Absible Elasticsearch

Install and configure Elasticsearch with java. This role is meant for my small need. It's inspired in part by the official role. If you need a more complete one, you can find it here: https://github.com/elastic/ansible-elasticsearch

Tested with v7.9.0

Requirements

You need a recent version of Java. Elasticsearch is now bundled with JVM, but you can use your own. Consult the official elasticsearch documentation if you don't want to use the bundled JVM.

Tested plateform

Debian 10 (Buster)

Role variables

The elasticsearch package version

elasticsearch_version: "7.9.0"

For now, it's only possible to install elasticsearch on Debian based distribution. It's also only possible ton install it via repository. The elasticsearch_major_version value is automatically set in a task, and is based on elasticsearch_version

elasticsearch_repo_base: "https://artifacts.elastic.co"
elasticsearch_apt_key: "{{ elasticsearch_repo_base }}/GPG-KEY-elasticsearch"
elasticsearch_apt_key_id: "46095ACC8548582C1A2699A9D27D666CD88E42B4"
elasticsearch_apt_url: "deb {{ elasticsearch_repo_base }}/packages/{{ elasticsearch_major_version }}/apt stable main"
elasticsearch_package_name: "elasticsearch"

Path used by Elasticsearch. The elasticsearch_snapshot_path is optional. You can set it to "" if you don't want to create snapshots on a daily basis in a cron task.

elasticsearch_conf_dir: "/etc/elasticsearch"
elasticsearch_pid_dir: "/var/run/elasticsearch"
elasticsearch_log_dir: "/var/log/elasticsearch"
elasticsearch_data_dirs:
  - /var/lib/elasticsearch
elasticsearch_snapshot_path: /var/local/elasticsearch
elasticsearch_default_file: "/etc/default/elasticsearch"
elasticsearch_home: "/usr/share/elasticsearch"

Lock the elasticsearch version. Its use, on debian, to mark the elasticsearch package on "hold" and to avoid unwanted upgrade.

elasticsearch_version_lock: false

The elasticsearch system user and group

elasticsearch_user: elasticsearch
elasticsearch_group: elasticsearch

Allow the elasticsearch service to be (re)started

elasticsearch_start_service: true

The elasticsearch host and port

elasticsearch_api_host: localhost
elasticsearch_api_port: 9200

The elasticsearch snapshot name. this value is optional, and can be set to "". But, if a value is set, elasticsearch_snapshot_path must be set to.

elasticsearch_snapshot_name: backup_es

Java heap size. Not defined by default and automatically set to 2g. Can be changed here.

elasticsearch_heap_size_xms: ""
elasticsearch_heap_size_xmx: ""

Default values of some JVM options. Can be changed.

elasticsearch_max_open_files: 65536
elasticsearch_max_map_count: 262144
elasticsearch_max_threads: 8192

Variables used in elastisearch.yml. By default, the cluster name is elasticsearch and the node name is default_node. Chane thoses to your liking.

elasticsearch_config_cluster_name: "elasticsearch"
elasticsearch_config_node_name: "default_node"
elasticsearch_discovery_seed_hosts: '["127.0.0.1"]'
elasticsearch_cluster_initial_master_nodes: '["master_node"]'

Examples

---
- hosts: somehost
  roles:
    - supertarto.elasticsearch
  vars:
    elasticsearch_config_cluster_name: supercluster
    elasticsearch_config_node_name: supernode
    elasticsearch_heap_size_xms: "4g"
    elasticsearch_heap_size_xmx: "4g"

Installation

ansible-galaxy install supertarto.elasticsearch

License

GPL V3.0