Elasticsearch 6.5.0 发布,分布式搜索与数据分析引擎

淡漠悠然
 淡漠悠然
发布于 2018年11月15日
收藏 9

Elasticsearch 6.5.0 已发布,Elasticsearch 是一个分布式的 RESTful 风格的搜索和数据分析引擎,能够解决不断涌现出的各种用例。作为 Elastic Stack 的核心,它集中存储你的数据,帮助你发现意料之中以及意料之外的情况。

Breaking changes

Aggregations

  • Require combine and reduce scripts in scripted metrics aggregation #33452 (issue: #32804)

Geo

  • Geo: Don’t flip longitude of envelopes crossing dateline #34535 (issue: #34418)

Mapping

  • Disallow "enabled" attribute change for types in mapping update (#33566) #33933

Plugins

Breaking Java changes

Authorization

  • Use RoleRetrievalResult for better caching #34197 (issue: #33205)

Core

  • Use generic AcknowledgedResponse instead of extended classes #32859

  • Switch WritePipelineResponse to AcknowledgedResponse #32722

Java High Level REST Client

  • HLRC XPack Protocol clean up: Licence, Misc #34469 (issue: #34451)

Plugins

  • Fix generics in ScriptPlugin#getContexts() #33426

Deprecations

Analysis

  • Deprecates the standard filter #33468

Scripting

  • Scripting: Conditionally use java time api in scripting #31441

Search

  • Deprecate negative query boost (#34486) #34512 (issues: #33309#34486)

  • Deprecate negative weight in Function Score Query #33624 (issue: #31927)

Suggesters

  • Make Geo Context Mapping Parsing More Strict (6.x version) #32862 (issue: #32821)

New features

Aggregations

  • Adds a new auto-interval date histogram #28993 (issue: #9572)

Analysis

  • Add script_filter tokenfilter #33431

  • Add conditional token filter to elasticsearch #31958

Audit

Authentication

  • Allow User/Password realms to disable authc #34033 (issue: #33292)

Authorization

  • [X-Pack] Beats centralized management: security role + licensing #34305 (issues: #30493#30520)

  • Add get-user-privileges API #33928 (issue: #32777)

  • Add support for "authorization_realms" #33262

CCR

  • [CCR] Added auto follow patterns feature #33118 (issue: #33007)

Index APIs

  • Add an index setting to control TieredMergePolicy#deletesPctAllowed #32907

Java High Level REST Client

  • add start trial API to HLRC #32799

Machine Learning

  • [ML] Label anomalies with multi_bucket_impact #34233

  • [ML] Add a file structure determination endpoint #33471

  • [ML] Partition-wise maximum scores #32748

Mapping

Monitoring

SQL

  • SQL: Implement CONVERT, an alternative to CAST #34660 (issue: #34513)

  • SQL: Implement IN(value1, value2, …) expression. #34581 (issue: #32955)

  • SQL: TRUNCATE and ROUND functions #33779 (issue: #33494)

  • SQL: Adds MONTHNAME, DAYNAME and QUARTER functions #33411 (issue: #33092)

Search

  • Add max_children limit to nested sort #33587 (issue: #33592)

Enhancements

Aggregations

  • Rollup adding support for date field metrics (#34185) #34200 (issue: #34185)

  • Add early termination support for min/max aggregations #33375

  • Add early termination support to BucketCollector #33279

  • Add interval response parameter to AutoDateInterval histogram #33254

  • Scripted metric aggregations: add deprecation warning and system property to control legacy params #31597 (issues: #29328#30111)

Allocation

  • Skip rebalancing when cluster_concurrent_rebalance threshold reached #33329 (issue: #27628)

Analysis

  • Allow TokenFilterFactories to rewrite themselves against their preceding chain #33702 (issue: #33609)

Audit

  • Add opaque_id to index audit logging #32260 (issue: #31521)

Authentication

  • Security: reduce memory usage of DnRoleMapper #34250 (issue: #34237)

  • Security: upgrade unboundid ldapsdk to 4.0.8 #34247 (issue: #33175)

  • [Kerberos] Add realm name & UPN to user metadata #33338

  • [SECURITY] Set Auth-scheme preference #33156 (issue: #32699)

  • Token API supports the client_credentials grant #33106

Authorization

  • [Authz] Allow update settings action for system user #34030 (issue: #33119)

  • Calculate changed roles on roles.yml reload #33525 (issue: #33205)

  • [Kerberos] Add authorization realms support to Kerberos realm #32392

  • [X-Pack] Beats centralized management: security role + licensing #30520 (issue: #30493)

Beats Plugin

  • [Monitoring] Update beats template to include apm-server metrics #33286

CRUD

  • Verify primary mode usage with assertions #32667 (issues: #10708#25692#32442)

  • Refactor TransportShardBulkAction to better support retries #31821

Circuit Breakers

  • Whitelisting / from Circuit Breaker Exception (#32325) #32666

Core

  • Improved IndexNotFoundException’s default error message #34649 (issue: #34628)

  • Prevent cause from being null in ShardOperationFailedException #32640 (issue: #32608)

  • Enable avoiding mmap bootstrap check #32421 (issue: #32267)

Discovery-Plugins

  • [GCE Discovery] Automatically set project-id and zone #33721 (issue: #13618)

Distributed

  • Add contains method to LocalCheckpointTracker #33871 (issue: #33656)

  • Introduce global checkpoint listeners #32696 (issue: #32651)

  • Expose whether or not the global checkpoint updated #32659 (issue: #32651)

  • Include translog path in error message when translog is corrupted #32251 (issue: #24929)

Docs Infrastructure

  • Docs: Allow snippets to have line continuation #32649

Engine

  • Do not alloc full buffer for small change requests #35158

  • Fill LocalCheckpointTracker with Lucene commit #34474 (issues: #0#2#33656)

  • Lock down Engine.Searcher #34363 (issue: #34357)

  • Fold EngineSearcher into Engine.Searcher #34082

  • Build DocStats from SegmentInfos in ReadOnlyEngine #34079 (issue: #33903)

  • Move CompletionStats into the Engine #33847 (issue: #33835)

  • Move DocsStats into Engine #33835

  • Add read-only Engine #33563 (issues: #32844#32867)

  • Allow engine to recover from translog upto a seqno #33032 (issue: #32867)

Index APIs

  • Introduce index settings version #34429

  • Add cluster-wide shard limit warnings #34021 (issues: #20705#32856)

  • Copy and validate soft-deletes setting on resize #33517 (issue: #33321)

  • Introduce mapping version to index metadata #33147

  • update rollover to leverage write-alias semantics #32216

Ingest

  • ingest: processor stats #34724 (issue: #34202)

  • ingest: better support for conditionals with simulate?verbose #34155

  • ingest: correctly measure chained pipeline stats #33912

  • ingest: support simulate with verbose for pipeline processor #33839

  • [ingest] geo-ip performance improvements #33029

  • ingest: Introduce the dissect processor #32884

  • INGEST: Add Pipeline Processor #32473 (issue: #31842)

  • Ingest: Add conditional per processor #32398 (issue: #21248)

  • Introduce the dissect library #32297

  • INGEST: Enable default pipelines #32286 (issue: #21101)

  • INGEST: Implement Drop Processor #32278 (issue: #23726)

  • Ingest: Support integer and long hex values in convert #32213 (issue: #32182)

  • Add region ISO code to GeoIP Ingest plugin #31669

Java High Level REST Client

  • HLRC: Delete role API #34620

  • [HLRC] Add support for Delete role mapping API #34531

  • [HLRC] Add Start/Stop Watch Service APIs. #34317

  • HLRC: ML Add preview datafeed api #34284 (issue: #29827)

  • HLRC: ML Adding get datafeed stats API #34271 (issue: #29827)

  • HLRC: Deactivate Watch API #34192 (issues: #29827#33988)

  • Create/Update role mapping API #34171 (issue: #33745)

  • HLRC: Get SSL Certificates API #34135

  • [ML][HLRC] Replace REST-based ML test cleanup with the ML client #34109

  • HLRC: Add activate watch action #33988 (issue: #29827)

  • Add support for ack watch to the HLRC. #33962 (issue: #29827)

  • HLRC: Add throttling for update & delete-by-query #33951

  • HLRC: ML Stop datafeed API #33946 (issue: #29827)

  • HLRC: Add get rollup job #33921

  • HLRC: ML start data feed API #33898 (issue: #29827)

  • HLRC: Add support for reindex rethrottling #33832

  • HLRC: Reindex should support requests_per_seconds parameter #33808

  • HLRC: Delete ML calendar #33775 (issue: #29827)

  • HLRC: Get ML calendars #33760 (issue: #29827)

  • [HLRC] Support for role mapper expression dsl #33745

  • [HLRC][ML] Add ML get datafeed API to HLRC #33715 (issue: #29827)

  • REST client: introduce a strict deprecation mode #33708 (issue: #33534)

  • [HLRC][ML] Add ML delete datafeed API to HLRC #33667 (issue: #29827)

  • HLRC: Add support for XPack Post Start Basic Licence API #33606 (issue: #29827)

  • [HLRC][ML] Add ML put datafeed API to HLRC #33603 (issue: #29827)

  • Create a WatchStatus class for the high-level REST client. #33527

  • HLRC: ML Delete Forecast API #33526 (issue: #29827)

  • Add create rollup job api to high level rest client #33521 (issues: #29827#32703)

  • HLRC: add change password API support #33509 (issue: #33481)

  • HLRC: ML Forecast Job #33506 (issue: #29827)

  • HLRC: add enable and disable user API support #33481 (issue: #29827)

  • HLRC: Add ML get categories API #33465 (issue: #29827)

  • HLRC: ML Post Data #33443 (issue: #29827)

  • add start trial API to HLRC #33406

  • HLRC: ML Update Job #33392 (issue: #29827)

  • HLRC: Add ML get influencers API #33389 (issue: #29827)

  • HLRC: ML PUT Calendar #33362 (issue: #29827)

  • HLRC: Add ML get overall buckets API #33297 (issue: #29827)

  • HLRC: create base timed request class #33216

  • HLRC: add client side RefreshPolicy #33209

  • HLRC: ML Flush job #33187 (issue: #29827)

  • HLRC: Adding ML Job stats #33183 (issue: #29827)

  • HLRC: Use Optional in validation logic #33104

  • HLRC: Add ML Get Records API #33085 (issue: #29827)

  • HLRC: Add ML Get Buckets API #33056 (issue: #29827)

  • GraphClient for the high level REST client and associated tests. #33025 (issue: #29827)

  • HLRC: Clear ML data after client tests #33023 (issue: #32993)

  • HLRC: Add ML Get Job #32960 (issue: #29827)

  • HLRC: ML Close Job #32943 (issue: #29827)

  • HLRC: Create server agnostic request and response #32912

  • Add GetRollupCaps API to high level rest client #32880 (issues: #29827#32703)

  • HLRC: adding machine learning open job #32860 (issue: #29827)

  • HLRC: Refactor WatchStatus #32842 (issue: #29827)

  • HLRC: adding machine learning delete job #32820 (issue: #29827)

  • HLRC: Refactor WatchStatus and implement activate watch #32802 (issue: #29827)

  • REST high-level client: add delete by query API #32782 (issues: #27205#32679#32760)

  • REST high-level client: add update by query API #32760 (issues: #27205#32679)

  • HLRC: migration get assistance API #32744 (issue: #29827)

  • Adding ML HLRC wrapper and put_job API call #32726

  • Add create rollup job api to high level rest client #32703 (issue: #29827)

  • REST high-level client: add reindex API #32679 (issue: #27205)

  • HLRC: Add Delete License API #32586 (issue: #29827)

  • Rest HL client: Add get license action #32438 (issue: #29827)

  • HLRC: Add delete watch action #32337 (issue: #29827)

  • HLRest: add xpack put user API #32332 (issue: #29827)

  • Rest HL client: Add put license action #32214 (issue: #29827)

  • Add Restore Snapshot High Level REST API #32155 (issue: #27205)

  • Add put stored script support to high-level rest client #31323 (issue: #27205)

Logging

  • Logging: Make node name consistent in logger #31588

Machine Learning

  • ML: Adding support for lazy nodes (#29991) #34538 (issue: #29991)

  • [ML] Add an ingest pipeline definition to structure finder #34350

  • [ML] Add a timeout option to file structure finder #34117

  • [ML] Allow asynchronous job deletion #34058 (issue: #32836)

  • Make certain ML node settings dynamic (#33565) #33961 (issue: #33565)

  • [ML] Display integers without .0 in file structure field stats #33947

  • [ML] Return both Joda and Java formats from structure finder #33900

  • Adding node_count to ML Usage (#33850) #33863 (issue: #33850)

  • Delete custom index if the only contained job is deleted #33788 (issue: #30075)

  • [ML] Allow overrides for some file structure detection decisions #33630

  • [ML] Minor improvements to categorization Grok pattern creation #33353

  • [ML] Delete forecast API (#31134) #33218 (issue: #31134)

Mapping

  • Preserve the order of nested documents in the Lucene index #34225 (issue: #33587)

  • Don’t count metadata fields towards index.mapping.total_fields.limit #33386 (issue: #24096)

  • Add expected mapping type to MapperException #31564 (issue: #31502)

Monitoring

  • [Monitoring] Add additional necessary mappings for apm-server #34392

  • Adding stack_monitoring_agent role #34369

  • [Monitoring] Add cluster metadata to cluster_stats docs #33860 (issue: #33691)

  • Implement xpack.monitoring.elasticsearch.collection.enabled setting #33474 (issue: #33290)

Network

  • Pass the host name on as server_name if proxy mode is on #34559

  • Bad regex in CORS settings should throw a nicer error #34035

  • Add sni name to SSLEngine in netty transport (#33144) #33513 (issue: #32517)

  • Add sni name to SSLEngine in netty transport #33144 (issue: #32517)

  • Add proxy support to RemoteClusterConnection #33062 (issues: #31840#32517)

  • Use a dedicated ConnectionManger for RemoteClusterConnection #32988 (issue: #31835)

Packaging

  • Add Ubuntu 18.04 to packaging tests #34139

Ranking

  • Add minimal sanity checks to custom/scripted similarities. (backport) #33893 (issue: #33564)

  • Use the global doc id to generate random scores #33599

Recovery

Rollup

  • [Rollup] Add support for date histo format when searching #34537 (issue: #34391)

  • [Rollup] Only allow aggregating on multiples of configured interval #32052

SQL

  • SQL: Improve CircuitBreaker logic for SqlParser #35300 (issue: #35299)

  • SQL: Optimizer rule for folding nullable expressions #35080 (issue: #34826)

  • SQL: Improve painless script generated from IN #35055 (issue: #34750)

  • SQL: Implement CAST between STRING and IP #34949 (issue: #34799)

  • SQL: Fix function args verification and error msgs #34926 (issues: #33469#34752)

  • SQL: Introduce ODBC mode, similar to JDBC #34825 (issue: #34720)

  • SQL: Introduce support for IP fields #34758 (issue: #32499)

  • SQL: Implement null handling for IN(v1, v2, ...) #34750 (issue: #34582)

  • SQL: handle X-Pack or X-Pack SQL not being available in a more graceful way #34736 (issue: #30009)

  • SQL: Support pattern against compatible indices #34718 (issues: #31611#31837#33803)

  • SQL: Allow min/max aggregates on date fields #34699 (issue: #34477)

  • SQL: Introduce support for NULL values (#34573) #34640 (issue: #32079)

  • SQL: return constants for all matching records in constants-containing SELECTs #34576 (issue: #31863)

  • SQL: Functions enhancements (OCTET_LENGTH function, order functions alphabetically, RANDOM function docs) #34101 (issue: #33477)

  • SQL: Internal refactoring of operators as functions #34097 (issue: #33975)

  • SQL: Remove more ANTLR4 grammar ambiguities #34074 (issue: #33854)

  • SQL: Move away internally from JDBCType to SQLType #33913 (issue: #33904)

  • SQL: Fix ANTL4 Grammar ambiguities. #33854 (issue: #31885)

  • SQL: Better handling of number parsing exceptions #33776 (issue: #33622)

  • SQL: Grammar tweak for number declarations #33767 (issue: #33765)

  • SQL: Return functions in JDBC driver metadata #33672 (issue: #33671)

  • SQL: Make Literal a NamedExpression #33583 (issue: #33523)

  • SQL: Improve alias vs index resolution #33393 (issue: #33363)

  • SQL: Align SYS TABLE for ODBC SQL_ALL_* args #33364 (issue: #33312)

  • SQL: Show/desc commands now support table ids #33363 (issue: #33294)

  • SQL: Support multi-index format as table identifier #33278

  • SQL: Multiple indices pattern #33162

  • SQL: skip uppercasing/lowercasing function tests for AZ locales as well #32910 (issue: #32589)

  • SQL: test coverage for JdbcResultSet #32813 (issue: #32078)

  • SQL: Added support for string manipulating functions with more than one parameter #32356(issue: #31604)

Scripting

  • Painless: Add Static Methods Shortcut #33440

  • Painless: Add Bindings #33042

Search

  • Replace version with reader cache key in IndicesRequestCache #34189 (issues: #27650#33473)

  • Handle terms query when detecting if a query can match nested docs #34072 (issue: #34067)

  • Add a limit for graph phrase query expansion #34031

  • Clarify RemoteClusterService#groupIndices behaviour #33899

  • Add nested and object fields to field capabilities response #33803 (issue: #33237)

  • Introduce a search_throttled threadpool #33732

  • Upgrade remote cluster settings #33537 (issues: #33413#33536)

  • Remove unsupported group_shard_failures parameter #33208 (issue: #32598)

  • Profiler: Don’t profile NEXTDOC for ConstantScoreQuery. #33196 (issue: #23430)

  • Change query field expansion #33020 (issues: #31655#31798)

  • Expose max_concurrent_shard_requests in _msearch #33016 (issue: #31877)

  • Search: Support of wildcard on docvalue_fields #32980 (issues: #26299#26390)

  • Ignore script fields when size is 0 #31917 (issue: #31824)

Security

  • Generate non-encrypted license public key #34626

  • Security: don’t call prepare index for reads #34568 (issues: #33205#34246)

  • Enable security automaton caching #34028

  • Add Debug/Trace logging to token service #34022

  • Security index expands to a single replica #33131 (issues: #29712#29933)

  • Introduce fips_mode setting and associated checks #32326

Settings

  • Introduce private settings #33327 (issue: #31286)

  • Add user-defined cluster metadata #33325 (issue: #33220)

  • Add settings updater for 2 affix settings #33050

Snapshot/Restore

  • Use more precise does S3 bucket exist method #34123

  • Add _source-only snapshot repository #32844

  • Increase max chunk size to 256Mb for repo-azure #32101 (issue: #12448)

  • Update AWS SDK to 1.11.340 in repository-s3 #30723 (issues: #22758#25552#30474)

Stats

  • Add cluster UUID to Cluster Stats API response #32206 (issue: #32205)

Store

  • add elasticsearch-shard tool to 6.x #33848 (issue: #31389)

  • drop index.shard.check_on_startup: fix #32279 (issue: #31389)

Suggesters

  • Completion types with multi-fields support #34081 (issue: #15115)

Watcher

  • Watcher: Reduce script cache churn by checking for mustache tags #33978 (issue: #29280)

  • [Watcher] Improved error messages for CronEvalTool #32800 (issue: #32735)

  • Watcher: Use Bulkprocessor in HistoryStore/TriggeredWatchStore #32490

  • Watcher: migrate PagerDuty v1 events API to v2 API #32285 (issue: #32243)

ZenDiscovery

  • Allow excluding folder names when scanning for dangling indices #34349

Bug fixes

Aggregations

  • Fix handling of empty keyword in terms aggregation #34457 (issue: #34434)

  • Check self references in metric agg after last doc collection (#33593) #34001

  • Unmapped aggs should not run pipelines if they delegate reduction #33528 (issue: #33514)

  • For filters aggregations, make sure that rewrites preserve other_bucket. #32921 (issue: #32834)

Allocation

  • DiskThresholdDecider#canAllocate can report negative free bytes #33641 (issue: #33596)

  • Don’t omit default values when updating routing exclusions (#32721) #33638

Analysis

  • Check stemmer language setting early #34601 (issue: #34170)

Authentication

  • ListenableFuture should preserve ThreadContext #34394

  • Allow an AuthenticationResult to return metadata #34382 (issues: #34290#34332)

  • Preserve thread context during authentication #34290

  • [Kerberos] Add debug log statement for exceptions #32663

  • [Kerberos] Remove Kerberos bootstrap checks #32451

Authorization

  • Handle missing user in user privilege APIs #34575 (issue: #34567)

  • Allow query caching by default again #33328 (issue: #33191)

  • Fix role query that can match nested documents #32705

  • Make get all app privs requires "*" permission #32460

CAT APIs

  • Fix potential NPE in _cat/shards/ with partial CommonStats #33858

  • Cat apis: Fix index creation time to use strict date format #32510 (issue: #32466)

CRUD

Circuit Breakers

  • Make accounting circuit breaker settings dynamic #34372 (issue: #34368)

Core

  • Fix AutoQueueAdjustingExecutorBuilder settings validation #33922

  • Fix Javadoc issues in 6.x for JDK11 #33579

  • Core: Add java time xcontent serializers #33120 (issue: #31853)

  • Protect scheduler engine against throwing listeners #32998

  • Fix content type detection with leading whitespace #32632 (issue: #32357)

Distributed

  • Only notify ready global checkpoint listeners #33690

  • Enable global checkpoint listeners to timeout #33620 (issue: #32696)

  • Fix race between replica reset and primary promotion #32442 (issues: #32118#32304#32431)

Engine

  • Acquire seacher on closing engine should throw AlreadyClosedException #33331 (issue: #33330)

  • Trim unreferenced translog when the safe commit advanced #32967 (issues: #28140#32089)

  • All Translog inner closes should happen after tragedy exception is set #32674 (issue: #32526)

Geo

  • Fix north pole overflow error in GeoHashUtils.bbox() #32891 (issue: #32857)

  • Use the determinant formula for calculating the orientation of a polygon #27967

Index APIs

  • Make XContentBuilder in AliasActions build is_write_index field #35071

  • Do not update number of replicas on no indices #34481

  • [Security] Get Alias API wildcard exclusion with Security #34144 (issues: #33518#33805)

  • Allow to clear the fielddata cache per field #33807 (issue: #33798)

  • CORE: Make Pattern Exclusion Work with Aliases #33518 (issue: #33395)

  • Fix IndexMetaData loads after rollover #33394 (issue: #33316)

  • Copy missing segment attributes in getSegmentInfo #32396

Ingest

  • INGEST: Create Index Before Pipeline Execute #32786 (issue: #32758)

Java High Level REST Client

  • HLRC: Fixing bug when getting a missing pipeline #34286 (issue: #34119)

  • Aggregations/HL Rest client fix: missing scores #32774 (issue: #32770)

  • HLRC: Ban LoggingDeprecationHandler #32756 (issue: #32151)

  • HLRC: Move commercial clients from XPackClient #32596

  • High-level client: fix clusterAlias parsing in SearchHit #32465

  • REST high-level client: parse back _ignored meta field #32362

License

  • Address license state update/read thread safety #33396

Logging

  • Logging: Configure the node name when we have it #32983 (issue: #32793)

Machine Learning

  • [ML] Prevent notifications being created on deletion of a non existent job #35337 (issues: #34058#35336)

  • [ML] Prevent default job values overwriting nulled fields #34804

  • Handle pre-6.x time fields #34373

  • [ML] Get job stats request should filter non-ML job tasks #33516 (issue: #33515)

  • [ML] Prevent NPE parsing the stop datafeed request. #33347

  • [ML] fix updating opened jobs scheduled events (#31651) #32881 (issue: #31651)

  • Clear Job#finished_time when it is opened (#32605) #32755

  • [ML] Fix thread leak when waiting for job flush (#32196) #32541 (issue: #32196)

Mapping

  • Fix field mapping updates with similarity #33634 (issue: #33611)

  • Ensure that _exists queries on keyword fields use norms when they’re available. #33006

  • Make sure that field collapsing supports field aliases. #32648 (issue: #32623)

  • Improve the error message when an index is incompatible with field aliases. #32482

Monitoring

  • Typo in x-pack template for thread_pool.management #34224

Network

  • NETWORKING: Add SSL Handler before other Handlers #34636 (issue: #33998)

  • Handle null SSLSessions during invalidation #34130 (issue: #32124)

  • Support PKCS#11 tokens as keystores and truststores #34063 (issue: #11)

  • Parse PEM Key files leniantly #33173 (issue: #33168)

  • NETWORKING: http.publish_host Should Contain CNAME #32806 (issue: #22029)

  • NETWORKING: Make RemoteClusterConn. Lazy Resolve DNS #32764 (issue: #28858)

  • Release requests in cors handle #32410

  • Release requests in cors handler #32364

Packaging

  • Fix use of hostname in Windows service #34193

  • Add temporary directory cleanup workarounds #32615 (issue: #31732)

Percolator

  • Ignore date ranges containing now when pre-processing a percolator query #35160

REST API

  • Core: Fix IndicesSegmentResponse.toXcontent() serialization #33414 (issue: #29120)

Recovery

  • Resync fails to notify on unavaiable exceptions #33615 (issues: #31179#33613)

  • Ensure to generate identical NoOp for the same failure #33141 (issue: #32986)

Rollup

  • [Rollup] Proactively resolve index patterns in RollupSearch endoint #34930 (issue: #34828)

  • Address BWC bug due to default metrics in (#34764) #34810 (issue: #34764)


  • [Rollup] Fix Caps Comparator to handle calendar/fixed time #33336 (issue: #32052)

  • [Rollup] Better error message when trying to set non-rollup index #32965

  • [Rollup] Return empty response when aggs are missing #32796 (issue: #32256)

  • [Rollup] Improve ID scheme for rollup documents #32558 (issue: #32372)

SQL

  • SQL: Fix null handling for AND and OR in SELECT #35277 (issue: #35240)

  • SQL: Handle null literal for AND and OR in WHERE #35236 (issue: #35088)

  • SQL: Introduce NotEquals node to simplify expressions #35234 (issues: #35210#35233)

  • SQL: handle wildcard expansion on incorrect fields #35134 (issue: #35092)

  • SQL: Fix null handling for IN ⇒ painless script #35124 (issues: #35108#35122)

  • SQL: Register missing processors #35121 (issue: #35119)

  • SQL: Fix NPE thrown if HAVING filter evals to null #35108 (issue: #35107)

  • SQL: Proper handling of nested fields at the beginning of the columns list #35068 (issue: #32951)

  • SQL: Fix incorrect AVG data type #34948 (issue: #33773)

  • SQL: Add CAST and CONVERT to SHOW FUNCTIONS #34940 (issue: #34939)

  • SQL: Handle aggregation for null group #34916 (issue: #34896)

  • SQL: Provide null-safe scripts for Not and Neg #34877 (issue: #34848)

  • SQL: Return error with ORDER BY on non-grouped. #34855 (issue: #34590)

  • SQL: Fix queries with filter resulting in NO_MATCH #34812 (issue: #34613)

  • SQL: Fix edge case: <field> IN (null) #34802 (issue: #34750)

  • SQL: Verifier allows aliases aggregates for sorting #34773 (issue: #34607)

  • SQL: the SSL default configuration shouldn&rsquo;t override the https protocol if used #34635 (issue: #33817)

  • JDBC: Fix artifactId in pom #34478 (issue: #34399)

  • SQL: Fix grammar for * in arithm expressions #34176 (issue: #33957)

  • SQL: Fix function resolution #34137 (issue: #34114)

  • SQL: Fix query translation of GroupBy with Having #34010 (issue: #33520)

  • SQL: Prevent StackOverflowError when parsing large statements #33902 (issue: #32942)

  • SQL: Fix issue with options for QUERY() and MATCH(). #33828 (issue: #32602)

  • SQL: Return correct catalog separator in JDBC #33670 (issue: #33654)

  • SQL: Fix result column names for CAST #33604 (issue: #33571)

  • SQL: Fix result column names for arithmetic functions #33500 (issues: #14#31869)

  • SQL: Fix bug in REPLACE function. Adds more tests to all string functions #33478

  • SQL: handle differently security connection related errors in the CLI #33255 (issue: #33230)

  • SQL: prevent duplicate generation for repeated aggs #33252 (issue: #30287)

  • SQL: Enable aggregations to create a separate bucket for missing values #32832 (issue: #32831)

  • SQL: Bug fix for the optional "start" parameter usage inside LOCATE function #32576 (issue: #32554)

  • SQL: Minor fix for javadoc #32573 (issue: #32553)

Scripting

  • Scripting: Add back lookup vars in score script #34833

  • Scripting: Add back params._source access in scripted metric aggs #34777 (issue: #33884)

  • Test: Fix last reference to SearchScript #34731 (issue: #34683)

  • Ensure map keys cannot be self referencing #34569

  • [Painless] Add a Map for java names to classes for use in the custom classloader #34424

  • [Painless] Allow statically imported methods without whitelisted class #34370

  • Painless: Remove caching of Painless scripts #34116

  • Painless: Fix Bindings Bug #33274

  • Painless: Fix Semicolon Regression #33212 (issue: #33193)

  • Scripting: Fix painless compiler loader to know about context classes #32385

Search

  • Fix inner_hits retrieval when stored fields are disabled #34652 (issues: #32941#33018)

  • Fix cross fields mode of the query_string query #34216 (issue: #34215)

  • Support string-style queries on metadata fields when reasonable. #34089 (issue: #34062)

  • Improves doc values format deprecation message #33576 (issue: #33572)

  • Fix nested _source retrieval with includes/excludes #33180 (issues: #33163#33170)

  • Fix quoted exists query #33019 (issue: #28922)

  • Fix multi fields empty query #33017 (issue: #33009)

  • XContentBuilder to handle BigInteger and BigDecimal #32888 (issue: #32395)

  • Do NOT allow termvectors on nested fields #32728 (issues: #21625#32652)

  • Cross-cluster search: preserve cluster alias in shard failures #32608

Security

  • Security: use x-pack config files when present #33688 (issue: #33464)

  • Security: use default scroll keepalive #33639

  • Enable FIPS140LicenseBootstrapCheck #32903 (issue: #32437)

Settings

  • CORE: Validate Type for String Settings #33503 (issue: #33135)

  • Fix deprecated setting specializations #33412

  • Apply settings filter to get cluster settings API #33247

Snapshot/Restore

  • Register Azure max_retries setting #35286

  • Do not override named S3 client credentials #33793 (issue: #33769)

  • Ensure fully deleted segments are accounted for correctly #33757 (issues: #32844#33689#33755)

Suggesters

  • Fix completion suggester&rsquo;s score tie-break #34508 (issue: #34378)

  • Null completion field should not throw IAE #33268

Transport API

  • Fix serialization of empty field capabilities response #33263

Watcher

  • watcher: Fix integration tests to ensure correct start/stop of Watcher #35271 (issues: #29877#30705#33291#34448#34462)

  • Make Watcher validation message copy/pasteable #33497 (issue: #33369)

  • Watcher: Reload properly on remote shard change #33167

  • Watcher: Fix race condition when reloading watches #33157

  • Guard against null in email admin watches #32923 (issue: #32590)

  • Watcher: Properly find next valid date in cron expressions #32734

ZenDiscovery

  • Fix logging of cluster state update descriptions #34182 (issue: #28941)

Regressions

Search

  • Preserve index_uuid when creating QueryShardException #32677 (issue: #32608)

Upgrades

Core

  • CORE: Upgrade to Jackson 2.8.11 #32670 (issue: #30352)

  • Dependencies: Upgrade to joda time 2.10 #32160

Ingest

  • Update geolite2 database in ingest geoip plugin #33840

Logging

Network

  • NETWORKING: Upgrade Netty to 4.1.30 #34417 (issue: #34411)

  • NETWORKING: Upgrade to Netty 4.1.29 #33984

Search

  • Upgrade to Lucene-7.5.0-snapshot-13b9e28f9d #32730

  • Upgrade to Lucene-7.5.0-snapshot-608f0277b0 #32390

Watcher

  • Dependencies: Update javax.mail in watcher to 1.6.2 #33664

下载地址:

https://www.elastic.co/cn/downloads/elasticsearch

本站文章除注明转载外,均为本站原创或编译。欢迎任何形式的转载,但请务必注明出处,尊重他人劳动共创开源社区。
转载请注明:文章转载自 开源中国社区 [http://www.oschina.net]
本文标题:Elasticsearch 6.5.0 发布,分布式搜索与数据分析引擎
加载中

最新评论(3

IT梁朝伟
IT梁朝伟
有毒吧?更新这么快。
MGL_TECH
MGL_TECH
66666666666666 这么多更新
GenerationWhy
GenerationWhy
程序客户端组件都跟不上啊
返回顶部
顶部