2.2.1. Provider Configuration and Aliases

💡 First Principle: The provider block configures how to talk to a platform (region, credentials, endpoints), and an alias lets you keep several such configurations side by side so each resource can point at the one it needs.

A bare provider block is the default configuration for that provider. Adding alias = "..." creates an additional named configuration:

provider "aws" {
  region = "us-east-1"           # default configuration
}

provider "aws" {
  alias  = "west"
  region = "us-west-2"           # additional configuration
}

resource "aws_instance" "east" {
  # uses the default aws provider
}

resource "aws_instance" "west" {
  provider = aws.west            # explicitly selects the aliased one
}

Resources use the default provider unless they set the provider meta-argument to an aliased reference like aws.west. This is the standard pattern for deploying the same resources across regions or accounts from one configuration. (Some providers add their own shortcut — AWS provider 6.0+ accepts a per-resource region argument — but aliases are the provider-agnostic mechanism, and what you still need when resources must use different credentials or accounts.)

⚠️ Exam Trap: Don't confuse the provider block (configures a provider instance) with the provider meta-argument (selects which configured instance a resource uses). The block defines the line; the meta-argument dials it.

Reflection Question: Why does deploying to two accounts with different credentials require two provider configurations, and how does each resource choose between them?