OpenSearch/docs/reference/release-notes/7.6.asciidoc

[[release-notes-7.6.2]]
== {es} version 7.6.2

Also see <<breaking-changes-7.6,Breaking changes in 7.6>>.

[[breaking-7.6.2]]
[discrete]
=== Breaking changes

Authorization::
* Creation of derived API keys (keys created by existing keys) now requires explicit "no privileges" configuration {es-pull}53647[#53647], https://www.elastic.co/community/security[CVE-2020-7009]

[[bug-7.6.2]]
[discrete]
=== Bug fixes

Allocation::
* Improve performance of shards limits decider {es-pull}53577[#53577] (issue: {es-issue}53559[#53559])

Authentication::
* Fix potential bug in concurrent token refresh support {es-pull}53668[#53668]

CCR::
* Handle no such remote cluster exception in ccr {es-pull}53415[#53415] (issue: {es-issue}53225[#53225])

Distributed::
* Execute retention lease syncs under system context {es-pull}53838[#53838] (issues: {es-issue}48430[#48430], {es-issue}53751[#53751])

Engine::
* Fix doc_stats and segment_stats of ReadOnlyEngine {es-pull}53345[#53345] (issues: {es-issue}51303[#51303], {es-issue}51331[#51331])

Features/ILM+SLM::
* Fix null config in SnapshotLifecyclePolicy.toRequest {es-pull}53328[#53328] (issues: {es-issue}44465[#44465], {es-issue}53171[#53171])
* ILM Freeze step retry when not acknowledged {es-pull}53287[#53287]

Features/Ingest::
* Fix ingest pipeline _simulate api with empty docs never returns a res… {es-pull}52937[#52937] (issue: {es-issue}52833[#52833])

Features/Java High Level REST Client::
* Add unsupported parameters to HLRC search request {es-pull}53745[#53745]
* Fix AbstractBulkByScrollRequest slices parameter via Rest {es-pull}53068[#53068] (issue: {es-issue}53044[#53044])

Features/Watcher::
* Disable Watcher script optimization for stored scripts {es-pull}53497[#53497] (issue: {es-issue}40212[#40212])

Infra/Core::
* Avoid self-suppression on grouped action listener {es-pull}53262[#53262] (issue: {es-issue}53174[#53174])

Infra/Packaging::
* Handle special characters and spaces in JAVA_HOME path in elasticsearch-service.bat {es-pull}52676[#52676]

Infra/Plugins::
* Ensure only plugin REST tests are run for plugins {es-pull}53184[#53184] (issues: {es-issue}52114[#52114], {es-issue}53183[#53183])

Machine Learning::
* Fix a bug in the calculation of the minimum loss leaf values for classification {ml-pull}1032[#1032]

Network::
* Invoke response handler on failure to send {es-pull}53631[#53631]

SQL::
* Fix NPE for parameterized LIKE/RLIKE {es-pull}53573[#53573] (issue: {es-issue}53557[#53557])
* Add support for index aliases for SYS COLUMNS command {es-pull}53525[#53525] (issue: {es-issue}31609[#31609])
* Fix issue with LIKE/RLIKE as painless script {es-pull}53495[#53495] (issue: {es-issue}53486[#53486])
* Fix column size for IP data type {es-pull}53056[#53056] (issue: {es-issue}52762[#52762])

Search::
* Fix Term Vectors with artificial docs and keyword fields {es-pull}53504[#53504] (issue: {es-issue}53494[#53494])
* Fix concurrent requests race over scroll context limit {es-pull}53449[#53449]
* Fix pre-sorting of shards in the can_match phase {es-pull}53397[#53397]
* Fix potential NPE in FuzzyTermsEnum {es-pull}53231[#53231] (issue: {es-issue}52894[#52894])

[[upgrade-7.6.2]]
[discrete]
=== Upgrades

Infra/Core::
* Update jackson-databind to 2.8.11.6 {es-pull}53522[#53522] (issue: {es-issue}45225[#45225])

[[release-notes-7.6.1]]
== {es} version 7.6.1

Also see <<breaking-changes-7.6,Breaking changes in 7.6>>.

[[bug-7.6.1]]
[discrete]
=== Bug fixes

Aggregations::
* Decode max and min optimization more carefully {es-pull}52336[#52336] (issue: {es-issue}52220[#52220])
* Fix a DST error in date_histogram {es-pull}52016[#52016] (issue: {es-issue}50265[#50265])

Audit::
* Logfile audit settings validation {es-pull}52537[#52537] (issues: {es-issue}47038[#47038], {es-issue}47711[#47711], {es-issue}52357[#52357])

CCR::
* Fix shard follow task cleaner under security {es-pull}52347[#52347] (issues: {es-issue}44702[#44702], {es-issue}51971[#51971])

Features/cat APIs::
* Fix NPE in RestPluginsAction {es-pull}52620[#52620] (issue: {es-issue}45321[#45321])

Features/ILM+SLM::
* ILM fix the init step to actually be retryable {es-pull}52076[#52076]

Features/Ingest::
* Handle errors when evaluating if conditions in processors {es-pull}52543[#52543] (issue: {es-issue}52339[#52339])

Features/Monitoring::
* Fix NPE in cluster state collector for monitoring {es-pull}52371[#52371] (issue: {es-issue}52317[#52317])

Features/Stats::
* Switch to AtomicLong for "IngestCurrent" metric to prevent negative values {es-pull}52581[#52581] (issues: {es-issue}52406[#52406], {es-issue}52411[#52411])

Infra/Packaging::
* Limit _FILE env var support to specific vars {es-pull}52525[#52525] (issue: {es-issue}52503[#52503])

Machine Learning::
* Don't return inflated definition when storing trained models {es-pull}52573[#52573]
* Validate tree feature index is within range {es-pull}52460[#52460]

Network::
* Remove seeds dependency for remote cluster settings {es-pull}52796[#52796]

Reindex::
* Allow comma separated source indices {es-pull}52044[#52044] (issue: {es-issue}51949[#51949])

SQL::
* Supplement input checks on received request parameters {es-pull}52229[#52229]
* Fix issue with timezone when paginating {es-pull}52101[#52101] (issue: {es-issue}51258[#51258])
* Fix ORDER BY on aggregates and GROUPed BY fields {es-pull}51894[#51894] (issue: {es-issue}50355[#50355])
* Fix milliseconds handling in intervals {es-pull}51675[#51675] (issue: {es-issue}41635[#41635])
* Selecting a literal from grouped by query generates error {es-pull}41964[#41964] (issues: {es-issue}41413[#41413], {es-issue}41951[#41951])

Snapshot/Restore::
* Fix Non-Verbose Snapshot List Missing Empty Snapshots {es-pull}52433[#52433]

Store::
* Fix synchronization in ByteSizeCachingDirectory {es-pull}52512[#52512]



[[upgrade-7.6.1]]
[discrete]
=== Upgrades

Authentication::
* Update oauth2-oidc-sdk to 7.0 {es-pull}52489[#52489] (issue: {es-issue}48409[#48409])

[[release-notes-7.6.0]]
== {es} version 7.6.0

Also see <<breaking-changes-7.6,Breaking changes in 7.6>>.

[[known-issues-7.6.0]]
[discrete]
=== Known issues

* Applying deletes or updates on an index after it has been shrunk may corrupt
the index. In order to prevent this issue, it is recommended to stop shrinking
read-write indices. For read-only indices, it is recommended to force-merge
indices after shrinking, which significantly reduces the likeliness of this
corruption in the case that deletes/updates would be applied by mistake. This
bug is fixed in {es} 7.7 and later versions. More details can be found on the
https://issues.apache.org/jira/browse/LUCENE-9300[corresponding issue].

* Indices created in 6.x with <<date,`date`>> and <<date_nanos,`date_nanos`>> fields using formats
that are incompatible with java.time patterns will cause parsing errors, incorrect date calculations or wrong search results.
https://github.com/elastic/elasticsearch/pull/52555
This is fixed in {es} 7.7 and later versions.

* Slow loggers can cause Log4j loggers to leak over time. When a new index is created,
 a new Log4j logger is associated with it. However, when an index is deleted,
  Log4j keeps an internal reference to its loggers that results in a memory leak (issue: {es-issue}56171[#56171])
+
This issue is fixed in {es} 6.8.10 and 7.7.1.

* Week-based date patterns are not working correctly with `Y`. Using `Y` with `w` will result in
a failed request and an exception in the logs (issue: {es-issue}57128[#57128]). Using `y` with `w` results in
incorrect date calculations. A workaround is to add the following line to the `jvm.options` file.
+
[source,shell]
--------------------------------------------
9-:-Djava.locale.providers=SPI,COMPAT
--------------------------------------------
+
This issue is fixed in {es} 7.7.0 and later versions (issue: {es-issue}50916[#50916]).

[[breaking-7.6.0]]
[discrete]
=== Breaking changes

Mapping::
* Add a cluster setting to disallow loading fielddata on _id field {es-pull}49166[#49166] (issues: {es-issue}26472[#26472], {es-issue}43599[#43599])



[[breaking-java-7.6.0]]
[discrete]
=== Breaking Java changes

Security::
* Support Client and RoleMapping in custom Realms {es-pull}50534[#50534] (issue: {es-issue}48369[#48369])



[[deprecation-7.6.0]]
[discrete]
=== Deprecations

Analysis::
* Deprecate and remove camel-case nGram and edgeNGram tokenizers {es-pull}50862[#50862] (issue: {es-issue}50561[#50561])

Authorization::
* Deprecating kibana_user and kibana_dashboard_only_user roles {es-pull}46456[#46456]

Distributed::
* Deprecate synced flush {es-pull}50835[#50835] (issue: {es-issue}50776[#50776])
* Deprecate indices without soft-deletes {es-pull}50502[#50502]

Features/Indices APIs::
* Emit warnings when index templates have multiple mappings {es-pull}50982[#50982]
* Ensure we emit a warning when using the deprecated 'template' field. {es-pull}50831[#50831] (issue: {es-issue}49460[#49460])

Infra/Core::
* Deprecate the 'local' parameter of /_cat/nodes {es-pull}50499[#50499] (issue: {es-issue}50088[#50088])

Reindex::
* Deprecate sorting in reindex {es-pull}49458[#49458] (issue: {es-issue}47567[#47567])

Search::
* Update the signature of vector script functions. {es-pull}48604[#48604]
* Deprecate the sparse_vector field type. {es-pull}48315[#48315]
* Add a deprecation warning regarding allocation awareness in search request {es-pull}48351[#48351] (issue: {es-issue}43453[#43453])


[[feature-7.6.0]]
[discrete]
=== New features

Aggregations::
* New Histogram field mapper that supports percentiles aggregations. {es-pull}48580[#48580] (issue: {es-issue}48578[#48578])
* Implement stats aggregation for string terms {es-pull}47468[#47468]

Analysis::
* Implement Lucene EstonianAnalyzer, Stemmer {es-pull}49149[#49149] (issue: {es-issue}48895[#48895])

Authentication::
* Password Protected Keystore (Feature Branch) {es-pull}49210[#49210]

Features/ILM+SLM::
* ILM action to wait for SLM policy execution {es-pull}50454[#50454] (issue: {es-issue}45067[#45067])
* Add ILM histore store index {es-pull}50287[#50287] (issue: {es-issue}49180[#49180])

Features/Ingest::
* CSV processor {es-pull}49509[#49509] (issue: {es-issue}49113[#49113])

Machine Learning::
* Implement `precision` and `recall` metrics for classification evaluation {es-pull}49671[#49671] (issue: {es-issue}48759[#48759])
* Explain data frame analytics API {es-pull}49455[#49455]
* Machine learning model inference ingest processor {es-pull}49052[#49052]
* Implement accuracy metric for multi-class classification {es-pull}47772[#47772] (issue: {es-issue}48759[#48759])
* Add feature importance values to classification and regression results (using tree
SHapley Additive exPlanation, or SHAP) {ml-pull}857[#857]

Mapping::
* Add per-field metadata. {es-pull}49419[#49419] (issue: {es-issue}33267[#33267])

Search::
* Add fuzzy intervals source {es-pull}49762[#49762] (issue: {es-issue}49595[#49595])
* Add a listener to track the progress of a search request locally {es-pull}49471[#49471] (issue: {es-issue}49091[#49091])



[[enhancement-7.6.0]]
[discrete]
=== Enhancements

Aggregations::
* Add reusable HistogramValue object   {es-pull}49799[#49799] (issue: {es-issue}49683[#49683])
* Optimize composite aggregation based on index sorting {es-pull}48399[#48399] (issue: {es-issue}48130[#48130])

Allocation::
* Auto-expand indices according to allocation filtering rules {es-pull}48974[#48974]
* Do not cancel ongoing recovery for noop copy on broken node {es-pull}48265[#48265] (issue: {es-issue}47974[#47974])
* Quieter logging from the DiskThresholdMonitor {es-pull}48115[#48115] (issue: {es-issue}48038[#48038])
* Faster access to INITIALIZING/RELOCATING shards {es-pull}47817[#47817] (issues: {es-issue}46941[#46941], {es-issue}48579[#48579])

Analysis::
* Check for deprecations when analyzers are built {es-pull}50908[#50908] (issue: {es-issue}42349[#42349])
* Make Multiplexer inherit filter chains analysis mode {es-pull}50662[#50662] (issue: {es-issue}50554[#50554])
* Allow custom characters in token_chars of ngram tokenizers {es-pull}49250[#49250] (issue: {es-issue}25894[#25894])

Authentication::
* Add Debug/Trace logging for authentication {es-pull}49575[#49575] (issue: {es-issue}49473[#49473])

Authorization::
* Increase Size and lower TTL on DLS BitSet Cache {es-pull}50535[#50535] (issues: {es-issue}43669[#43669], {es-issue}49260[#49260])
* Add 'monitor_snapshot' cluster privilege {es-pull}50489[#50489] (issue: {es-issue}50210[#50210])
* Remove reserved roles for code search {es-pull}50068[#50068] (issue: {es-issue}49842[#49842])
* [Code] Remove code_admin/code_user roles {es-pull}48164[#48164]
* Resolve the role query and the number of docs lazily {es-pull}48036[#48036]

CCR::
* Improve error message when pausing index {es-pull}48915[#48915]
* Use MultiFileTransfer in CCR remote recovery {es-pull}44514[#44514] (issue: {es-issue}44468[#44468])

CRUD::
* print id detail when id is too long. {es-pull}49433[#49433]
* Add preflight check to dynamic mapping updates {es-pull}48817[#48817] (issue: {es-issue}35564[#35564])

Cluster Coordination::
* Move metadata storage to Lucene {es-pull}50907[#50907] (issue: {es-issue}48701[#48701])
* Remove custom metadata tool {es-pull}50813[#50813] (issue: {es-issue}48701[#48701])

Distributed::
* Use retention lease in peer recovery of closed indices {es-pull}48430[#48430] (issue: {es-issue}45136[#45136])

Engine::
* Do not force refresh when write indexing buffer {es-pull}50769[#50769]
* Deleted docs disregarded for if_seq_no check {es-pull}50526[#50526]
* Allow realtime get to read from translog {es-pull}48843[#48843]
* Do not warm up searcher in engine constructor {es-pull}48605[#48605] (issue: {es-issue}47186[#47186])
* Add a new merge policy that interleaves old and new segments on force merge {es-pull}48533[#48533] (issue: {es-issue}37043[#37043])
* Refresh should not acquire readLock {es-pull}48414[#48414] (issue: {es-issue}47186[#47186])

Features/ILM+SLM::
* Refresh cached phase policy definition if possible on new poli… {es-pull}50820[#50820] (issue: {es-issue}48431[#48431])
* Make the UpdateRolloverLifecycleDateStep retryable {es-pull}50702[#50702] (issue: {es-issue}48183[#48183])
* Make InitializePolicyContextStep retryable {es-pull}50685[#50685] (issue: {es-issue}48183[#48183])
* ILM retryable async action steps {es-pull}50522[#50522] (issues: {es-issue}44135[#44135], {es-issue}48183[#48183])
* Make the TransportRolloverAction execute in one cluster state update {es-pull}50388[#50388]
* ILM open/close steps are noop if idx is open/close {es-pull}48614[#48614]
* ILM Make the `check-rollover-ready` step retryable {es-pull}48256[#48256] (issue: {es-issue}44135[#44135])

Features/Ingest::
* Foreach processor - fork recursive call  {es-pull}50514[#50514]
* Sync grok patterns with logstash patterns {es-pull}50381[#50381]
* Replace required pipeline with final pipeline {es-pull}49470[#49470] (issue: {es-issue}49247[#49247])
* Add templating support to enrich processor {es-pull}49093[#49093]
* Introduce on_failure_pipeline ingest metadata inside on_failure block {es-pull}49076[#49076] (issue: {es-issue}44920[#44920])
* Add templating support to pipeline processor. {es-pull}49030[#49030] (issue: {es-issue}39955[#39955])
* Add option to split processor for preserving trailing empty fields {es-pull}48664[#48664] (issue: {es-issue}48498[#48498])
* Change grok watch dog to be Matcher based instead of thread based. {es-pull}48346[#48346] (issues: {es-issue}43673[#43673], {es-issue}47374[#47374])
* update ingest-user-agent regexes.yml {es-pull}47807[#47807]

Features/Java High Level REST Client::
* Add remote info to the HLRC {es-pull}49657[#49657] (issue: {es-issue}47678[#47678])
* Add delete alias to the HLRC {es-pull}48819[#48819] (issue: {es-issue}47678[#47678])

Features/Monitoring::
* Significantly Lower Monitoring HttpExport Memory Footprint {es-pull}48854[#48854]
* Validate proxy base path at parse time {es-pull}47912[#47912] (issue: {es-issue}47711[#47711])
* Validate index name time format setting at parse time {es-pull}47911[#47911] (issue: {es-issue}47711[#47711])
* Validate monitoring header overrides at parse time {es-pull}47848[#47848] (issue: {es-issue}47711[#47711])
* Validate monitoring username at parse time {es-pull}47821[#47821] (issue: {es-issue}47711[#47711])
* Validate monitoring password at parse time {es-pull}47740[#47740] (issue: {es-issue}47711[#47711])

Features/Stats::
* Add ingest info to Cluster Stats {es-pull}48485[#48485] (issue: {es-issue}46146[#46146])

Features/Watcher::
* Log attachment generation failures {es-pull}50080[#50080]
* Don't dump a stacktrace for invalid patterns when executing elasticse… {es-pull}49744[#49744] (issue: {es-issue}49642[#49642])

Geo::
* "CONTAINS" support for BKD-backed geo_shape and shape fields {es-pull}50141[#50141] (issue: {es-issue}41204[#41204])
* Adds support for geo-bounds filtering in geogrid aggregations {es-pull}50002[#50002]
* Introduce faster approximate sinh/atan math functions {es-pull}49009[#49009] (issue: {es-issue}41166[#41166])
* Add IndexOrDocValuesQuery to GeoPolygonQueryBuilder {es-pull}48449[#48449]

Infra/Core::
* Add "did you mean" to ObjectParser {es-pull}50938[#50938]
* Consistent case in CLI option descriptions {es-pull}49635[#49635]
* Improve resiliency to formatting JSON in server {es-pull}48553[#48553] (issue: {es-issue}48450[#48450])
* Don't close stderr under `--quiet` {es-pull}47208[#47208] (issue: {es-issue}46900[#46900])

Infra/Packaging::
* Respect ES_PATH_CONF on package install {es-pull}50158[#50158]
* Restrict support for CMS to pre-JDK 14 {es-pull}49123[#49123] (issue: {es-issue}46973[#46973])
* Remove parsed JVM settings from general settings in Windows service daemon manager {es-pull}49061[#49061] (issue: {es-issue}48796[#48796])
* Package the JDK into jdk.app on macOS {es-pull}48765[#48765]
* Add UBI-based Docker images {es-pull}48710[#48710] (issue: {es-issue}48429[#48429])

Infra/Plugins::
* Report progress of multiple plugin installs {es-pull}51001[#51001] (issue: {es-issue}50924[#50924])
* Allow installing multiple plugins as a transaction {es-pull}50924[#50924] (issue: {es-issue}50443[#50443])

Infra/Scripting::
* Scripting: ScriptFactory not required by compile {es-pull}50344[#50344] (issue: {es-issue}49466[#49466])
* Scripting: Cache script results if deterministic {es-pull}50106[#50106] (issue: {es-issue}49466[#49466])
* Scripting: Groundwork for caching script results {es-pull}49895[#49895] (issue: {es-issue}49466[#49466])
* Scripting: add available languages & contexts API {es-pull}49652[#49652] (issue: {es-issue}49463[#49463])
* Scripting: fill in get contexts REST API {es-pull}48319[#48319] (issue: {es-issue}47411[#47411])
* Scripting: get context names REST API {es-pull}48026[#48026] (issue: {es-issue}47411[#47411])

Infra/Settings::
* Add parameter to make sure that log of updating IndexSetting be more detailed {es-pull}49969[#49969] (issue: {es-issue}49818[#49818])
* Enable dependent settings values to be validated {es-pull}49942[#49942]
* Do not reference values for filtered settings {es-pull}48066[#48066]

License::
* Add max_resource_units to enterprise license {es-pull}50735[#50735]
* Add setting to restrict license types {es-pull}49418[#49418] (issue: {es-issue}48508[#48508])
* Support "enterprise" license types {es-pull}49223[#49223] (issue: {es-issue}48510[#48510])

Machine Learning::
* Add audit warning for 1000 categories found early in job {es-pull}51146[#51146] (issue: {es-issue}50749[#50749])
* Add `num_top_feature_importance_values` param to regression and classification {es-pull}50914[#50914]
* Implement force deleting a data frame analytics job {es-pull}50553[#50553] (issue: {es-issue}48124[#48124])
* Delete unused data frame analytics state {es-pull}50243[#50243]
* Make each analysis report desired field mappings to be copied {es-pull}50219[#50219] (issue: {es-issue}50119[#50119])
* Retry bulk indexing of state docs {es-pull}50149[#50149] (issue: {es-issue}50143[#50143])
* Persist/restore state for data frame analytics classification {es-pull}50040[#50040]
* Introduce `randomize_seed` setting for regression and classification {es-pull}49990[#49990]
* Pass `prediction_field_type` to C++ analytics process {es-pull}49861[#49861] (issue: {es-issue}49796[#49796])
* Add optional source filtering during data frame reindexing {es-pull}49690[#49690] (issue: {es-issue}49531[#49531])
* Add default categorization analyzer definition to ML info {es-pull}49545[#49545]
* Add graceful retry for anomaly detector result indexing failures {es-pull}49508[#49508] (issue: {es-issue}45711[#45711])
* Lower minimum model memory limit value for data frame analytics jobs from 1MB to 1kB {es-pull}49227[#49227] (issue: {es-issue}49168[#49168])
* Improve `model_memory_limit` user experience for data frame analytics jobs {es-pull}44699[#44699]
* Improve performance of boosted tree training for both classification and regression {ml-pull}775[#775]
* Reduce the peak memory used by boosted tree training and fix an overcounting bug
estimating maximum memory usage {ml-pull}781[#781]
* Stratified fractional cross validation for regression {ml-pull}784[#784]
* Added `geo_point` supported output for `lat_long` function records {ml-pull}809[#809], {es-pull}47050[#47050]
* Use a random bag of the data to compute the loss function derivatives for each
new tree which is trained for both regression and classification {ml-pull}811[#811]
* Emit `prediction_probability` field alongside prediction field in ml results {ml-pull}818[#818]
* Reduce memory usage of {ml} native processes on Windows {ml-pull}844[#844]
* Reduce runtime of classification and regression {ml-pull}863[#863]
* Stop early training a classification and regression forest when the validation
error is no longer decreasing {ml-pull}875[#875]
* Emit `prediction_field_name` in data frame analytics results using the type
provided as `prediction_field_type` parameter {ml-pull}877[#877]
* Improve performance updating quantile estimates {ml-pull}881[#881]
* Migrate to use Bayesian optimisation for initial hyperparameter value line
searches and stop early if the expected improvement is too small {ml-pull}903[#903]
* Stop cross-validation early if the predicted test loss has a small chance of
being smaller than for the best parameter values found so far {ml-pull}915[#915]
* Optimize decision threshold for classification to maximize minimum class recall {ml-pull}926[#926]
* Include categorization memory usage in the `model_bytes` field in
`model_size_stats`, so that it is taken into account in node assignment
decisions {ml-pull}927[#927] (issue: {ml-issue}724[#724])

Mapping::
* Add telemetry for flattened fields. {es-pull}48972[#48972]

Network::
* Add certutil http command {es-pull}49827[#49827]
* Do not load SSLService in plugin contructor {es-pull}49667[#49667] (issue: {es-issue}44536[#44536])
* Netty4: switch to composite cumulator {es-pull}49478[#49478]
* Add the simple strategy to cluster settings {es-pull}49414[#49414] (issue: {es-issue}49067[#49067])
* Deprecate misconfigured SSL server config {es-pull}49280[#49280] (issue: {es-issue}45892[#45892])
* Improved diagnostics for TLS trust failures {es-pull}48911[#48911]

Percolator::
* Refactor percolator's QueryAnalyzer to use QueryVisitors {es-pull}49238[#49238] (issue: {es-issue}45639[#45639])

Ranking::
* Support `search_type` in Rank Evaluation API {es-pull}48542[#48542] (issue: {es-issue}48503[#48503])

Recovery::
* Use peer recovery retention leases for indices without soft-deletes {es-pull}50351[#50351] (issues: {es-issue}45136[#45136], {es-issue}46959[#46959])
* Recovery buffer size 16B smaller {es-pull}50100[#50100]

Reindex::
* Reindex sort deprecation warning take 2 {es-pull}49855[#49855] (issue: {es-issue}49458[#49458])

SQL::
* SQL: Handle uberjar scenario where the ES jdbc driver file is bundled in another jar {es-pull}51856[#51856] (issue: {es-issue}50201[#50201])
* SQL: add trace logging for search responses coming from server {es-pull}50530[#50530]
* SQL: Add TRUNC alias for TRUNCATE {es-pull}49571[#49571] (issue: {es-issue}41195[#41195])
* SQL: binary communication implementation for drivers and the CLI {es-pull}48261[#48261] (issue: {es-issue}47785[#47785])
* SQL: Verify Full-Text Search functions not allowed in SELECT {es-pull}51568[#51568] (issue: {es-issue}47446[#47446])


Search::
* Add Validation for maxQueryTerms to be greater than 0 for MoreLikeThisQuery {es-pull}49966[#49966] (issue: {es-issue}49927[#49927])
* Optimize numeric sort on match_all queries {es-pull}49717[#49717] (issue: {es-issue}48804[#48804])
* Pre-sort shards based on the max/min value of the primary sort field {es-pull}49092[#49092] (issue: {es-issue}49091[#49091])
* Optimize sort on long field {es-pull}48804[#48804]
* Search optimisation - add canMatch early aborts for queries on "_index" field {es-pull}48681[#48681] (issue: {es-issue}48473[#48473])
* #48475 Pure disjunctions should rewrite to a MatchNoneQueryBuilder {es-pull}48557[#48557]
* Disable caching when queries are profiled {es-pull}48195[#48195] (issue: {es-issue}33298[#33298])
* BlendedTermQuery's equals method should consider boosts {es-pull}48193[#48193] (issue: {es-issue}48184[#48184])
* Increase the number of vector dims to 2048 {es-pull}46895[#46895]

Security::
* Make .async-search-* a restricted namespace {es-pull}50294[#50294]
* Security should not reload files that haven't changed {es-pull}50207[#50207] (issue: {es-issue}50063[#50063])

Snapshot/Restore::
* Use Cluster State to Track Repository Generation {es-pull}49729[#49729] (issue: {es-issue}49060[#49060])
* Track Repository Gen. in BlobStoreRepository {es-pull}48944[#48944] (issues: {es-issue}38941[#38941], {es-issue}47520[#47520], {es-issue}47834[#47834], {es-issue}49048[#49048])
* Restore from Individual Shard Snapshot Files in Parallel {es-pull}48110[#48110] (issue: {es-issue}42791[#42791])
* Track Shard-Snapshot Index Generation at Repository Root  {es-pull}46250[#46250] (issues: {es-issue}38941[#38941], {es-issue}45736[#45736])

Store::
* mmap dim files in HybridDirectory {es-pull}49272[#49272] (issue: {es-issue}48509[#48509])

Transform::
* Improve force stop robustness in case of an error {es-pull}51072[#51072]
* Add actual timeout in message {es-pull}50140[#50140]
* Automatic deletion of old checkpoints {es-pull}49496[#49496]
* Improve error handling of script errors {es-pull}48887[#48887] (issue: {es-issue}48467[#48467])
* Add `wait_for_checkpoint` flag to stop {es-pull}47935[#47935] (issue: {es-issue}45293[#45293])



[[bug-7.6.0]]
[discrete]
=== Bug fixes

Aggregations::
* Use #name() instead of #simpleName() when generating doc values {es-pull}51920[#51920] (issues: {es-issue}50307[#50307], {es-issue}51847[#51847])
* Fix a sneaky bug in rare_terms {es-pull}51868[#51868] (issue: {es-issue}51020[#51020])
* Support time_zone on composite's date_histogram {es-pull}51172[#51172] (issues: {es-issue}45199[#45199], {es-issue}45200[#45200])
* Fix format problem in composite of unmapped {es-pull}50869[#50869] (issue: {es-issue}50600[#50600])
* SingleBucket aggs need to reduce their bucket's pipelines first {es-pull}50103[#50103] (issue: {es-issue}50054[#50054])
* Avoid precision loss in DocValueFormat.RAW#parseLong {es-pull}49063[#49063] (issue: {es-issue}38692[#38692])
* Fix ignoring missing values in min/max aggregations {es-pull}48970[#48970] (issue: {es-issue}48905[#48905])

Allocation::
* Collect shard sizes for closed indices {es-pull}50645[#50645] (issue: {es-issue}33888[#33888])
* Auto-expand replicated closed indices {es-pull}48973[#48973]
* Ignore dangling indices created in newer versions {es-pull}48652[#48652] (issue: {es-issue}34264[#34264])
* Handle negative free disk space in deciders {es-pull}48392[#48392] (issue: {es-issue}48380[#48380])

Analysis::
* Fix caching for PreConfiguredTokenFilter {es-pull}50912[#50912] (issue: {es-issue}50734[#50734])
* Throw Error on deprecated nGram and edgeNGram custom filters {es-pull}50376[#50376] (issue: {es-issue}50360[#50360])
* _analyze api does not correctly use normalizers when specified {es-pull}48866[#48866] (issue: {es-issue}48650[#48650])

Audit::
* Audit log filter and marker {es-pull}45456[#45456] (issue: {es-issue}47251[#47251])

Authentication::
* Preserve ApiKey credentials for async verification {es-pull}51244[#51244]
* Don't fallback to anonymous for tokens/apikeys {es-pull}51042[#51042] (issue: {es-issue}50171[#50171])
* Populate User metadata with OpenIDConnect collections {es-pull}50521[#50521] (issue: {es-issue}50250[#50250])
* Always return 401 for not valid tokens {es-pull}49736[#49736] (issue: {es-issue}38866[#38866])
* Fix iterate-from-1 bug in smart realm order {es-pull}49473[#49473]
* Remove unnecessary details logged for OIDC {es-pull}48746[#48746]
* Add owner flag parameter to the rest spec {es-pull}48500[#48500] (issue: {es-issue}48499[#48499])

Authorization::
* Fix memory leak in DLS bitset cache {es-pull}50635[#50635] (issue: {es-issue}49261[#49261])
* Validate field permissions when creating a role {es-pull}50212[#50212] (issues: {es-issue}46275[#46275], {es-issue}48108[#48108])
* Validate field permissions when creating a role {es-pull}48108[#48108] (issue: {es-issue}46275[#46275])

CCR::
* CCR should auto-retry rejected execution exceptions {es-pull}49213[#49213]

CRUD::
* Block too many concurrent mapping updates {es-pull}51038[#51038] (issue: {es-issue}50670[#50670])
* Ensure meta and document field maps are never null in GetResult {es-pull}50112[#50112] (issue: {es-issue}48215[#48215])
* Replicate write actions before fsyncing them {es-pull}49746[#49746]
* Do not mutate request on scripted upsert {es-pull}49578[#49578] (issue: {es-issue}48670[#48670])
* Fix Transport Stopped Exception {es-pull}48930[#48930] (issue: {es-issue}42612[#42612])
* Return consistent source in updates {es-pull}48707[#48707]
* Close query cache on index service creation failure {es-pull}48230[#48230] (issue: {es-issue}48186[#48186])

Cluster Coordination::
* Import replicated closed dangling indices {es-pull}50649[#50649]
* Ignore metadata of deleted indices at start {es-pull}48918[#48918]
* Make elasticsearch-node tools custom metadata-aware {es-pull}48390[#48390]

Discovery-Plugins::
* Make EC2 Discovery Cache Empty Seed Hosts List {es-pull}50607[#50607] (issue: {es-issue}50550[#50550])
* Make EC2 Discovery Plugin Retry Requests {es-pull}50550[#50550] (issue: {es-issue}50462[#50462])

Distributed::
* Exclude nested documents in LuceneChangesSnapshot {es-pull}51279[#51279]
* Closed shard should never open new engine {es-pull}47186[#47186] (issues: {es-issue}45263[#45263], {es-issue}47060[#47060])
* Fix meta version of task index mapping {es-pull}50363[#50363] (issue: {es-issue}48393[#48393])

Engine::
* Do not wrap soft-deletes reader for segment stats {es-pull}51331[#51331] (issues: {es-issue}51192[#51192], {es-issue}51303[#51303])
* Account soft-deletes in FrozenEngine {es-pull}51192[#51192] (issue: {es-issue}50775[#50775])
* Account trimAboveSeqNo in committed translog generation {es-pull}50205[#50205] (issue: {es-issue}49970[#49970])
* Greedily advance safe commit on new global checkpoint {es-pull}48559[#48559] (issue: {es-issue}48532[#48532])
* Do not ignore exception when trim unreferenced readers {es-pull}48470[#48470]

Features/Features::
* Fix X-Pack SchedulerEngine Shutdown {es-pull}48951[#48951]

Features/ILM+SLM::
* Fix SLM check for restore in progress {es-pull}50868[#50868]
* Handle failure to retrieve ILM policy step better {es-pull}49193[#49193] (issue: {es-issue}49128[#49128])
* Don't halt policy execution on policy trigger exception {es-pull}49128[#49128]
* Re-read policy phase JSON when using ILM's move-to-step API {es-pull}48827[#48827]
* Don't schedule SLM jobs when services have been stopped {es-pull}48658[#48658] (issue: {es-issue}47749[#47749])
* Ensure SLM stats does not block an in-place upgrade from 7.4 {es-pull}48367[#48367]
* Ensure SLM stats does not block an in-place upgrade from 7.4 {es-pull}48361[#48361]
* Add SLM support to xpack usage and info APIs {es-pull}48096[#48096] (issue: {es-issue}43663[#43663])
* Change policy_id to list type in slm.get_lifecycle {es-pull}47766[#47766] (issue: {es-issue}47765[#47765])

Features/Ingest::
* Fix ignore_missing in CsvProcessor {es-pull}51600[#51600]
* Don't overwrite target field with SetSecurityUserProcessor {es-pull}51454[#51454] (issue: {es-issue}51428[#51428])
* Fix ingest simulate response document order if processor executes async {es-pull}50244[#50244]
* Allow list of IPs in geoip ingest processor {es-pull}49573[#49573] (issue: {es-issue}46193[#46193])
* Do not wrap ingest processor exception with IAE {es-pull}48816[#48816] (issue: {es-issue}48810[#48810])
* Introduce dedicated ingest processor exception {es-pull}48810[#48810] (issue: {es-issue}48803[#48803])

Features/Java High Level REST Client::
* Support es7 node http publish_address format {es-pull}49279[#49279] (issue: {es-issue}48950[#48950])
* Add slices to delete and update by query in HLRC {es-pull}48420[#48420]
* fix incorrect comparison {es-pull}48208[#48208]
* Fix HLRC parsing of CancelTasks response {es-pull}47017[#47017]
* Prevent deadlock by using separate schedulers {es-pull}48697[#48697] (issues: {es-issue}41451[#41451], {es-issue}47599[#47599])

Features/Java Low Level REST Client::
* Improve warning value extraction performance in Response {es-pull}50208[#50208] (issue: {es-issue}24114[#24114])

Features/Monitoring::
* Validate exporter type is HTTP for HTTP exporter {es-pull}49992[#49992] (issues: {es-issue}47246[#47246], {es-issue}47711[#47711], {es-issue}49942[#49942])
* APM system_user {es-pull}47668[#47668] (issues: {es-issue}2708[#2708], {es-issue}40876[#40876])

Geo::
* Guard against null geoBoundingBox {es-pull}50506[#50506] (issue: {es-issue}50505[#50505])
* Geo: Switch generated GeoJson type names to camel case (#50285) {es-pull}50400[#50400] (issue: {es-issue}49568[#49568])
* Geo: Switch generated WKT to upper case {es-pull}50285[#50285] (issue: {es-issue}49568[#49568])
* Fix typo when assigning null_value in GeoPointFieldMapper  {es-pull}49645[#49645]
* Fix handling of circles in legacy geo_shape queries {es-pull}49410[#49410] (issue: {es-issue}49296[#49296])
* GEO: intersects search for geo_shape return wrong result {es-pull}49017[#49017]
* Geo: improve handling of out of bounds points in linestrings {es-pull}47939[#47939] (issue: {es-issue}43916[#43916])

Highlighting::
* Fix invalid break iterator highlighting on keyword field {es-pull}49566[#49566]

Infra/Core::
* Ignore virtual ethernet devices that disappear {es-pull}51581[#51581] (issue: {es-issue}49914[#49914])
* Guess root cause support unwrap {es-pull}50525[#50525] (issue: {es-issue}50417[#50417])
* Allow parsing timezone without fully provided time {es-pull}50178[#50178] (issue: {es-issue}49351[#49351])
* [Java.time] Retain prefixed date pattern in formatter {es-pull}48703[#48703] (issue: {es-issue}48698[#48698])
* Don't drop user's MaxDirectMemorySize flag on jdk8/windows {es-pull}48657[#48657] (issues: {es-issue}44174[#44174], {es-issue}48365[#48365])
* Warn when MaxDirectMemorySize may be incorrect (Windows/JDK8 only issue) {es-pull}48365[#48365] (issue: {es-issue}47384[#47384])
* [Java.time] Calculate week of a year with ISO rules {es-pull}48209[#48209] (issues: {es-issue}41670[#41670], {es-issue}42588[#42588], {es-issue}43275[#43275], {es-issue}43652[#43652])

Infra/Logging::
* Slow log must use separate underlying logger for each index {es-pull}47234[#47234] (issue: {es-issue}42432[#42432])

Infra/Packaging::
* Extend systemd timeout during startup {es-pull}49784[#49784] (issue: {es-issue}49593[#49593])

Infra/REST API::
* Return 400 when handling invalid JSON {es-pull}49552[#49552] (issue: {es-issue}49428[#49428])
* Slash missed in indices.put_mapping url {es-pull}49468[#49468]

Machine Learning::
* Fix 2 digit year regex in find_file_structure {es-pull}51469[#51469]
* Validate classification `dependent_variable` cardinality is at least two {es-pull}51232[#51232]
* Do not copy mapping from dependent variable to prediction field in regression analysis {es-pull}51227[#51227]
* Handle nested and aliased fields correctly when copying mapping {es-pull}50918[#50918] (issue: {es-issue}50787[#50787])
* Fix off-by-one error in `ml_classic` tokenizer end offset {es-pull}50655[#50655]
* Improve uniqueness of result document IDs {es-pull}50644[#50644] (issue: {es-issue}50613[#50613])
* Fix accuracy metric in multi-class confusion matrix {es-pull}50310[#50310] (issue: {es-issue}48759[#48759])
* Fix race condition when stopping a data frame analytics jobs immediately after starting it {es-pull}50276[#50276] (issues: {es-issue}49680[#49680], {es-issue}50177[#50177])
* Apply source query on data frame analytics memory estimation {es-pull}49517[#49517] (issue: {es-issue}49454[#49454])
* Fix r_squared eval when variance is 0 {es-pull}49439[#49439]
* Blacklist a number of prediction field names {es-pull}49371[#49371] (issue: {es-issue}48808[#48808])
* Make data frame analytics more robust for very short-lived analyses {es-pull}49282[#49282] (issue: {es-issue}49095[#49095])
* Fixes potential memory corruption when determining seasonality {ml-pull}852[#852]
* Prevent `prediction_field_name` clashing with other fields in {ml} results {ml-pull}861[#861]
* Include out-of-order as well as in-order terms in categorization reverse searches {ml-pull}950[#950] (issue: {ml-issue}949[#949])

Mapping::
* Ensure that field collapsing works with field aliases. {es-pull}50722[#50722] (issues: {es-issue}32648[#32648], {es-issue}50121[#50121])
* Improve DateFieldMapper `ignore_malformed` handling {es-pull}50090[#50090] (issues: {es-issue}46675[#46675], {es-issue}50081[#50081])
* Annotated text type should extend TextFieldType {es-pull}49555[#49555] (issue: {es-issue}49289[#49289])
* Ensure parameters are updated when merging flattened mappings. {es-pull}48971[#48971] (issue: {es-issue}48907[#48907])

Network::
* Fix TransportMasterNodeAction not Retrying NodeClosedException {es-pull}51325[#51325]

Percolator::
* Correctly handle MSM for nested disjunctions {es-pull}50669[#50669] (issue: {es-issue}50305[#50305])
* Fix query analyzer logic for mixed conjunctions of terms and ranges {es-pull}49803[#49803] (issue: {es-issue}49684[#49684])

Recovery::
* Check allocation id when failing shard on recovery {es-pull}50656[#50656] (issue: {es-issue}50508[#50508])
* Migrate peer recovery from translog to retention lease {es-pull}49448[#49448] (issue: {es-issue}45136[#45136])
* Ignore Lucene index in peer recovery if translog corrupted {es-pull}49114[#49114]

Reindex::
* Reindex and friends fail on RED shards {es-pull}45830[#45830] (issues: {es-issue}42612[#42612], {es-issue}45739[#45739])

SQL::
* SQL: Fix milliseconds handling in intervals {es-pull}51675[#51675] (issue: {es-issue}41635[#41635])
* SQL: Fix ORDER BY YEAR() function {es-pull}51562[#51562] (issue: {es-issue}51224[#51224])
* SQL: change the way unsupported data types fields are handled {es-pull}50823[#50823]
* SQL: Optimisation fixes for conjunction merges {es-pull}50703[#50703] (issue: {es-issue}49637[#49637])
* SQL: Fix issue with CAST and NULL checking. {es-pull}50371[#50371] (issue: {es-issue}50191[#50191])
* SQL: fix NPE for JdbcResultSet.getDate(param, Calendar) calls {es-pull}50184[#50184] (issue: {es-issue}50174[#50174])
* SQL: COUNT DISTINCT returns 0 instead of NULL for no matching docs {es-pull}50037[#50037] (issue: {es-issue}50013[#50013])
* Fix LOCATE function optional parameter handling  {es-pull}49666[#49666] (issue: {es-issue}49557[#49557])
* Fix NULL handling for FLOOR and CEIL functions {es-pull}49644[#49644] (issue: {es-issue}49556[#49556])
* Handle NULL arithmetic operations with INTERVALs {es-pull}49633[#49633] (issue: {es-issue}49297[#49297])
* Fix issue with GROUP BY YEAR() {es-pull}49559[#49559] (issue: {es-issue}49386[#49386])
* Fix issue with CASE/IIF pre-calculating results {es-pull}49553[#49553] (issue: {es-issue}49388[#49388])
* Fix issue with folding of CASE/IIF {es-pull}49449[#49449] (issue: {es-issue}49387[#49387])
* Fix issues with WEEK/ISO_WEEK/DATEDIFF {es-pull}49405[#49405] (issues: {es-issue}48209[#48209], {es-issue}49376[#49376])
* SQL: Fix issue with mins & hours for DATEDIFF {es-pull}49252[#49252]
* SQL: Failing Group By queries due to different ExpressionIds {es-pull}43072[#43072] (issues: {es-issue}33361[#33361], {es-issue}34543[#34543], {es-issue}36074[#36074], {es-issue}37044[#37044], {es-issue}40001[#40001], {es-issue}40240[#40240], {es-issue}41159[#41159], {es-issue}42041[#42041], {es-issue}46316[#46316])

Search::
* Fix upgrade of custom similarity {es-pull}50851[#50851] (issue: {es-issue}50763[#50763])
* Fix NPE bug inner_hits {es-pull}50709[#50709] (issue: {es-issue}50539[#50539])
* Collect results in a local list when notifying partial results {es-pull}49828[#49828] (issue: {es-issue}49778[#49778])
* Fixes a bug in interval filter serialization {es-pull}49793[#49793] (issue: {es-issue}49519[#49519])
* Correctly handle duplicates in unordered interval matching {es-pull}49775[#49775]
* Correct rewritting of script_score query {es-pull}48425[#48425] (issue: {es-issue}48081[#48081])
* Do not throw errors on unknown types in SearchAfterBuilder {es-pull}48147[#48147] (issue: {es-issue}48074[#48074])

Security::
* Always consume the body in has privileges {es-pull}50298[#50298] (issue: {es-issue}50288[#50288])

Snapshot/Restore::
* Fix Overly Aggressive Request DeDuplication {es-pull}51270[#51270] (issue: {es-issue}51253[#51253])
* Guard Repository#getRepositoryData for exception throw  {es-pull}50970[#50970]
* Fix Index Deletion During Partial Snapshot Create {es-pull}50234[#50234] (issues: {es-issue}50200[#50200], {es-issue}50202[#50202])
* Fix Index Deletion during Snapshot Finalization {es-pull}50202[#50202] (issues: {es-issue}45689[#45689], {es-issue}50200[#50200])
* Fix RepoCleanup not Removed on Master-Failover {es-pull}49217[#49217]
* Make FsBlobContainer Listing Resilient to Concurrent Modifications {es-pull}49142[#49142] (issue: {es-issue}37581[#37581])
* Fix SnapshotShardStatus Reporting for Failed Shard {es-pull}48556[#48556] (issue: {es-issue}48526[#48526])
* Cleanup Concurrent RepositoryData Loading {es-pull}48329[#48329] (issue: {es-issue}48122[#48122])

Transform::
* Fix mapping deduction for scaled_float {es-pull}51990[#51990] (issue: {es-issue}51780[#51780])
* Fix stats can return old state information if security is enabled {es-pull}51732[#51732] (issue: {es-issue}51728[#51728])
* Fail to start/put on missing pipeline {es-pull}50701[#50701] (issue: {es-issue}50135[#50135])
* Fix possible audit logging disappearance after rolling upgrade {es-pull}49731[#49731] (issue: {es-issue}49730[#49730])
* Do not fail checkpoint creation due to global checkpoint mismatch {es-pull}48423[#48423] (issue: {es-issue}48379[#48379])



[[upgrade-7.6.0]]
[discrete]
=== Upgrades

Engine::
* Upgrade to Lucene 8.4.0. {es-pull}50518[#50518]

Infra/Packaging::
* Upgrade the bundled JDK to JDK 13.0.2 {es-pull}51511[#51511]