf

@@ -38,7 +38,6 @@ module "prereq" {
   vpc_cidr            = var.vpc_cidr
   vpc_private_subnets = var.vpc_private_subnets
 
-  create_iam_service_linked_role_es = var.create_iam_service_linked_role_es
 }
 
 resource "kubernetes_namespace" "datahub" {

analytics/terraform/datahub-on-eks/datahub-addon/modules/prereq/main.tf

@@ -1,7 +1,3 @@
-locals {
-  cidr_blocks = [coalesce(var.vpc_cidr, "10.1.0.0/16")]
-}
-
 #---------------------------------------------------------------
 # OpenSearch For DataHub metadata
 #---------------------------------------------------------------
@@ -13,15 +9,10 @@ resource "aws_security_group" "es" {
     from_port   = 0
     to_port     = 0
     protocol    = "-1"
-    cidr_blocks = local.cidr_blocks
+    cidr_blocks = [var.vpc_cidr]
   }
 }
 
-resource "aws_iam_service_linked_role" "es" {
-  aws_service_name = "opensearchservice.amazonaws.com"
-  count            = var.create_iam_service_linked_role_es ? 1 : 0
-}
-
 resource "random_password" "master_password" {
   length      = 16
   special     = true
@@ -32,7 +23,6 @@ resource "random_password" "master_password" {
 }
 
 resource "aws_opensearch_domain" "es" {
-  depends_on     = [aws_iam_service_linked_role.es]
   domain_name    = "${var.prefix}-es-domain"
   engine_version = "OpenSearch_1.1"
   cluster_config {
@@ -78,6 +68,7 @@ resource "aws_opensearch_domain" "es" {
 
 # Creating the AWS Elasticsearch domain policy
 resource "aws_opensearch_domain_policy" "main" {
+  depends_on      = [aws_msk_cluster.msk]
   domain_name     = aws_opensearch_domain.es.domain_name
   access_policies = <<POLICIES
 {
@@ -105,7 +96,7 @@ resource "aws_security_group" "msk" {
     from_port   = 0
     to_port     = 0
     protocol    = "-1"
-    cidr_blocks = local.cidr_blocks
+    cidr_blocks = [var.vpc_cidr]
   }
 }
 
@@ -184,7 +175,7 @@ resource "aws_security_group" "rds" {
     from_port   = 0
     to_port     = 3306
     protocol    = "tcp"
-    cidr_blocks = local.cidr_blocks
+    cidr_blocks = [var.vpc_cidr]
   }
 }
 

analytics/terraform/datahub-on-eks/datahub-addon/modules/prereq/variables.tf

@@ -21,9 +21,3 @@ variable "vpc_cidr" {
   type        = string
   default     = ""
 }
-
-variable "create_iam_service_linked_role_es" {
-  type        = bool
-  default     = true
-  description = "Whether to create `AWSServiceRoleForAmazonOpensearchService` service-linked role. Set it to `false` if the role already exists"
-}

analytics/terraform/datahub-on-eks/datahub-addon/values/datahub_values.yaml

@@ -251,7 +251,7 @@ global:
       #   secretKey: postgres-password
 
   datahub:
-    version: v0.10.4
+    version: v0.12.0
 
     systemUpdate:
       enabled: true
@@ -277,7 +277,7 @@ global:
 
     managed_ingestion:
       enabled: true
-      defaultCliVersion: "0.10.4"
+      defaultCliVersion: "0.12.0"
 
     metadata_service_authentication:
       enabled: true

analytics/terraform/datahub-on-eks/datahub-addon/variables.tf

@@ -33,9 +33,3 @@ variable "vpc_cidr" {
   type        = string
   default     = ""
 }
-
-variable "create_iam_service_linked_role_es" {
-  type        = bool
-  default     = true
-  description = "Whether to create `AWSServiceRoleForAmazonOpensearchService` service-linked role. Set it to `false` if the role already exists"
-}

analytics/terraform/datahub-on-eks/datahub.tf

@@ -2,9 +2,8 @@ module "datahub" {
   depends_on          = [module.eks, module.eks_blueprints_addons]
   source              = "./datahub-addon"
   prefix              = local.name
-  vpc_id              = module.vpc.vpc_id
+  vpc_id              = local.vpc_id
   vpc_cidr            = local.vpc_cidr
-  vpc_private_subnets = module.vpc.private_subnets
+  vpc_private_subnets = local.private_subnets
 
-  create_iam_service_linked_role_es = var.create_iam_service_linked_role_es
 }

analytics/terraform/datahub-on-eks/install.sh

@@ -9,7 +9,6 @@ terraform init || echo "\"terraform init\" failed"
 # List of Terraform modules to apply in sequence
 targets=(
   "module.vpc"
-  "module.vpc_endpoints_sg"
   "module.vpc_endpoints"
   "module.eks"
   "module.ebs_csi_driver_irsa"
@@ -22,7 +21,7 @@ targets=(
 for target in "${targets[@]}"
 do
   echo "Applying module $target..."
-  apply_output=$(terraform apply -target="$target" -auto-approve 2>&1 | tee /dev/tty)
+  apply_output=$(terraform apply -target="$target" -var="region=$region" -auto-approve 2>&1 | tee /dev/tty)
   if [[ ${PIPESTATUS[0]} -eq 0 && $apply_output == *"Apply complete"* ]]; then
     echo "SUCCESS: Terraform apply of $target completed successfully"
   else
@@ -33,7 +32,7 @@ done
 
 # Final apply to catch any remaining resources
 echo "Applying remaining resources..."
-apply_output=$(terraform apply -auto-approve 2>&1 | tee /dev/tty)
+apply_output=$(terraform apply -var="region=$region" -auto-approve 2>&1 | tee /dev/tty)
 if [[ ${PIPESTATUS[0]} -eq 0 && $apply_output == *"Apply complete"* ]]; then
   echo "SUCCESS: Terraform apply of all modules completed successfully"
 else

analytics/terraform/datahub-on-eks/locals.tf

@@ -9,4 +9,7 @@ locals {
     Blueprint  = local.name
     GithubRepo = "github.com/awslabs/data-on-eks"
   })
+
+  vpc_id          = var.create_vpc ? module.vpc.vpc_id : var.vpc_id
+  private_subnets = var.create_vpc ? module.vpc.private_subnets : var.private_subnet_ids
 }

analytics/terraform/datahub-on-eks/main.tf

@@ -8,8 +8,8 @@ module "eks" {
   cluster_endpoint_private_access = true # if true, Kubernetes API requests within your cluster's VPC (such as node to control plane communication) use the private VPC endpoint
   cluster_endpoint_public_access  = true # if true, Your cluster API server is accessible from the internet. You can, optionally, limit the CIDR blocks that can access the public endpoint.
 
-  vpc_id     = module.vpc.vpc_id
-  subnet_ids = module.vpc.private_subnets
+  vpc_id     = local.vpc_id
+  subnet_ids = local.private_subnets
 
 
   eks_managed_node_group_defaults = {
@@ -26,7 +26,7 @@ module "eks" {
     core_node_group = {
       name        = "core-node-group"
       description = "EKS managed node group example launch template"
-      subnet_ids  = module.vpc.private_subnets
+      subnet_ids  = local.private_subnets
 
       min_size     = 1
       max_size     = 9

analytics/terraform/datahub-on-eks/providers.tf

@@ -15,11 +15,12 @@ provider "helm" {
     token                  = data.aws_eks_cluster_auth.this.token
   }
 }
-
+/*
 provider "kubectl" {
   apply_retry_count      = 30
   host                   = module.eks.cluster_endpoint
   cluster_ca_certificate = base64decode(module.eks.cluster_certificate_authority_data)
   load_config_file       = false
   token                  = data.aws_eks_cluster_auth.this.token
 }
+*/

analytics/terraform/datahub-on-eks/variables.tf

@@ -23,11 +23,29 @@ variable "tags" {
 }
 
 variable "vpc_cidr" {
-  description = "VPC CIDR"
+  description = "VPC CIDR - for existing or new VPC"
   default     = "10.1.0.0/16"
   type        = string
 }
 
+variable "create_vpc" {
+  description = "Create VPC"
+  default     = true
+  type        = bool
+}
+
+variable "vpc_id" {
+  description = "VPC Id - with create_vpc set to false"
+  default     = "vpc-0bcc567a949321822"
+  type        = string
+}
+
+variable "private_subnet_ids" {
+  description = "Private subnet ids - needed with create_vpc set to false"
+  default     = ["subnet-066c6579a95470d55", "subnet-0b4bf52219857602f"]
+  type        = list(string)
+}
+
 variable "enable_vpc_endpoints" {
   description = "Enable VPC Endpoints"
   default     = false
@@ -46,9 +64,3 @@ variable "private_subnets" {
   default     = ["10.1.0.0/17", "10.1.128.0/18"]
   type        = list(string)
 }
-
-variable "create_iam_service_linked_role_es" {
-  type        = bool
-  default     = true
-  description = "Whether to create `AWSServiceRoleForAmazonOpensearchService` service-linked role. Set it to `false` if the role already exists"
-}

analytics/terraform/datahub-on-eks/vpc.tf

@@ -8,6 +8,8 @@ module "vpc" {
   source  = "terraform-aws-modules/vpc/aws"
   version = "~> 5.0"
 
+  create_vpc = var.create_vpc
+
   name           = local.name
   cidr           = local.vpc_cidr
   azs            = local.azs
@@ -36,43 +38,24 @@ module "vpc" {
   tags = local.tags
 }
 
-module "vpc_endpoints_sg" {
-  source  = "terraform-aws-modules/security-group/aws"
-  version = "~> 5.0"
-
-  create = var.enable_vpc_endpoints
-
-  name        = "${local.name}-vpc-endpoints"
-  description = "Security group for VPC endpoint access"
-  vpc_id      = module.vpc.vpc_id
-
-  ingress_with_cidr_blocks = [
-    {
-      rule        = "https-443-tcp"
-      description = "VPC CIDR HTTPS"
-      cidr_blocks = join(",", module.vpc.private_subnets_cidr_blocks)
-    },
-  ]
-
-  egress_with_cidr_blocks = [
-    {
-      rule        = "https-443-tcp"
-      description = "All egress HTTPS"
-      cidr_blocks = "0.0.0.0/0"
-    },
-  ]
-
-  tags = local.tags
-}
-
 module "vpc_endpoints" {
   source  = "terraform-aws-modules/vpc/aws//modules/vpc-endpoints"
-  version = "~> 5.0"
+  version = "~> 5.1"
 
   create = var.enable_vpc_endpoints
 
-  vpc_id             = module.vpc.vpc_id
-  security_group_ids = [module.vpc_endpoints_sg.security_group_id]
+  vpc_id = module.vpc.vpc_id
+
+  # Security group
+  create_security_group      = true
+  security_group_name_prefix = "${local.name}-vpc-endpoints-"
+  security_group_description = "VPC endpoint security group"
+  security_group_rules = {
+    ingress_https = {
+      description = "HTTPS from VPC"
+      cidr_blocks = [module.vpc.vpc_cidr_block]
+    }
+  }
 
   endpoints = merge({
     s3 = {

website/docs/blueprints/data-analytics/datahub-on-eks.md

@@ -24,16 +24,19 @@ DataHub also depends on many underlying infrastructure and services to function,
 
 ## Deploying the Solution
 
-This blueprint deploys an EKS Cluster into a new VPC.
+This blueprint deploys an EKS Cluster into a new VPC by default:
 
 - Creates a new sample VPC, 2 Private Subnets and 2 Public Subnets
 - Creates Internet gateway for Public Subnets and NAT Gateway for Private Subnets
+
+You may also deploy to an existing VPC by setting value for `create_vpc` variable to `false` and specify `vpc_id`, `private_subnet_ids`, and `vpc_cidr` values.
+
 - Creates EKS Cluster Control plane with public endpoint (for demo reasons only) with core managed node group, on-demand node group and Spot node group for Spark workloads.
 - Deploys Metrics server, Cluster Autoscaler, Prometheus server and AMP workspace, and AWS LoadBalancer Controller.
 
 It then provisions the storage services for DataHub.
 
-- Creates service-linked role, security group, and an OpenSearch domain with one data node in each of the private subnets / AZs that EKS cluster is deployed on.
+- Creates security group, and an OpenSearch domain with one data node in each of the private subnets / AZs that EKS cluster is deployed on.
 - Creates security group, kms key, and configuration for MSK.  Creates the MSK cluster with one broker in each of the private subnets.
 - Creates an RDS MySQL db instance with multi-AZ enabled.
 
@@ -55,6 +58,14 @@ Ensure that you have installed the following tools on your machine.
 2. [kubectl](https://Kubernetes.io/docs/tasks/tools/)
 3. [terraform](https://learn.hashicorp.com/tutorials/terraform/install-cli)
 
+Also, you need opensearch service-linked role created in the account.  To verify and create the role if needed, run:
+```
+aws iam create-service-linked-role --aws-service-name opensearchservice.amazonaws.com || true
+```
+
+# If the role has already been successfully created, you will see:
+# An error occurred (InvalidInput) when calling the CreateServiceLinkedRole operation: Service role name AWSServiceRoleForOpenSearch has been taken in this account, please try a different suffix.
+
 ### Deploy
 
 Clone the repository