S3

Stores the state as a given key in a given bucket on Amazon S3. This backend also supports state locking and consistency checking via Dynamo DB, which can be enabled by setting the dynamodb_table field to an existing DynamoDB table name. A single DynamoDB table can be used to lock multiple remote state files. Terraform generates key names that include the values of the bucket and key variables.

Example Configuration

terraform {
  backend "s3" {
    bucket = "mybucket"
    key    = "path/to/my/key"
    region = "us-east-1"
  }
}

This assumes we have a bucket created called mybucket. The Terraform state is written to the key path/to/my/key.

Note that for the access credentials we recommend using a partial configuration.

S3 Bucket Permissions

Terraform will need the following AWS IAM permissions on the target backend bucket:

• s3:ListBucket on arn:aws:s3:::mybucket
• s3:GetObject on arn:aws:s3:::mybucket/path/to/my/key
• s3:PutObject on arn:aws:s3:::mybucket/path/to/my/key
• s3:DeleteObject on arn:aws:s3:::mybucket/path/to/my/key

This is seen in the following AWS IAM Statement:

{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Effect": "Allow",
      "Action": "s3:ListBucket",
      "Resource": "arn:aws:s3:::mybucket"
    },
    {
      "Effect": "Allow",
      "Action": ["s3:GetObject", "s3:PutObject", "s3:DeleteObject"],
      "Resource": "arn:aws:s3:::mybucket/path/to/my/key"
    }
  ]
}

DynamoDB Table Permissions

If you are using state locking, Terraform will need the following AWS IAM permissions on the DynamoDB table (arn:aws:dynamodb:::table/mytable):

• dynamodb:GetItem
• dynamodb:PutItem
• dynamodb:DeleteItem

This is seen in the following AWS IAM Statement:

{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Effect": "Allow",
      "Action": [
        "dynamodb:GetItem",
        "dynamodb:PutItem",
        "dynamodb:DeleteItem"
      ],
      "Resource": "arn:aws:dynamodb:*:*:table/mytable"
    }
  ]
}

Data Source Configuration

To make use of the S3 remote state in another configuration, use the terraform_remote_state data source.

data "terraform_remote_state" "network" {
  backend = "s3"
  config = {
    bucket = "terraform-state-prod"
    key    = "network/terraform.tfstate"
    region = "us-east-1"
  }
}

The terraform_remote_state data source will return all of the root module outputs defined in the referenced remote state (but not any outputs from nested modules unless they are explicitly output again in the root). An example output might look like:

data.terraform_remote_state.network:
  id = 2016-10-29 01:57:59.780010914 +0000 UTC
  addresses.# = 2
  addresses.0 = 52.207.220.222
  addresses.1 = 54.196.78.166
  backend = s3
  config.% = 3
  config.bucket = terraform-state-prod
  config.key = network/terraform.tfstate
  config.region = us-east-1
  elb_address = web-elb-790251200.us-east-1.elb.amazonaws.com
  public_subnet_id = subnet-1e05dd33

Configuration

This backend requires the configuration of the AWS Region and S3 state storage. Other configuration, such as enabling DynamoDB state locking, is optional.

Credentials and Shared Configuration

The following configuration is required:

• region - (Required) AWS Region of the S3 Bucket and DynamoDB Table (if used). This can also be sourced from the AWS_DEFAULT_REGION and AWS_REGION environment variables.

The following configuration is optional:

• access_key - (Optional) AWS access key. If configured, must also configure secret_key. This can also be sourced from the AWS_ACCESS_KEY_ID environment variable, AWS shared credentials file (e.g. ~/.aws/credentials), or AWS shared configuration file (e.g. ~/.aws/config).
• secret_key - (Optional) AWS access key. If configured, must also configure access_key. This can also be sourced from the AWS_SECRET_ACCESS_KEY environment variable, AWS shared credentials file (e.g. ~/.aws/credentials), or AWS shared configuration file (e.g. ~/.aws/config).
• iam_endpoint - (Optional) Custom endpoint for the AWS Identity and Access Management (IAM) API. This can also be sourced from the AWS_IAM_ENDPOINT environment variable.
• max_retries - (Optional) The maximum number of times an AWS API request is retried on retryable failure. Defaults to 5.
• profile - (Optional) Name of AWS profile in AWS shared credentials file (e.g. ~/.aws/credentials) or AWS shared configuration file (e.g. ~/.aws/config) to use for credentials and/or configuration. This can also be sourced from the AWS_PROFILE environment variable.
• shared_credentials_file - (Optional) Path to the AWS shared credentials file. Defaults to ~/.aws/credentials.
• skip_credentials_validation - (Optional) Skip credentials validation via the STS API.
• skip_region_validation - (Optional) Skip validation of provided region name.
• skip_metadata_api_check - (Optional) Skip usage of EC2 Metadata API.
• sts_endpoint - (Optional) Custom endpoint for the AWS Security Token Service (STS) API. This can also be sourced from the AWS_STS_ENDPOINT environment variable.
• token - (Optional) Multi-Factor Authentication (MFA) token. This can also be sourced from the AWS_SESSION_TOKEN environment variable.

Assume Role Configuration

The following configuration is optional:

• assume_role_duration_seconds - (Optional) Number of seconds to restrict the assume role session duration.
• assume_role_policy - (Optional) IAM Policy JSON describing further restricting permissions for the IAM Role being assumed.
• assume_role_policy_arns - (Optional) Set of Amazon Resource Names (ARNs) of IAM Policies describing further restricting permissions for the IAM Role being assumed.
• assume_role_tags - (Optional) Map of assume role session tags.
• assume_role_transitive_tag_keys - (Optional) Set of assume role session tag keys to pass to any subsequent sessions.
• external_id - (Optional) External identifier to use when assuming the role.
• role_arn - (Optional) Amazon Resource Name (ARN) of the IAM Role to assume.
• session_name - (Optional) Session name to use when assuming the role.

S3 State Storage

The following configuration is required:

• bucket - (Required) Name of the S3 Bucket.
• key - (Required) Path to the state file inside the S3 Bucket. When using a non-default workspace, the state path will be /workspace_key_prefix/workspace_name/key (see also the workspace_key_prefix configuration).

The following configuration is optional:

• acl - (Optional) Canned ACL to be applied to the state file.
• encrypt - (Optional) Enable server side encryption of the state file.
• endpoint - (Optional) Custom endpoint for the AWS S3 API. This can also be sourced from the AWS_S3_ENDPOINT environment variable.
• force_path_style - (Optional) Enable path-style S3 URLs (https://<HOST>/<BUCKET> instead of https://<BUCKET>.<HOST>).
• kms_key_id - (Optional) Amazon Resource Name (ARN) of a Key Management Service (KMS) Key to use for encrypting the state. Note that if this value is specified, Terraform will need kms:Encrypt, kms:Decrypt and kms:GenerateDataKey permissions on this KMS key.
• sse_customer_key - (Optional) The key to use for encrypting state with Server-Side Encryption with Customer-Provided Keys (SSE-C). This is the base64-encoded value of the key, which must decode to 256 bits. This can also be sourced from the AWS_SSE_CUSTOMER_KEY environment variable, which is recommended due to the sensitivity of the value. Setting it inside a terraform file will cause it to be persisted to disk in terraform.tfstate.
• workspace_key_prefix - (Optional) Prefix applied to the state path inside the bucket. This is only relevant when using a non-default workspace. Defaults to env:.

DynamoDB State Locking

The following configuration is optional:

• dynamodb_endpoint - (Optional) Custom endpoint for the AWS DynamoDB API. This can also be sourced from the AWS_DYNAMODB_ENDPOINT environment variable.
• dynamodb_table - (Optional) Name of DynamoDB Table to use for state locking and consistency. The table must have a partition key named LockID with type of String. If not configured, state locking will be disabled.

Multi-account AWS Architecture

A common architectural pattern is for an organization to use a number of separate AWS accounts to isolate different teams and environments. For example, a "staging" system will often be deployed into a separate AWS account than its corresponding "production" system, to minimize the risk of the staging environment affecting production infrastructure, whether via rate limiting, misconfigured access controls, or other unintended interactions.

The S3 backend can be used in a number of different ways that make different tradeoffs between convenience, security, and isolation in such an organization. This section describes one such approach that aims to find a good compromise between these tradeoffs, allowing use of Terraform's workspaces feature to switch conveniently between multiple isolated deployments of the same configuration.

Use this section as a starting-point for your approach, but note that you will probably need to make adjustments for the unique standards and regulations that apply to your organization. You will also need to make some adjustments to this approach to account for existing practices within your organization, if for example other tools have previously been used to manage infrastructure.

Terraform is an administrative tool that manages your infrastructure, and so ideally the infrastructure that is used by Terraform should exist outside of the infrastructure that Terraform manages. This can be achieved by creating a separate administrative AWS account which contains the user accounts used by human operators and any infrastructure and tools used to manage the other accounts. Isolating shared administrative tools from your main environments has a number of advantages, such as avoiding accidentally damaging the administrative infrastructure while changing the target infrastructure, and reducing the risk that an attacker might abuse production infrastructure to gain access to the (usually more privileged) administrative infrastructure.

Administrative Account Setup

Your administrative AWS account will contain at least the following items:

• One or more IAM user for system administrators that will log in to maintain infrastructure in the other accounts.
• Optionally, one or more IAM groups to differentiate between different groups of users that have different levels of access to the other AWS accounts.
• An S3 bucket that will contain the Terraform state files for each workspace.
• A DynamoDB table that will be used for locking to prevent concurrent operations on a single workspace.

Provide the S3 bucket name and DynamoDB table name to Terraform within the S3 backend configuration using the bucket and dynamodb_table arguments respectively, and configure a suitable workspace_key_prefix to contain the states of the various workspaces that will subsequently be created for this configuration.

Environment Account Setup

For the sake of this section, the term "environment account" refers to one of the accounts whose contents are managed by Terraform, separate from the administrative account described above.

Your environment accounts will eventually contain your own product-specific infrastructure. Along with this it must contain one or more IAM roles that grant sufficient access for Terraform to perform the desired management tasks.

Delegating Access

Each Administrator will run Terraform using credentials for their IAM user in the administrative account. IAM Role Delegation is used to grant these users access to the roles created in each environment account.

Full details on role delegation are covered in the AWS documentation linked above. The most important details are:

• Each role's Assume Role Policy must grant access to the administrative AWS account, which creates a trust relationship with the administrative AWS account so that its users may assume the role.
• The users or groups within the administrative account must also have a policy that creates the converse relationship, allowing these users or groups to assume that role.

Since the purpose of the administrative account is only to host tools for managing other accounts, it is useful to give the administrative accounts restricted access only to the specific operations needed to assume the environment account role and access the Terraform state. By blocking all other access, you remove the risk that user error will lead to staging or production resources being created in the administrative account by mistake.

When configuring Terraform, use either environment variables or the standard credentials file ~/.aws/credentials to provide the administrator user's IAM credentials within the administrative account to both the S3 backend and to Terraform's AWS provider.

Use conditional configuration to pass a different assume_role value to the AWS provider depending on the selected workspace. For example:

variable "workspace_iam_roles" {
  default = {
    staging    = "arn:aws:iam::STAGING-ACCOUNT-ID:role/Terraform"
    production = "arn:aws:iam::PRODUCTION-ACCOUNT-ID:role/Terraform"
  }
}

provider "aws" {
  # No credentials explicitly set here because they come from either the
  # environment or the global credentials file.

  assume_role = {
    role_arn = "${var.workspace_iam_roles[terraform.workspace]}"
  }
}

If workspace IAM roles are centrally managed and shared across many separate Terraform configurations, the role ARNs could also be obtained via a data source such as terraform_remote_state to avoid repeating these values.

Creating and Selecting Workspaces

With the necessary objects created and the backend configured, run terraform init to initialize the backend and establish an initial workspace called "default". This workspace will not be used, but is created automatically by Terraform as a convenience for users who are not using the workspaces feature.

Create a workspace corresponding to each key given in the workspace_iam_roles variable value above:

$ terraform workspace new staging
Created and switched to workspace "staging"!
 
...
 
$ terraform workspace new production
Created and switched to workspace "production"!
 
...

Due to the assume_role setting in the AWS provider configuration, any management operations for AWS resources will be performed via the configured role in the appropriate environment AWS account. The backend operations, such as reading and writing the state from S3, will be performed directly as the administrator's own user within the administrative account.

$ terraform workspace select staging
$ terraform apply
...

Running Terraform in Amazon EC2

Teams that make extensive use of Terraform for infrastructure management often run Terraform in automation to ensure a consistent operating environment and to limit access to the various secrets and other sensitive information that Terraform configurations tend to require.

When running Terraform in an automation tool running on an Amazon EC2 instance, consider running this instance in the administrative account and using an instance profile in place of the various administrator IAM users suggested above. An IAM instance profile can also be granted cross-account delegation access via an IAM policy, giving this instance the access it needs to run Terraform.

To isolate access to different environment accounts, use a separate EC2 instance for each target account so that its access can be limited only to the single account.

Similar approaches can be taken with equivalent features in other AWS compute services, such as ECS.

Protecting Access to Workspace State

In a simple implementation of the pattern described in the prior sections, all users have access to read and write states for all workspaces. In many cases it is desirable to apply more precise access constraints to the Terraform state objects in S3, so that for example only trusted administrators are allowed to modify the production state, or to control reading of a state that contains sensitive information.

Amazon S3 supports fine-grained access control on a per-object-path basis using IAM policy. A full description of S3's access control mechanism is beyond the scope of this guide, but an example IAM policy granting access to only a single state object within an S3 bucket is shown below:

{
  "Version": "2012-10-17",
  "Statement": [
    {
      "Effect": "Allow",
      "Action": "s3:ListBucket",
      "Resource": "arn:aws:s3:::myorg-terraform-states"
    },
    {
      "Effect": "Allow",
      "Action": ["s3:GetObject", "s3:PutObject"],
      "Resource": "arn:aws:s3:::myorg-terraform-states/myapp/production/tfstate"
    }
  ]
}

It is also possible to apply fine-grained access control to the DynamoDB table used for locking. When Terraform puts the state lock in place during terraform plan, it stores the full state file as a document and sets the s3 object key as the partition key for the document. After the state lock is released, Terraform places a digest of the updated state file in DynamoDB. The key is similar to the one for the original state file, but is suffixed with -md5.

The example below shows a simple IAM policy that allows the backend operations role to perform these operations:

{
  "Version": "2012-10-17",
  "Statement": [
      {
        "Effect" : "Allow",
        "Action" : [
          "dynamodb:DeleteItem",
          "dynamodb:GetItem",
          "dynamodb:PutItem",
          "dynamodb:Query",
          "dynamodb:UpdateItem"
        ],
        "Resource" : ["arn:aws:dynamodb:*:*:table/myorg-state-lock-table"],
        "Condition" : {
          "ForAllValues:StringEquals" : {
            "dynamodb:LeadingKeys" : [
              "myorg-terraform-states/myapp/production/tfstate", // during a state lock the full state file is stored with this key
              "myorg-terraform-states/myapp/production/tfstate-md5" // after the lock is released a hash of the statefile's contents are stored with this key
            ]
          }
        }
      }
  ]
}

Refer to the AWS documentation on DynamoDB fine-grained locking for more details.

Configuring Custom User-Agent Information

Note this feature is optional and only available in Terraform v0.13.1+.

By default, the underlying AWS client used by the Terraform AWS Provider creates requests with User-Agent headers including information about Terraform and AWS Go SDK versions. To provide additional information in the User-Agent headers, the TF_APPEND_USER_AGENT environment variable can be set and its value will be directly added to HTTP requests. e.g.

$ export TF_APPEND_USER_AGENT="JenkinsAgent/i-12345678 BuildID/1234 (Optional Extra Information)"