terraform-datadog-belgie/README.md

# Terraform Datadog Belgie (RDS Dashboard) Module

## Overview

This Terraform module creates an AWS RDS database monitoring dashboard in Datadog with comprehensive metrics for performance, storage, connections, and replication lag. Designed specifically for Belgian/EU deployments with pre-configured alert recipients and monitoring thresholds.

## Features

- **Comprehensive RDS Metrics**: CPU, memory, connections, storage, disk queue, latency
- **Dynamic Dashboard**: 25+ preconfigured widgets with automatic metric visualization
- **Flexible Alerting**: Configurable alert recipients for different severity levels
- **CloudWatch Integration**: Leverages AWS RDS CloudWatch metrics
- **Customizable Monitors**: Map-based monitor configuration for easy customization

## Resources Created

- `datadog_dashboard`: RDS Database Dashboard with 25+ widgets including:
  - Query value widgets for read/write latency and IOPS
  - Timeseries for replication lag, connections, CPU, memory, disk metrics
  - Toplist widgets for metric ranking
  - Note widgets for dashboard organization

## Dashboard Widgets

The dashboard includes comprehensive monitoring for:
- Read/Write Latency (query value widgets)
- Replication Lag (timeseries)
- Database Connections (timeseries)
- CPU Utilization (timeseries + toplist)
- Read/Write Operations (timeseries)
- Freeable Memory (timeseries + toplist)
- Disk Queue Depth (timeseries)
- Free Storage Space (timeseries + toplist)

## Requirements

| Name | Version |
|------|---------|
| terraform | >= 0.12 |
| datadog | >= 3.1.2 |
| aws | >= 2.0 |

## Usage

```hcl
module "rds_dashboard" {
  source = "./terraform-datadog-belgie"

  region         = "eu-west-1"
  api_key        = var.datadog_api_key
  app_key        = var.datadog_app_key
  datadog_site   = "https://api.datadoghq.eu/"
  aws_profile    = "production"
  cfa_slug       = "my-cfa"
  team           = "platform-team"
  application    = "myapp"
  stage          = "prd"
  
  alert_recipients   = ["team@example.com"]
  recipients         = ["team@example.com"]
  warning_recipients = ["team@example.com"]
  
  dd_rds_monitors = {
    cpu = {
      enabled   = true
      warning   = 75
      critical  = 90
      name      = "RDS CPU High"
    }
    # ... additional monitors
  }
}
```

## Inputs

| Name | Description | Type | Required | Default |
|------|-------------|------|----------|---------|
| `region` | AWS region for resources | `string` | yes | - |
| `api_key` | Datadog API key | `string` | yes | - |
| `app_key` | Datadog APP key | `string` | yes | - |
| `datadog_site` | Datadog site (EU or US) | `string` | no | `"https://api.datadoghq.eu/"` |
| `aws_profile` | AWS account this integration belongs to | `string` | yes | - |
| `cfa_slug` | CFA this integration belongs to | `string` | yes | - |
| `team` | Team this integration belongs to | `string` | yes | - |
| `application` | Application name | `string` | yes | - |
| `stage` | Stage (dev, tst, acc, prd) | `string` | yes | - |
| `alert_recipients` | Alert notification recipients | `list(string)` | no | `["patrick.de.ruiter@sanoma.com"]` |
| `recipients` | General notification recipients | `list(string)` | no | `["patrick.de.ruiter@sanoma.com"]` |
| `warning_recipients` | Warning notification recipients | `list(string)` | no | `["patrick.de.ruiter@sanoma.com"]` |
| `dd_rds_monitors` | RDS monitor configuration map | `map(any)` | no | See variables.tf |

## RDS Monitors Configuration

The `dd_rds_monitors` variable accepts a map with the following monitor types:
- `cpu`: CPU utilization monitoring
- `memory`: Freeable memory monitoring
- `connections`: Database connections monitoring
- `storage`: Free storage space monitoring
- `disk_queue`: Disk queue depth monitoring
- `read_latency`: Read latency monitoring
- `write_latency`: Write latency monitoring
- `replication_lag`: Replication lag monitoring

Each monitor can be configured with:
```hcl
{
  enabled   = bool    # Enable/disable the monitor
  warning   = number  # Warning threshold
  critical  = number  # Critical threshold
  name      = string  # Monitor name
}
```

## Outputs

Currently, this module does not export any outputs (outputs are commented out).

## Local Values

The module uses several local values for dynamic configuration:
- `dbidentifier`: Formatted as `{application}-{stage}`
- `rdsgraphs`: Map of 8 RDS metric queries
- `full_message`: Constructed alert message with links
- `tags`: Standard tags (team, stage, application)

## RDS Metrics Monitored

| Metric | CloudWatch Metric Name | Description |
|--------|----------------------|-------------|
| CPU | `aws.rds.cpuutilization` | CPU utilization percentage |
| Memory | `aws.rds.freeable_memory` | Freeable memory in bytes |
| Connections | `aws.rds.database_connections` | Number of database connections |
| Storage | `aws.rds.free_storage_space` | Free storage space in bytes |
| Read Latency | `aws.rds.read_latency` | Read operation latency |
| Write Latency | `aws.rds.write_latency` | Write operation latency |
| Disk Queue | `aws.rds.disk_queue_depth` | Disk queue depth |
| Replication Lag | `aws.rds.replica_lag` | Replication lag in seconds |

## Dashboard Layout

- **Layout Type**: Free layout (allows custom positioning)
- **Read-Only**: No (allows modifications in UI)
- **Widget Organization**: Sections with note widgets as headers
- **Conditional Formatting**: Metric values color-coded based on thresholds

## Tagging Strategy

All resources are tagged with:
- `team:{team}`
- `stage:{stage}`
- `application:{application}`

## Notes

- Default Datadog site is EU (GDPR compliance)
- Dashboard uses free layout for flexible widget positioning
- Metrics are sourced from AWS RDS CloudWatch integration
- Database identifier is auto-generated from application and stage variables
- Alert messages include links to related monitors

## License

Internal use only - Sanoma/WeBuildYourCloud

## Authors

Created and maintained by the Platform Engineering team.