Product release

Elasticsearch 5.5.0 released

By
Clinton Gormley

Today we are pleased to announce the release of Elasticsearch 5.5.0, based on Lucene 6.6.0. This is the latest stable release, and is already available for deployment on Elastic Cloud, our Elasticsearch-as-a-service platform. This release includes a security fix for users of Elasticsearch X-Pack Security.

Latest stable release in 5.x:

You can read about all the changes in the release notes linked above, but there are a few changes which are worth highlighting:

Sensitive Information Leak (ESA-2017-10)

Elasticsearch X-Pack Security versions 5.0.0 to 5.4.3, when enabled, can result in the Elasticsearch _nodes API leaking sensitive configuration information, such as the paths and passphrases of SSL keys that were configured as part of an authentication realm. This could allow an authenticated Elasticsearch user to improperly view these details.

All users of X-Pack security should upgrade to version 5.5.0. This update will prevent the _nodes API from returning sensitive settings. If you cannot upgrade, any sensitive settings can be hidden by using the X-Pack hide_settings configuration option.

CVE-ID: CVE-2017-8442

Windows MSI Installer release

With this release of the Elastic stack, we're proud to announce the first release of the Windows MSI Installer for Elasticsearch. The installer provides a graphical user interface with a step-by-step wizard to guide users through the process of getting started with Elasticsearch on Windows:


Elasticsearch Node Configuration using the Windows Installer

In addition, it supports many command line options for flexible, silent installations. Please download the Windows Installer for Elasticsearch 5.5.0 and try it out, and look out for a blog post coming soon with more details.


ICU Collation Keyword Fields

Collations are used for sorting documents in a language-specific word order. The icu_collation_keyword field type allows you to convert a string (e.g. a surname) into bytes which represent the sort order for that string in a particular language. These bytes are indexed as a single token, just like the keyword field, and stored in doc values for fast sorting. Multiple collation fields can exist for the same string. For instance, the following will allow sorting in English order and in German “phonebook” order:

PUT my_index
{
  "mappings": {
    "user": {
      "properties": {
        "name": {
          "type": "text",
          "fields": {
            "de_phonebook": {
              "type": "icu_collation_keyword",
              "index": false,
              "language": "de",
              "country": "DE",
              "variant": "@collation=phonebook"
            },
            "en": {
              "type": "icu_collation_keyword",
              "index": false,
              "language": "en",
              "country": "US"
            }
          }
        }
      }
    }
  }
}

Other Changes

  • A new ip_range field types (similar to date and numeric range fields) enables the indexing of IPv6 ranges, and allows these ranges to be queried with a range.
  • The index.mapping.single_type setting will help you to prepare for the one-type-per-index change coming in 6.0.
  • Terms aggregations which are sorted on a sub-aggregation will use less memory than before.
  • The circuit breaker used by aggregations was allocating memory before accounting for the requested memory, which could result in an OOM exception.

Please download Elasticsearch 5.5.0, try it out, and let us know what you think on Twitter (@elastic) or in our forum. You can report any problems on the GitHub issues page.