aws_dynamodb_table

Provides a DynamoDB table resource

Note: It is recommended to use lifecycle ignore_changes for read_capacity and/or write_capacity if there's autoscaling policy attached to the table.

Example Usage

The following dynamodb table description models the table and GSI shown in the AWS SDK example documentation

resource "aws_dynamodb_table" "basic-dynamodb-table" {
  name           = "GameScores"
  read_capacity  = 20
  write_capacity = 20
  hash_key       = "UserId"
  range_key      = "GameTitle"

  attribute {
    name = "UserId"
    type = "S"
  }

  attribute {
    name = "GameTitle"
    type = "S"
  }

  attribute {
    name = "TopScore"
    type = "N"
  }

  ttl {
    attribute_name = "TimeToExist"
    enabled = false
  }

  global_secondary_index {
    name               = "GameTitleIndex"
    hash_key           = "GameTitle"
    range_key          = "TopScore"
    write_capacity     = 10
    read_capacity      = 10
    projection_type    = "INCLUDE"
    non_key_attributes = ["UserId"]
  }

  tags {
    Name        = "dynamodb-table-1"
    Environment = "production"
  }
}

Notes: attribute can be lists

  attribute = [{
    name = "UserId"
    type = "S"
  }, {
    name = "GameTitle"
    type = "S"
  }, {
    name = "TopScore"
    type = "N"
  }]

Argument Reference

The following arguments are supported:

• name - (Required) The name of the table, this needs to be unique within a region.
• hash_key - (Required, Forces new resource) The attribute to use as the hash (partition) key. Must also be defined as an attribute, see below.
• range_key - (Optional, Forces new resource) The attribute to use as the range (sort) key. Must also be defined as an attribute, see below.
• write_capacity - (Required) The number of write units for this table
• read_capacity - (Required) The number of read units for this table
• attribute - (Required) List of nested attribute definitions. Only required for hash_key and range_key attributes. Each attribute has two properties:
  • name - (Required) The name of the attribute
  • type - (Required) Attribute type, which must be a scalar type: S, N, or B for (S)tring, (N)umber or (B)inary data
• ttl - (Optional) Defines ttl, has two properties, and can only be specified once:
  • enabled - (Required) Indicates whether ttl is enabled (true) or disabled (false).
  • attribute_name - (Required) The name of the table attribute to store the TTL timestamp in.
• local_secondary_index - (Optional, Forces new resource) Describe an LSI on the table; these can only be allocated at creation so you cannot change this definition after you have created the resource.
• global_secondary_index - (Optional) Describe a GSO for the table; subject to the normal limits on the number of GSIs, projected attributes, etc.
• stream_enabled - (Optional) Indicates whether Streams are to be enabled (true) or disabled (false).
• stream_view_type - (Optional) When an item in the table is modified, StreamViewType determines what information is written to the table's stream. Valid values are KEYS_ONLY, NEW_IMAGE, OLD_IMAGE, NEW_AND_OLD_IMAGES.
• server_side_encryption - (Optional) Encrypt at rest options.
• tags - (Optional) A map of tags to populate on the created table.
• point_in_time_recovery - (Optional) Point-in-time recovery options.

Timeouts

The timeouts block allows you to specify timeouts for certain actions:

• create - (Defaults to 10 mins) Used when creating the table
• update - (Defaults to 10 mins) Used when updating the table
• delete - (Defaults to 10 mins) Used when deleting the table

Nested fields

local_secondary_index

• name - (Required) The name of the index
• range_key - (Required) The name of the range key; must be defined
• projection_type - (Required) One of ALL, INCLUDE or KEYS_ONLY where ALL projects every attribute into the index, KEYS_ONLY projects just the hash and range key into the index, and INCLUDE projects only the keys specified in the non_key_attributes parameter.
• non_key_attributes - (Optional) Only required with INCLUDE as a projection type; a list of attributes to project into the index. These do not need to be defined as attributes on the table.

global_secondary_index

• name - (Required) The name of the index
• write_capacity - (Required) The number of write units for this index
• read_capacity - (Required) The number of read units for this index
• hash_key - (Required) The name of the hash key in the index; must be defined as an attribute in the resource.
• range_key - (Optional) The name of the range key; must be defined
• projection_type - (Required) One of ALL, INCLUDE or KEYS_ONLY where ALL projects every attribute into the index, KEYS_ONLY projects just the hash and range key into the index, and INCLUDE projects only the keys specified in the non_key_attributes parameter.
• non_key_attributes - (Optional) Only required with INCLUDE as a projection type; a list of attributes to project into the index. These do not need to be defined as attributes on the table.

server_side_encryption

• enabled - (Required) Whether to enable encryption at rest. If the server_side_encryption block is not provided then this defaults to false.

point_in_time_recovery

• enabled - (Required) Whether to enable point-in-time recovery - note that it can take up to 10 minutes to enable for new tables. If the point_in_time_recovery block is not provided then this defaults to false.

A note about attributes

Only define attributes on the table object that are going to be used as:

• Table hash key or range key
• LSI or GSI hash key or range key

The DynamoDB API expects attribute structure (name and type) to be passed along when creating or updating GSI/LSIs or creating the initial table. In these cases it expects the Hash / Range keys to be provided; because these get re-used in numerous places (i.e the table's range key could be a part of one or more GSIs), they are stored on the table object to prevent duplication and increase consistency. If you add attributes here that are not used in these scenarios it can cause an infinite loop in planning.

Attributes Reference

In addition to all arguments above, the following attributes are exported:

• arn - The arn of the table
• id - The name of the table
• stream_arn - The ARN of the Table Stream. Only available when stream_enabled = true
• stream_label - A timestamp, in ISO 8601 format, for this stream. Note that this timestamp is not a unique identifier for the stream on its own. However, the combination of AWS customer ID, table name and this field is guaranteed to be unique. It can be used for creating CloudWatch Alarms. Only available when stream_enabled = true

Import

DynamoDB tables can be imported using the name, e.g.

$ terraform import aws_dynamodb_table.basic-dynamodb-table GameScores