IMPORTANT: No additional bug fixes or documentation updates
will be released for this version. For the latest information, see the
current release documentation.
7.10 Release notes
edit
IMPORTANT: This documentation is no longer updated. Refer to Elastic's version policy and the latest documentation.
7.10 Release notes
edit7.10.1 Release notes
editClient
edit- Updates for connecting with Cloud.
7.10 Release notes
editClient
edit-
Support for Elasticsearch version
7.10.0. -
Fixes a bug when building the complete endpoint URL could end with duplicate slashes
//. - Fixes a bug when building the complete endpoint URL with cloud id could end with duplicate ports #1081.
API
edit- Fix in RubyDoc comments, some parameters were being duplicated.
-
Deprecation notice: Synced flush (
indices.flush_synced) is deprecated and will be removed in 8.0. Use flush instead.
New API Endpoints
edit-
snapshot.clone
API Changes
edit-
bulk,index,update: new parameterrequire_alias(boolean): When true, requires destination to be an alias (default: false) forindexandupdate. Forbulkit setsrequire_aliasfor all incoming documents. Defaults to unset (false).
X-Pack
editDeprecation notice: searchable_snapshots.repository_stats is deprecated and is replaced by the Repositories Metering API.
New API Endpoints
edit-
close_point_in_time -
open_point_in_time -
security.clear_api_key_cache -
security.grant_api_key
API Changes
edit-
cat.ml_datafeeds,cat.ml_jobs,machine_learning.close_job,machine_learning.get_datafeed_stats,machine_learning.get_datafeeds,machine_learning.get_job_stats,machine_learning.get_jobs,machine_learning.get_overall_buckets,machine_learning.stop_datafeed: new parameterallow_no_match(boolean): Whether to ignore if a wildcard expression matches no datafeeds (this includes_allstring or when no datafeeds have been specified). -machine_learning.get_data_frame_analytics: new parameterverbose(boolean), whether the stats response should be verbose. -
machine_learning.get_trained_models: new parameterinclude(string), a comma-separate list of fields to optionally include. Valid options are definition and total_feature_importance. Default is none. -
machine_learning.stop_datafeed: endpoint now accepts abody: the URL params optionally sent in the body. -
security.get_role,security/get_role_mapping: The name parameter is now a comma-separated list of role-mapping names. -
machine_learning.delete_trained_model,machine_learning.get_trained_models,machine_learning.get_trained_models_stats,machine_learning.put_trained_model: Internal change, url changed from_ml/inferenceto_ml/trained_models.