See issues on GitHub

Release Notes

Upgrade Notes:

  • This is a release candidate release. We recommend testing this version out in developement, and not running it in production yet.
  • Upgrading from 0.20 requires a full cluster restart.

Breaking changes:

  • BooleanFieldMapper should set IndexOptions.DOCS_ONLY by default. (#2866)
  • ignore_indices doesn't fail if all the indices requested are not present (#2837)
  • Filter / Id Cache Stats: Add to Indices Stats API, revise node stats API (#2862)
  • Thread Pool: Update default settings (move from default cached to fixed) (#2858)

New features:

  • Scripts not casting integers to doubles (#2834)

Enhancements:

  • Optimize aliases processing (#2832)
  • List of existing plugins with Node Info API (#2668)
  • support geo sorting on multiple geo point values per doc (#1846)
  • Field Data: Simplify field data cache settings (#2843)
  • FieldData Stats: Add field data stats to indices stats API (#2870)
  • Node Stats: Allow to explicitly get specific indices level node stats element (#2871)
  • Segments API: Add version & compound for each segment (#2823)
  • Settings / Config: Allow to explicitly specify external environment variable syntax, in which case its optional (#2855)
  • Stemmer override with large rule set causing java.lang.OutOfMemoryError: Java heap space (#2800)
  • Warmers: Have an explicit warmer thread pool (#2815)
  • {sort: "field"} throws misleading errors (#2835)

Bug fixes:

  • Make sure all_terms works consistently (#2861)
  • Can't retrieve a value from an empty GeoPointValues while doing Geo Distance sorting (#2851)
  • Uncaught exception in javascript (#2848)
  • ES 0.90 Multiple Numeric Range filters within boolean.should return incorrect results (#2826)
  • Least used store distributor allocates all data to the last directory on the list (#2820)
  • Phrase suggester throws ArrayIndexOutOfBoundsException when stopwords are the only text values (#2817)
  • Alias filter is ignored if a sort field is specified (#2816)
  • 0.90.0.RC1 Percolator, Not Matching Properly on Custom Analyzer (#2814)
  • TooManyOpenFiles might cause data-loss in ElasticSearch (Lucene) (#2812)
  • Date math in query_string caches now() (#2808)
  • When faceting on integer field using TERMS , 0.9 fails to count records where the facet field is missing (#2807)
  • 0.90.0.RC1 Percolator Query, NPE (#2806)