Logstash 7.6.0 Release Notes

  • Feature: Introduce deprecation logger for internal classes and plugins. #11260 and #11486

    • The Deprecation logger is a unified way for Logstash components to log deprecation notices into a separate file, located by default at log/logstash-deprecation.log. This file gives users a single location to see if they are using features that may stop working after a major upgrade.
  • Feature: Add support for cloud-id/auth for Logstash monitoring/management #11496
  • Feature: Initial release of the Jdbc Integration Plugin, which combines previously-separate Jdbc plugins and shared dependencies into a single codebase
  • Fixed: Regression where compilation of multiple pipelines experiences slowdown #11560

    • The fix for the Java execution pipeline compilation slowdown relative to the number of workers in #11482 introduced a regression which caused a slowdown of pipeline compilation when using multiple pipelines. This fix solves that regression and the original issue when using multiple workers.
  • Updated puma to 4.x #11241
  • Updated jruby to 9.2.9.0 #11281
  • Fixed: Correct directory for versions.yml file when building plugins #11318 This fixes an issue where a versions.yml was unnecessarily required when trying to build native Java plugins
  • Updated sinatra and rack to 2.x #11354
  • Changed: base JRUBY_OPTS to default to --dev (for fast scripts) #11355
  • Fixed: Removed use of deprecated Thread.exclusive method, which caused a warning message every time logstash started. #11388
  • Add Enterprise license level subscription #11407
  • [DOC] Remove module-only disclaimer for cloud id #11469
  • [DOC] Add details about pipeline.workers #11474
  • [DOC] Add deprecation notice to internal collectors for monitoring #11526
  • Build: Fail license report job on missing licenses #11554
  • Fixed: Updated log4j2.properties file that the Docker container image uses to also log the pipeline.id. #11567

Plugins

Jdbc Integration

  • Initial release of the Jdbc Integration Plugin, which combines previously-separate Jdbc plugins and shared dependencies into a single codebase

Cef Codec

  • Fixed CEF short to long name translation for ahost/agentHostName field, according to documentation #75

Fluent Codec

  • Handle EventTime msgpack extension to handle nanosecond precision time and add its parameter #18

Dns Filter

  • Fixed an issue where each missed lookup could result in unreclaimed memory (jruby bug) by handling lookup misses without raising exceptions #61
  • Added restriction on JRuby resolv.rb patch to versions prior to 9.2.9.0 #58
  • Fixed asciidoc formatting for unordered list and a code sample in docs #57
  • Added search domains to the nameserver option #56

Elasticsearch Filter

  • Feat: support cloud_id / cloud_auth configuration #122

Beats Input

  • Updated Jackson dependencies

Elasticsearch Input

  • Feat: Added support for cloud_id / cloud_auth configuration #112
  • Changed Elasticsearch Client transport to use Manticore #111

File Input

  • Fix regression in exclude handling. Patterns are matched against the filename, not full path. #237

Http Input

  • Revert updates to netty and tcnative since CBC ciphers are still used in many contexts

Csv Output

  • Docs: Correct typos #19
  • Docs: Fix formatting after code sample #22

Elasticsearch Output

  • Feat: Added support for cloud_id and cloud_auth #906

S3 Output

  • Added ability to specify ONEZONE_IA as storage_class

Udp Output

  • Fixed plugin crash upon socket write exception #10
  • Added support for the retry_count and retry_backoff_ms options #12