Apa itu Datasource? #

# resource — Terraform MEMILIKI ini
# Dibuat, diupdate, dan dihapus oleh Terraform
resource "aws_vpc" "main" {
  cidr_block = "10.0.0.0/16"
  # VPC ini ada karena Terraform membuatnya
  # Jika blok ini dihapus dari konfigurasi → VPC dihapus
}

# data — Terraform MEMBACA ini
# Sudah ada di luar, Terraform hanya query
data "aws_vpc" "existing" {
  id = "vpc-0abcdef1234567890"
  # VPC ini sudah ada sebelum Terraform — dibuat secara manual atau oleh tim lain
  # Terraform hanya membacanya, tidak pernah mengubah atau menghapusnya
}

# Data source dievaluasi dulu, baru resource yang menggunakannya diplan

data "aws_ami" "ubuntu" {
  most_recent = true
  owners      = ["099720109477"]  # Canonical

  filter {
    name   = "name"
    values = ["ubuntu/images/hvm-ssd/ubuntu-*-22.04-amd64-server-*"]
  }
}

resource "aws_instance" "web" {
  # Saat plan, data.aws_ami.ubuntu.id sudah diketahui nilainya
  # (bukan "known after apply" seperti resource)
  ami           = data.aws_ami.ubuntu.id
  instance_type = "t3.micro"
}

# 1. AMI terbaru
data "aws_ami" "ubuntu_22_04" {
  most_recent = true
  owners      = ["099720109477"]

  filter {
    name   = "name"
    values = ["ubuntu/images/hvm-ssd/ubuntu-jammy-22.04-amd64-server-*"]
  }

  filter {
    name   = "virtualization-type"
    values = ["hvm"]
  }
}

# 2. Informasi akun AWS saat ini
data "aws_caller_identity" "current" {}

data "aws_region" "current" {}

# Gunakan dalam konfigurasi:
resource "aws_s3_bucket" "logs" {
  bucket = "logs-${data.aws_caller_identity.current.account_id}-${data.aws_region.current.name}"
}

# 3. Availability Zones yang tersedia
data "aws_availability_zones" "available" {
  state = "available"
}

resource "aws_subnet" "public" {
  count             = 3
  availability_zone = data.aws_availability_zones.available.names[count.index]
  vpc_id            = aws_vpc.main.id
  cidr_block        = "10.0.${count.index}.0/24"
}

# 4. VPC atau resource yang dikelola Terraform lain
data "aws_vpc" "shared_services" {
  tags = {
    Name = "shared-services-vpc"
    Environment = "production"
  }
}

# 5. Secret dari AWS Secrets Manager
data "aws_secretsmanager_secret_version" "api_key" {
  secret_id = "production/myapp/api-key"
}

# 6. IAM policy document
data "aws_iam_policy_document" "assume_role" {
  statement {
    effect  = "Allow"
    actions = ["sts:AssumeRole"]

    principals {
      type        = "Service"
      identifiers = ["ec2.amazonaws.com"]
    }
  }
}

resource "aws_iam_role" "instance" {
  assume_role_policy = data.aws_iam_policy_document.assume_role.json
}

# ANTI-PATTERN: Filter yang terlalu lebar — mungkin mengembalikan >1 hasil
data "aws_instance" "web" {
  filter {
    name   = "instance-state-name"
    values = ["running"]  # Ada banyak instance yang running!
  }
  # Error: Your query returned more than one result. Please try a more specific search criteria.
}

# BENAR: Filter yang cukup spesifik untuk mengembalikan tepat 1 hasil
data "aws_instance" "web" {
  filter {
    name   = "tag:Name"
    values = ["web-server-production"]
  }

  filter {
    name   = "instance-state-name"
    values = ["running"]
  }
}

# DATA SOURCE: Ambil dari API cloud
data "aws_ami" "latest" {
  most_recent = true
  owners      = ["amazon"]
  filter {
    name   = "name"
    values = ["amzn2-ami-hvm-*"]
  }
}

# VARIABLE: Diberikan oleh user atau tfvars
variable "ami_id" {
  type    = string
  default = "ami-12345"
}

# Kapan data source lebih baik:
# - Nilai berubah seiring waktu (AMI terbaru, zone availability)
# - Nilai bergantung pada resource yang sudah ada di cloud
# - Butuh informasi yang tidak diketahui user

# Kapan variable lebih baik:
# - Nilai sudah diketahui dan spesifik
# - Butuh validasi custom
# - Nilai berbeda per environment

# ANTI-PATTERN: Mengambil terlalu banyak data
data "aws_instances" "all" {
  # Mengambil SEMUA instance di account → API call mahal
  # Tidak ada filter → response sangat besar
}

# BENAR: Filter sedini mungkin
data "aws_instances" "running" {
  filter {
    name   = "instance-state-name"
    values = ["running"]
  }
  
  filter {
    name   = "tag:Environment"
    values = [var.environment]
  }
}

# Ambil info VPC yang sudah ada (shared infra)
data "aws_vpc" "main" {
  filter {
    name   = "tag:Environment"
    values = [var.environment]
  }
}

data "aws_subnets" "private" {
  filter {
    name   = "vpc-id"
    values = [data.aws_vpc.main.id]
  }
  filter {
    name   = "tag:Tier"
    values = ["private"]
  }
}

# Gunakan data source untuk conditional resource
resource "aws_instance" "web" {
  count         = var.environment == "production" ? 3 : 1
  ami           = data.aws_ami.latest.id
  instance_type = var.environment == "production" ? "m5.large" : "t3.micro"
  subnet_id     = data.aws_subnets.private.ids[0]
}

# Error umum:

# 1. Resource tidak ditemukan
# Error: no matching AMI found
# → Periksa filter, pastikan resource benar-benar ada

# 2. Multiple matches
# Error: multiple VPCs matched; use additional constraints
# → Tambahkan filter lebih spesifik

# 3. Permission denied
# Error: UnauthorizedOperation
# → Pastikan IAM policy punya izin read untuk resource type

# 4. API rate limit
# Error: Throttling
# → Kurangi data source, gunakan caching, atau retry

# Tips: gunakan try() untuk graceful degradation
locals {
  # Jika data source gagal, gunakan default
  vpc_cidr = try(data.aws_vpc.main.cidr_block, "10.0.0.0/16")
  
  # Jika data source mengembalikan kosong
  subnet_ids = length(data.aws_subnets.private.ids) > 0     ? data.aws_subnets.private.ids     : [aws_subnet.fallback.id]
}

DATA SOURCE LIFECYCLE:

terraform init
  └── Download providers

terraform plan
  ├── Read data source (API call)
  ├── Store result in state
  ├── Use result untuk compute resource arguments
  └── Plan resource changes

terraform apply
  ├── Read data source lagi (refresh)
  ├── Apply resource changes
  └── Update state

terraform destroy
  └── Hapus managed resources (data source TIDAK dihapus)

# Data source SELALU di-refresh saat plan/apply
# Ini bisa lambat jika banyak data source

# Percepat: gunakan -refresh=false (tidak recommended)
terraform plan -refresh=false

# Lebih baik: kurangi jumlah data source
# atau gunakan caching mechanism