Skip to main content

Documentation Index

Fetch the complete documentation index at: https://mintlify.com/hashicorp/terraform/llms.txt

Use this file to discover all available pages before exploring further.

​
Provider Configuration Basics

Providers are configured using provider blocks in your Terraform configuration. Each provider has its own configuration schema that defines the arguments it accepts.

​
Declaring Provider Requirements

Before using a provider, declare it in a required_providers block: 
terraform {
  required_providers {
    aws = {
      source  = "hashicorp/aws"
      version = "~> 5.0"
    }
    azurerm = {
      source  = "hashicorp/azurerm"
      version = ">= 3.0"
    }
  }
}

​
Provider Source Addresses

The source attribute uses the format: [hostname/]namespace/type
  • hostname (optional): Registry hostname (defaults to registry.terraform.io)
  • namespace: Organizational namespace (e.g., hashicorp, aws)
  • type: Provider type name (e.g., aws, azurerm)
Examples:
  • hashicorp/aws → registry.terraform.io/hashicorp/aws
  • aws → legacy provider reference (not recommended)
  • example.com/myorg/custom → custom registry provider

​
Version Constraints

Version constraints use semantic versioning operators:
  • = 1.0.0 - Exact version
  • != 1.0.0 - Exclude specific version
  • > 1.0.0 - Greater than
  • >= 1.0.0 - Greater than or equal to
  • < 2.0.0 - Less than
  • <= 2.0.0 - Less than or equal to
  • ~> 1.2.0 - Pessimistic constraint (>= 1.2.0, < 1.3.0)
Multiple constraints can be combined: 
version = ">= 1.0.0, < 2.0.0"

​
Configuring Providers

After declaring requirements, configure the provider with a provider block: 
provider "aws" {
  region     = "us-west-2"
  access_key = var.aws_access_key
  secret_key = var.aws_secret_key
  
  default_tags {
    tags = {
      Environment = "production"
      Terraform   = "true"
    }
  }
}

​
Provider Configuration Validation

When you configure a provider, Terraform performs validation in multiple stages:

​
1. Schema Validation

Terraform validates the configuration against the provider’s schema from GetProviderSchema. This checks:
  • Required attributes are present
  • Attribute types match the schema
  • Unknown attributes are flagged as errors

​
2. Provider-Specific Validation

Providers can implement custom validation logic in the ValidateProviderConfig method (internal/providers/provider.go:30): 
type ValidateProviderConfigRequest struct {
    Config cty.Value // Raw configuration value
}

type ValidateProviderConfigResponse struct {
    PreparedConfig cty.Value // Unused in protocol v5
    Diagnostics    tfdiags.Diagnostics
}
Providers can return warnings or errors based on business logic, such as:
  • Checking for conflicting configuration options
  • Validating credential formats
  • Verifying API endpoint reachability

​
3. Configuration Application

During initialization, Terraform calls ConfigureProvider (internal/providers/provider.go:437): 
type ConfigureProviderRequest struct {
    TerraformVersion   string
    Config             cty.Value
    ClientCapabilities ClientCapabilities
}

type ConfigureProviderResponse struct {
    Diagnostics tfdiags.Diagnostics
}
At this stage, the provider:
  • Establishes connections to remote APIs
  • Authenticates with credentials
  • Initializes internal clients and state
  • Returns diagnostics if configuration is invalid

​
Multiple Provider Configurations

You can define multiple configurations for the same provider using aliases: 
provider "aws" {
  region = "us-west-2"
}

provider "aws" {
  alias  = "east"
  region = "us-east-1"
}
Reference aliased providers in resources: 
resource "aws_instance" "west" {
  # Uses the default provider configuration
  ami           = "ami-12345678"
  instance_type = "t2.micro"
}

resource "aws_instance" "east" {
  provider      = aws.east
  ami           = "ami-87654321"
  instance_type = "t2.micro"
}

​
Provider Configuration Inheritance

In module hierarchies, provider configurations are inherited:

​
Root Module

provider "aws" {
  region = "us-west-2"
}

module "network" {
  source = "./modules/network"
  # Inherits the aws provider configuration
}

​
Explicit Provider Passing

You can explicitly pass provider configurations to modules: 
module "network" {
  source = "./modules/network"
  
  providers = {
    aws = aws.east
  }
}

​
Environment Variables

Many providers support environment-based configuration: 
export AWS_REGION=us-west-2
export AWS_ACCESS_KEY_ID=AKIAIOSFODNN7EXAMPLE
export AWS_SECRET_ACCESS_KEY=wJalrXUtnFEMI/K7MDENG/bPxRfiCYEXAMPLEKEY
Provider configuration attributes typically override environment variables: 
provider "aws" {
  # This takes precedence over AWS_REGION
  region = "us-east-1"
}

​
Provider Meta-Arguments

Some meta-arguments can be used in provider blocks:

​
alias

Create multiple configurations for the same provider: 
provider "aws" {
  alias  = "backup"
  region = "us-east-2"
}

​
version (deprecated)

Older syntax for version constraints (use required_providers instead): 
# Deprecated - use required_providers
provider "aws" {
  version = "~> 5.0"
}

​
Sensitive Values

Provider configurations often include sensitive values. Use variables with sensitive flag: 
variable "database_password" {
  type      = string
  sensitive = true
}

provider "mysql" {
  endpoint = "example.com:3306"
  username = "admin"
  password = var.database_password
}
Terraform masks sensitive values in output and logs.

​
Provider Configuration Context

Providers receive additional context during configuration (internal/providers/provider.go:437):
  • TerraformVersion: The version of Terraform executing the provider
  • ClientCapabilities: Features supported by the Terraform client
Providers can use this information to:
  • Ensure compatibility with the Terraform version
  • Enable/disable features based on client capabilities
  • Provide version-specific warnings or errors

​
Configuration Lifecycle

The provider configuration lifecycle follows these steps:
  1. Parse: Terraform parses the HCL configuration
  2. Schema Lookup: Retrieves provider schema via GetProviderSchema
  3. Decode: Decodes configuration against the schema
  4. Validate: Calls ValidateProviderConfig for custom validation
  5. Configure: Calls ConfigureProvider to initialize the provider
  6. Ready: Provider is ready for resource operations
If any step fails, Terraform reports diagnostics and halts execution.

​
Best Practices

​
Use Variables for Credentials

Never hardcode credentials: 
# Bad
provider "aws" {
  access_key = "AKIAIOSFODNN7EXAMPLE"
  secret_key = "wJalrXUtnFEMI/K7MDENG/bPxRfiCYEXAMPLEKEY"
}

# Good
provider "aws" {
  access_key = var.aws_access_key
  secret_key = var.aws_secret_key
}

​
Pin Provider Versions

Use version constraints to ensure reproducible deployments: 
terraform {
  required_providers {
    aws = {
      source  = "hashicorp/aws"
      version = "~> 5.0.0" # Pin to minor version
    }
  }
}

​
Use Required Providers Block

Always declare providers in required_providers for explicit dependency management: 
# Recommended
terraform {
  required_providers {
    aws = {
      source  = "hashicorp/aws"
      version = "~> 5.0"
    }
  }
}

# Not recommended - implicit provider
resource "aws_instance" "example" {
  # Terraform infers the provider
}

​
Next Steps

Build docs developers (and LLMs) love

Get started for freeTalk to us