Best Practice: Greenfield FinOps by Design

new-service/
├── docs/
│   ├── adr/
│   │   └── ADR-TEMPLATE.md          # Mit Cost-Impact-Sektion
│   ├── cost-debt-register.yml        # Initialisiert, leer
│   └── retention-strategy.yml        # Tier-Strategie dokumentiert
├── infrastructure/
│   ├── modules/
│   │   ├── mandatory-tags/           # Pflicht: Cost-Center, Owner, etc.
│   │   └── budget-alert/             # Pflicht: Budget + Alert
│   ├── environments/
│   │   ├── production/
│   │   │   ├── main.tf
│   │   │   ├── budget.tf             # Production Budget
│   │   │   └── variables.tf
│   │   └── staging/
│   │       └── ...
│   └── shared/
│       └── lifecycle-defaults.tf     # Default-Lifecycle für Storage/Logs
├── .github/
│   └── workflows/
│       ├── cost-compliance.yml       # CI-Gate: Tagging, Lifecycle, Budget
│       └── monthly-finops-report.yml
└── tagging-taxonomy.yml

# modules/mandatory-tags/main.tf

variable "cost_center" {
  type = string
  validation {
    condition     = can(regex("^[a-z][a-z0-9-]+$", var.cost_center))
    error_message = "cost-center must be lowercase kebab-case."
  }
}

variable "owner" {
  type = string
  validation {
    condition     = can(regex("^[a-z][a-z0-9-]+$", var.owner))
    error_message = "owner must be lowercase kebab-case (team name, not person)."
  }
}

variable "environment" {
  type = string
  validation {
    condition     = contains(["production", "staging", "development", "testing"], var.environment)
    error_message = "environment must be: production, staging, development, or testing."
  }
}

variable "workload" {
  type        = string
  description = "Service or workload name."
}

variable "additional_tags" {
  type    = map(string)
  default = {}
}

locals {
  base_tags = {
    cost-center          = var.cost_center
    owner                = var.owner
    environment          = var.environment
    workload             = var.workload
    managed-by           = "terraform"
    wafpp-cost-compliant = "true"
  }
}

output "tags" {
  value = merge(local.base_tags, var.additional_tags)
}

# modules/budget-alert/main.tf

variable "budget_name" {
  type = string
}

variable "monthly_limit" {
  type        = number
  description = "Monthly budget limit in USD."
}

variable "workload" {
  type = string
}

variable "alert_emails" {
  type        = list(string)
  description = "Email addresses for budget alerts."
}

resource "aws_budgets_budget" "workload_budget" {
  name         = "${var.budget_name}-monthly"
  budget_type  = "COST"
  limit_amount = tostring(var.monthly_limit)
  limit_unit   = "USD"
  time_unit    = "MONTHLY"

  cost_filter {
    name   = "TagKeyValue"
    values = ["workload$${var.workload}"]
  }

  notification {
    comparison_operator        = "GREATER_THAN"
    threshold                  = 80
    threshold_type             = "PERCENTAGE"
    notification_type          = "ACTUAL"
    subscriber_email_addresses = var.alert_emails
  }

  notification {
    comparison_operator        = "GREATER_THAN"
    threshold                  = 100
    threshold_type             = "PERCENTAGE"
    notification_type          = "ACTUAL"
    subscriber_email_addresses = var.alert_emails
  }

  notification {
    comparison_operator        = "GREATER_THAN"
    threshold                  = 110
    threshold_type             = "PERCENTAGE"
    notification_type          = "FORECASTED"
    subscriber_email_addresses = var.alert_emails
  }
}

# shared/lifecycle-defaults.tf – Immer für alle Environments

# CloudWatch Log Groups MÜSSEN Retention haben
# Wird als Variable übergeben; Default je nach Environment

variable "log_retention_operational" {
  type        = number
  default     = 30
  description = "Retention in days for operational logs (Hot-Tier)."
  validation {
    condition     = var.log_retention_operational > 0
    error_message = "Log retention must be > 0 days. 0 means infinite (not allowed)."
  }
}

variable "log_retention_audit" {
  type        = number
  default     = 365
  description = "Retention in days for audit logs (Cold-Tier)."
}

# S3 Lifecycle-Policy-Modul (verpflichtend für alle Buckets)
module "s3_lifecycle" {
  source  = "../../modules/s3-lifecycle"
  bucket  = aws_s3_bucket.main.id

  transition_to_ia_days       = 30
  transition_to_glacier_days  = 90
  expiration_days             = 365  # Anpassen je nach Datenklasse
  delete_old_versions_after   = 30
}

# environments/production/main.tf

module "tags" {
  source = "../../modules/mandatory-tags"

  cost_center = "fintech-platform"
  owner       = "payments-team"
  environment = "production"
  workload    = "payment-service"
}

module "budget" {
  source = "../../modules/budget-alert"

  budget_name   = "payment-service-prod"
  monthly_limit = 5000  # USD/Monat
  workload      = "payment-service"
  alert_emails  = ["payments-team@company.com", "finops@company.com"]
}

resource "aws_instance" "app" {
  ami           = data.aws_ami.ubuntu.id
  instance_type = "t3.medium"

  tags = merge(module.tags.tags, {
    rightsizing-reviewed = "2025-03-01"
    capacity-commitment  = "on-demand"
  })
}

resource "aws_cloudwatch_log_group" "app" {
  name              = "/app/production/payment-service"
  retention_in_days = 30  # Hot-Tier: 30 Tage
  kms_key_id        = aws_kms_key.logging.arn
  tags              = module.tags.tags
}

resource "aws_s3_bucket" "data" {
  bucket = "acme-payment-data-prod"
  tags   = module.tags.tags
}

resource "aws_s3_bucket_lifecycle_configuration" "data" {
  bucket = aws_s3_bucket.data.id

  rule {
    id     = "default-tiering"
    status = "Enabled"

    transition {
      days          = 30
      storage_class = "STANDARD_IA"
    }

    transition {
      days          = 90
      storage_class = "GLACIER_IR"
    }

    noncurrent_version_expiration {
      noncurrent_days = 30
    }
  }
}