Elasticsearch version 8.11.2

Elasticsearch version 8.11.2

Also see Breaking changes in 8.11.

Known issues

• High Memory Pressure due to a GC change in JDK 21

  This version of Elasticsearch is bundled with JDK 21. In JDK 21 Preventive GC has been removed. This may lead to increased memory pressure and an increased number of CircuitBreakerExceptions when retrieving large documents under some particular load. (issue: #99592)

  If you needed to explicitly enable Preventive GC for your workload in a previous Elasticsearch version, we recommend you avoid to upgrade to this version, as the settings to enable Preventive GC have been removed from JDK 21.

Security updates

• The 8.11.2 patch release contains a fix for a potential security vulnerability. Please see our security advisory for more details.

Bug fixes

Allocation

• Improve failure handling in ContinuousComputation #102281

Application

• Default run_ml_inference should be true #102151
• [Query Rules] Fix bug where combining the same metadata with text/numeric values leads to error #102891 (issue: #102827)

Cluster Coordination

• Synchronize Coordinator#onClusterStateApplied #100986 (issue: #99023)

Data streams

• [Usage API] Count all the data streams that have lifecycle #102259

ES|QL

• ES|QL: Fix drop of renamed grouping #102282 (issue: #102121)
• ES|QL: Fix layout management for Project #102399 (issue: #102120)
• Fix DISSECT with empty patterns #102580 (issue: #102577)
• Fix leaking blocks in TopN #102715 (issue: #102646)
• Fix leaking blocks in BlockUtils #102716
• Fix memory tracking in TopN.Row #102831 (issues: #100640, #102784, #102790, #102683)

ILM+SLM

• [ILM] Fix downsample to skip already downsampled indices #102250 (issue: #102249)

Infra/Circuit Breakers

• Add more logging to the real memory circuit breaker and lower minimum interval #102396

Ingest Node

• Better processor stat merge #102821

Machine Learning

• Ensure datafeed previews with no start or end time don’t search the cold or frozen tiers #102492
• Recreate the Elasticsearch private temporary directory if it doesn’t exist when an ML job is opened #102599

Mapping

• Fix dense_vector cluster stats indexed_vector_dim_min/max values #102467 (issue: #102416)

Search

• Allow mismatched sort-by field types if there are no docs to sort #102779

Security

• Fix double-completion in SecurityUsageTransportAction #102114 (issue: #102111)

Snapshot/Restore

• Set region for the STS client via privileged calls in AWS SDK #102230 (issue: #102173)
• Simplify BlobStoreRepository idle check #102057 (issue: #101948)

Transform

• Ensure transform updates only modify the expected transform task #102934 (issue: #102933)
• Exclude stack traces from transform audit messages and health #102240

Enhancements

Machine Learning

• Add inference counts by model to the machine learning usage stats #101915

Security

• Upgrade xmlsec to 2.3.4 #102220

Upgrades

Snapshot/Restore

• Upgrade reactor netty http version #102311