Skip to main content
Service Catalog Version 0.85.1Last updated in version 0.85.0

Amazon Elasticsearch Service

View SourceRelease Notes

Overview​

This service contains code to deploy an Amazon Elasticsearch Service cluster. See the Amazon Elasticsearch Service documentation and the Getting Started page for more information.

Features​

  • A fully-managed native Elasticsearch cluster in a VPC
  • A fully functional Kibana UI
  • VPC-based security
  • Zone awareness, i.e., deployment of Elasticsearch nodes across Availability Zones
  • Automatic nightly snapshots
  • CloudWatch Alarms for alerting when CPU, memory, and disk metrics exceed certain thresholds

Learn​

note

This repo is a part of the Gruntwork Service Catalog, a collection of reusable, battle-tested, production ready infrastructure code. If you’ve never used the Service Catalog before, make sure to read How to use the Gruntwork Service Catalog!

Deploy​

Non-production deployment (quick start for learning)​

If you just want to try this repo out for experimenting and learning, check out the following resources:

  • examples/for-learning-and-testing folder: The examples/for-learning-and-testing folder contains standalone sample code optimized for learning, experimenting, and testing (but not direct production usage).

  • AWS Free tier: Using Amazon ES on Amazon’s free tier is a great way to get started, but it has limited features and does not include encryption at rest, ultra warm data notes, or advanced security options such as fine-grained access control. The free tier does allow multiple availability zones, VPC-based access control, TLS-only requests, and node-to-node encryption.

Production deployment​

If you want to deploy this repo in production, check out the following resources:

Reference​

Required​

domain_namestring(required)

The name of the Elasticsearch cluster. It must be unique to your account and region, start with a lowercase letter, contain between 3 and 28 characters, and contain only lowercase letters a-z, the numbers 0-9, and the hyphen (-).

instance_countnumber(required)

The number of instances to deploy in the Elasticsearch cluster. This must be an even number if zone_awareness_enabled is true.

instance_typestring(required)

The instance type to use for Elasticsearch data nodes (e.g., t2.small.elasticsearch, or m4.large.elasticsearch). For supported instance types see https://docs.aws.amazon.com/elasticsearch-service/latest/developerguide/aes-supported-instance-types.html.

volume_sizenumber(required)

The size in GiB of the EBS volume for each node in the cluster (e.g. 10, or 512). For volume size limits see https://docs.aws.amazon.com/elasticsearch-service/latest/developerguide/aes-limits.html.

volume_typestring(required)

The type of EBS volumes to use in the cluster. Must be one of: standard, gp2, io1, sc1, or st1. For a comparison of EBS volume types, see https://docs.aws.amazon.com/AWSEC2/latest/WindowsGuide/ebs-volume-types.html.

Whether to deploy the Elasticsearch nodes across two Availability Zones instead of one. Note that if you enable this, the instance_count MUST be an even number.

Optional​

advanced_optionsmap(optional)

Key-value string pairs to specify advanced configuration options. Note that the values for these configuration options must be strings (wrapped in quotes).

map(any)
{}

Enable fine grain access control

false
alarm_sns_topic_arnslist(optional)

ARNs of the SNS topics associated with the CloudWatch alarms for the Elasticsearch cluster.

list(string)
[]

The list of network CIDR blocks to allow network access to Aurora from. One of allow_connections_from_cidr_blocks or allow_connections_from_security_groups must be specified for the database to be reachable.

[]

The list of IDs or Security Groups to allow network access to Aurora from. All security groups must either be in the VPC specified by vpc_id, or a peered VPC with the VPC specified by vpc_id. One of allow_connections_from_cidr_blocks or allow_connections_from_security_groups must be specified for the database to be reachable.

[]

Hour during which the service takes an automated daily snapshot of the indices in the domain. This setting has no effect on Elasticsearch 5.3 and later.

0
availability_zone_countnumber(optional)

Number of Availability Zones for the domain to use with zone_awareness_enabled. Defaults to 2. Valid values: 2 or 3.

2

Whether or not the Service Linked Role for Elasticsearch should be created within this module. Normally the service linked role is created automatically by AWS when creating the Elasticsearch domain in the web console, but API does not implement this logic. You can either have AWS automatically manage this by creating a domain manually in the console, or manage it in terraform using the landing zone modules or this variable.

false
custom_endpointstring(optional)

Fully qualified domain for your custom endpoint.

null

ACM certificate ARN for your custom endpoint.

null

Whether to enable custom endpoint for the Elasticsearch domain.

false
custom_tagsmap(optional)

A map of custom tags to apply to the ElasticSearch Domain. The key is the tag name and the value is the tag value.

map(string)
{}
dedicated_master_countnumber(optional)

The number of dedicated master nodes to run. We recommend setting this to 3 for production deployments. Only used if dedicated_master_enabled is true.

null

Whether to deploy separate nodes specifically for performing cluster management tasks (e.g. tracking number of nodes, monitoring health, replicating changes). This increases the stability of large clusters and is required for clusters with more than 10 nodes.

false
dedicated_master_typestring(optional)

The instance type for the dedicated master nodes. These nodes can use a different instance type than the rest of the cluster. Only used if dedicated_master_enabled is true.

null
ebs_enabledbool(optional)

Set to false to disable EBS volumes. This is useful for nodes that have optimized instance storage, like hosts running the i3 instance type.

true
elasticsearch_versionstring(optional)

The version of Elasticsearch to deploy.

7.7

Set to true to enable several basic CloudWatch alarms around CPU usage, memory usage, and disk space usage. If set to true, make sure to specify SNS topics to send notifications to using alarms_sns_topic_arns.

true

False by default because encryption at rest is not included in the free tier. When true, the Elasticsearch domain storage will be encrypted at rest using the KMS key described with encryption_kms_key_id. We strongly recommend configuring a custom KMS key instead of using the shared service key for a better security posture when configuring encryption at rest.

true

Whether to enable node-to-node encryption.

true
encryption_kms_key_idstring(optional)

The ID of the KMS key to use to encrypt the Elasticsearch domain storage. Only used if enable_encryption_at_rest. When null, uses the aws/es service KMS key.

null
iam_principal_arnslist(optional)

The ARNS of the IAM users and roles to which to allow full access to the Elasticsearch cluster. Setting this to a restricted list is useful when using a public access cluster.

list(string)
[ '*' ]

Whether the internal user database is enabled. Enable this to use master accounts. Only used if advanced_security_options is set to true.

false
iopsnumber(optional)

The baseline input/output (I/O) performance of EBS volumes attached to data nodes. Must be between 1000 and 4000. Applicable only if volume_type is io1.

null
is_publicbool(optional)

Whether the cluster is publicly accessible.

false
master_user_arnstring(optional)

ARN of the master user. Only used if advanced_security_options and internal_user_database_enabled are set to true.

null
master_user_namestring(optional)

Master account user name. Only used if advanced_security_options and internal_user_database_enabled are set to true.

null
master_user_passwordstring(optional)

Master account user password. Only used if advanced_security_options and internal_user_database_enabled are set to true. WARNING: this password will be stored in Terraform state.

null
subnet_idslist(optional)

List of VPC Subnet IDs for the Elasticsearch domain endpoints to be created in. If zone_awareness_enabled is true, the first 2 or 3 provided subnet ids are used, depending on availability_zone_count. Otherwise only the first one is used.

list(string)
[]
tls_security_policystring(optional)

The name of the TLS security policy that needs to be applied to the HTTPS endpoint. Valid values are Policy-Min-TLS-1-0-2019-07 and Policy-Min-TLS-1-2-2019-07. Terraform performs drift detection if this is configured.

Policy-Min-TLS-1-2-2019-07
update_timeoutstring(optional)

How long to wait for updates to the ES cluster before timing out and reporting an error.

90m
vpc_idstring(optional)

The id of the VPC to deploy into. It must be in the same region as the Elasticsearch domain and its tenancy must be set to Default. If zone_awareness_enabled is false, the Elasticsearch cluster will have an endpoint in one subnet of the VPC; otherwise it will have endpoints in two subnets.

null