Elasticsearch Install

Installs ElasticSearch on the specified hosts. It's divided into four main steps. The configuration of the Elastics Search Users and Group, the configuration of Kerberos, the installation of Elastics Search from rpm repositories and the configuration of Elastic Search environment

module.exports = header: 'ES Install', handler: ->
  {elasticsearch ,realm} = @config.ryba

Identities

By default, the "zookeeper" package create the following entries:

cat /etc/passwd | grep elasticsearch
elasticsearch:x:493:493:elasticsearch user:/home/elasticsearch:/sbin/nologin
cat /etc/group | grep elasticsearch
elasticsearch:x:493:

  @system.group header: 'Group', elasticsearch.group
  @system.user header: 'User', elasticsearch.user

Kerberos

  # @krb5.addprinc krb5,
  #   skip: true
  #   header: 'Kerberos'
  #   principal: elasticsearch.principal
  #   randkey: true
  #   keytab: elasticsearch.keytab
  #   uid: elasticsearch.user.name
  #   gid: elasticsearch.group.name

Install

ElasticSearch archive comes with an RPM

  @call header: 'Packages', ->
    @file.download
      source: elasticsearch.source
      target: "/var/tmp/elasticsearch-#{elasticsearch.version}.noarch.rpm"
      # unless_exec: "rpm -q --queryformat '%{VERSION}' elasticsearch | grep '#{elasticsearch.version}'"
      unless_exists: true
    @system.execute
      cmd:"""
      yum localinstall -y --nogpgcheck /var/tmp/elasticsearch-#{elasticsearch.version}.noarch.rpm
      chkconfig --add elasticsearch
      """
      unless_exec: "rpm -q --queryformat '%{VERSION}' elasticsearch | grep '#{elasticsearch.version}'"

Env

  @call header: 'Environment', ->
    @file.yaml
      target: '/etc/elasticsearch/elasticsearch.yml'
      content:
        'cluster.name': "#{elasticsearch.cluster.name}"
        'index.number_of_shards': "#{elasticsearch.number_of_shards}"
        'index.number_of_replicas': "#{elasticsearch.number_of_replicas}"
      merge: true
      backup: true
    # write = [
    #   match: /^.*cluster.name: .*/m
    #   replace: "cluster.name: \"#{elasticsearch.cluster.name}\" # RYBA CONF `elasticsearch.cluster.name`, DON'T OVERWRITE"
    # ,
    #   match: /^.*index.number_of_shards: .*/m
    #   replace: "index.number_of_shards: #{elasticsearch.number_of_shards} # RYBA CONF `elasticsearch.number_of_shards`, DON'T OVERWRITE"
    # ,
    #   match: /^.*index.number_of_replicas: .*/m
    #   replace: "index.number_of_replicas: #{elasticsearch.number_of_replicas} # RYBA CONF `elasticsearch.number_of_replicas`, DON'T OVERWRITE"
    # ]
    # if elasticsearch.node?
    #   {node} = elasticsearch
    #   if node.name? then write.push
    #     match: /^.*node.name: .*/m
    #     replace: "node.name: \"#{elasticsearch.node.name}\" # RYBA CONF `elasticsearch.node.name`, DON'T OVERWRITE"
    #   if node.master? then write.push
    #     match: /^node.master: .*/m
    #     replace: "node.master: #{elasticsearch.node.master} # RYBA CONF `elasticsearch.node.master`, DON'T OVERWRITE"
    #     append: true
    #   if node.data? then write.push
    #     match: /^node.data: .*/m
    #     replace: "node.data: #{elasticsearch.node.data} # RYBA CONF `elasticsearch.node.data`, DON'T OVERWRITE"
    #     append: true
    # @file
    #   target: 
    #   write: write