Elasticsearch 9.3.4 → 9.4.4
9 releases · 0 breaking changes · 20 security fixes · 10 advisories fixed
20security0breaking0deprecated127feature160fix
Vulnerabilities fixed by this upgrade
- 5.4 CVE-2026-102407 Incorrect Authorization (CWE-863) in Elasticsearch can lead to unauthorized data stream modification via Accessing Functionality Not Properly Constrained by ACLs (CAPEC-1). An authenticated user wi…
- 5.4 CVE-2026-78607 Missing Authorization (CWE-862) in the Elasticsearch custom inference service can lead to information disclosure via Privilege Abuse (CAPEC-122). A user holding only inference execution privileges…
- 6.5 CVE-2026-72683 A flaw in Elasticsearch allows an authenticated user with the privileges required to invoke the simulate pipeline API endpoint (https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation…
- 6.5 CVE-2026-63263 Uncontrolled Resource Consumption (CWE-400) in Elasticsearch can lead to denial of service via Exponential Data Expansion (CAPEC-197). An authenticated user may submit a specially crafted query to…
- 6.5 CVE-2026-63144 Uncontrolled Recursion (CWE-674) in Elasticsearch can lead to denial of service via a specially crafted search request submitted by a low-privileged authenticated user. A user with read-level index…
- 6.5 CVE-2026-63140 Reachable Assertion (CWE-617) in Elasticsearch can lead to denial of service via Input Data Manipulation (CAPEC-153). A specially crafted search request containing a null value in a specific query…
- 5.3 CVE-2026-56144 Incorrect Authorization (CWE-863) in Elasticsearch can allow an authenticated user with limited index privileges to exploit insufficient authorization controls in the ingest simulation feature. By…
- 6.5 CVE-2026-56145 Uncontrolled Resource Consumption (CWE-400) in Elasticsearch can lead to denial of service via Excessive Allocation (CAPEC-130). A low-privileged authenticated user with permission to execute EQL s…
- 6.5 CVE-2026-56148 Uncontrolled Recursion (CWE-674) in Elasticsearch can lead to a denial of service via Excessive Allocation (CAPEC-130). An authenticated user can submit a specially crafted query that causes excess…
- 4.9 CVE-2026-56149 Allocation of Resources Without Limits or Throttling (CWE-770) in Elasticsearch can lead to a denial of service via Excessive Allocation (CAPEC-130). A user with elevated privileges can submit a sp…
Breaking changes
None in the release notes.
Security changes
- security Fixes CVE-2026-72683: A flaw in Elasticsearch allows an authenticated user with the privileges required to invoke the simulate pipeline API endpoint (https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation… in 9.4.4
- security Fixes CVE-2026-63263: Uncontrolled Resource Consumption (CWE-400) in Elasticsearch can lead to denial of service via Exponential Data Expansion (CAPEC-197). An authenticated user may submit a specially crafted query to… in 9.4.4
- security Fixes CVE-2026-63144: Uncontrolled Recursion (CWE-674) in Elasticsearch can lead to denial of service via a specially crafted search request submitted by a low-privileged authenticated user. A user with read-level index… in 9.4.4
- security Fixes CVE-2026-63140: Reachable Assertion (CWE-617) in Elasticsearch can lead to denial of service via Input Data Manipulation (CAPEC-153). A specially crafted search request containing a null value in a specific query… in 9.4.4
- security Fixes CVE-2026-56145: Uncontrolled Resource Consumption (CWE-400) in Elasticsearch can lead to denial of service via Excessive Allocation (CAPEC-130). A low-privileged authenticated user with permission to execute EQL s… in 9.4.4
- security Fixes CVE-2026-102407: Incorrect Authorization (CWE-863) in Elasticsearch can lead to unauthorized data stream modification via Accessing Functionality Not Properly Constrained by ACLs (CAPEC-1). An authenticated user wi… in 9.4.4
- security Fixes CVE-2026-78607: Missing Authorization (CWE-862) in the Elasticsearch custom inference service can lead to information disclosure via Privilege Abuse (CAPEC-122). A user holding only inference execution privileges… in 9.4.4
- security Fixes CVE-2026-56144: Incorrect Authorization (CWE-863) in Elasticsearch can allow an authenticated user with limited index privileges to exploit insufficient authorization controls in the ingest simulation feature. By… in 9.4.4
- security Fixes CVE-2026-56148: Uncontrolled Recursion (CWE-674) in Elasticsearch can lead to a denial of service via Excessive Allocation (CAPEC-130). An authenticated user can submit a specially crafted query that causes excess… in 9.4.3
- security Fixes CVE-2026-56149: Allocation of Resources Without Limits or Throttling (CWE-770) in Elasticsearch can lead to a denial of service via Excessive Allocation (CAPEC-130). A user with elevated privileges can submit a sp… in 9.4.3
- security Fixes CVE-2026-72683: A flaw in Elasticsearch allows an authenticated user with the privileges required to invoke the simulate pipeline API endpoint (https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation… in 9.3.8
- security Fixes CVE-2026-63263: Uncontrolled Resource Consumption (CWE-400) in Elasticsearch can lead to denial of service via Exponential Data Expansion (CAPEC-197). An authenticated user may submit a specially crafted query to… in 9.3.8
- security Fixes CVE-2026-63144: Uncontrolled Recursion (CWE-674) in Elasticsearch can lead to denial of service via a specially crafted search request submitted by a low-privileged authenticated user. A user with read-level index… in 9.3.8
- security Fixes CVE-2026-63140: Reachable Assertion (CWE-617) in Elasticsearch can lead to denial of service via Input Data Manipulation (CAPEC-153). A specially crafted search request containing a null value in a specific query… in 9.3.8
- security Fixes CVE-2026-102407: Incorrect Authorization (CWE-863) in Elasticsearch can lead to unauthorized data stream modification via Accessing Functionality Not Properly Constrained by ACLs (CAPEC-1). An authenticated user wi… in 9.3.8
- security Fixes CVE-2026-78607: Missing Authorization (CWE-862) in the Elasticsearch custom inference service can lead to information disclosure via Privilege Abuse (CAPEC-122). A user holding only inference execution privileges… in 9.3.8
- security Fixes CVE-2026-56145: Uncontrolled Resource Consumption (CWE-400) in Elasticsearch can lead to denial of service via Excessive Allocation (CAPEC-130). A low-privileged authenticated user with permission to execute EQL s… in 9.3.7
- security Fixes CVE-2026-56144: Incorrect Authorization (CWE-863) in Elasticsearch can allow an authenticated user with limited index privileges to exploit insufficient authorization controls in the ingest simulation feature. By… in 9.3.7
- security Fixes CVE-2026-56148: Uncontrolled Recursion (CWE-674) in Elasticsearch can lead to a denial of service via Excessive Allocation (CAPEC-130). An authenticated user can submit a specially crafted query that causes excess… in 9.3.6
- security Fixes CVE-2026-56149: Allocation of Resources Without Limits or Throttling (CWE-770) in Elasticsearch can lead to a denial of service via Excessive Allocation (CAPEC-130). A user with elevated privileges can submit a sp… in 9.3.6
Every release in between
securityFixes CVE-2026-72683: A flaw in Elasticsearch allows an authenticated user with the privileges required to invoke the simulate pipeline API endpoint (https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation…
securityFixes CVE-2026-63263: Uncontrolled Resource Consumption (CWE-400) in Elasticsearch can lead to denial of service via Exponential Data Expansion (CAPEC-197). An authenticated user may submit a specially crafted query to…
securityFixes CVE-2026-63144: Uncontrolled Recursion (CWE-674) in Elasticsearch can lead to denial of service via a specially crafted search request submitted by a low-privileged authenticated user. A user with read-level index…
securityFixes CVE-2026-63140: Reachable Assertion (CWE-617) in Elasticsearch can lead to denial of service via Input Data Manipulation (CAPEC-153). A specially crafted search request containing a null value in a specific query…
securityFixes CVE-2026-56145: Uncontrolled Resource Consumption (CWE-400) in Elasticsearch can lead to denial of service via Excessive Allocation (CAPEC-130). A low-privileged authenticated user with permission to execute EQL s…
securityFixes CVE-2026-102407: Incorrect Authorization (CWE-863) in Elasticsearch can lead to unauthorized data stream modification via Accessing Functionality Not Properly Constrained by ACLs (CAPEC-1). An authenticated user wi…
securityFixes CVE-2026-78607: Missing Authorization (CWE-862) in the Elasticsearch custom inference service can lead to information disclosure via Privilege Abuse (CAPEC-122). A user holding only inference execution privileges…
securityFixes CVE-2026-56144: Incorrect Authorization (CWE-863) in Elasticsearch can allow an authenticated user with limited index privileges to exploit insufficient authorization controls in the ingest simulation feature. By…
featureUpgrade OpenTelemetry in ms-graph-authz #151968
featureImplement watchdog timeout for GROK #152851
featureUpdate commons-text library version to match commons-lang3 #152092
featureUpgrade jackson #152163
featureAdd EuroBERT and Jina v5 ops to graph validation allowlist #3015
featureDowngrade log severity for a batch of recoverable errors #2889
featureHarden pytorch_inference with TorchScript model graph validation #3008 (issue: #2890)
featureImprove adherence to memory limits for the bucket gatherer #2848
featureImprove error handling for quantiles state documents #2894
featureImprove handling of invalid JSON state documents #2895
featureImprove messaging for OOM process termination #2841
featureReport the actual memory usage of the autodetect process #2846
featureRestrict file system access for PyTorch models #2851
featureUpdate the PyTorch library to version 2.7.1 #2863
fixAvoid OOMs by accounting for memory during cardinality agg reduction phase #152773 (issue: #150290)
fixFix NPE when aggregation param scripts return null #152152 (issue: #136639)
fixFix Clone API Key silently dropping fields that follow a null expiration #152874
fixFix data stream privilege checking #148179
fixImprove simulate bulk action authz #152148
fixAvoid dense run-length allocation in HLL #152035
fixCall Project.output() only when Project.expressionsResolved() is true #151697 (issue: #149681)
fixFix UnsupportedOperationException for partially unmapped keyword fields with multi-fields #150676 (issue: #150667)
fixFix errors and warnings with spatial functions that parse invalid BBOX #152877 (issue: #152876)
fixResolve NoSuchElementException in ViewUnionAll analyzer retries #152867
fixResolve prefixed searchable-snapshot backing indices to their data stream in METRICS_INFO/TS_INFO #153128
fixTreat empty list query parameters as null #152098 (issue: #152271)
fixFix infinite loop in GeoLineDecomposer dateline crossing checks #152096 (issue: #152066)
fixTranslate tools and tool_choice to Anthropic format for Google Model Garden Anthropic chat completion #152427
fixInclude ingest metadata in self reference checks #152932
fixAutomatically repair ML anomaly results aliases pointing at a .reindexed-v7 index #147688 (issue: #147686)
fixExtend AD results template for reindexed indices and generalize heal beyond job_id mapping #153755 (issue: #147686)
fixFix ML jobs stuck in starting state on Serverless trial projects #150362
+15 more items in the release notessecurityFixes CVE-2026-56148: Uncontrolled Recursion (CWE-674) in Elasticsearch can lead to a denial of service via Excessive Allocation (CAPEC-130). An authenticated user can submit a specially crafted query that causes excess…
securityFixes CVE-2026-56149: Allocation of Resources Without Limits or Throttling (CWE-770) in Elasticsearch can lead to a denial of service via Excessive Allocation (CAPEC-130). A user with elevated privileges can submit a sp…
featureAdd filter capture for ESQL query #149535
featureUpgrade Apache Iceberg to 1.10.2 #151664
feature[Inference] Add requires org.apache.commons.lang3 to module-info #151794
featureAdd query filter logging to SQL and EQL queries #150127
featureAdd EuroBERT and Jina v5 ops to graph validation allowlist #3015
featureBetter error handling regarding quantiles state documents #2894
featureBetter handling of invalid JSON state documents #2895
featureBetter messaging regarding OOM process termination #2841
featureDowngrade log severity for a batch of recoverable errors #2889
featureHarden pytorch_inference with TorchScript model graph validation #3008 (issue: #2890)
featureImprove adherence to memory limits for the bucket gatherer #2848
featureReport the actual memory usage of the autodetect process #2846
featureRestrict file system access for pytorch models #2851
featureUpdate the PyTorch library to version 2.7.1 #2863
fixFix AVLTreeDigest CDF monotonicity for penultimate centroid #151181 (issue: #151012)
fixFix non-monotonic T-Digest centroids caused by floating-point errors #151979
fixAvoid long overflows on AVG by casting to double #148519 (issue: #99575)
fixCarry over synthetic attributes from conversions to the Project above UnionAll #149775 (issue: #149509)
fixFix ValuesReader assertion error #149683
fixFix subquery/fork with inline stats pruning #149783 (issue: #149589)
fixFix wildcard view resolution losing duplicate copies #149418 (issue: #149416)
fixMake subquery with counter type not present in all branches behave more consistently #150704 (issues: #150640, #150641, #150644, #150645)
fixResolve surrogates in union type field resolution before plan serialization #151633 (issue: #151475)
fixPass through includeSourceInError #142335
fixFix bug causing incorrect error on force deleting already deleted model #107188 (issue: #105420)
fixFix flaky CIoManagerTest/testFileIoGood test #3017
fixRevert "[ML] Add extra validation in trained_model creation" #152000
fixDon't allow runtime fields to shadow fields used in index sort #150636 (issue: #149500)
fixFix division of max_docs between slices in reindex #150808
fixFix CLI tests failures due to unconsumed output #145967 (issues: #143646, #143645)
fixGuard against null zeroTermsQuery in cross_fields multi_match #149935 (issue: #149934)
fixIds Query: Throw IllegalArgumentException #151234 (issue: #150305)
fixRun remote child task cancellation as system user #150560
fixFix prefix escape for synthetic id #150433 (issue: #150389)
fixFix GPU SQ format name to match CPU equivalent #149512 (issue: #148975)
featureUpgrade to log4j 2.26.0 #132166 (issue: #132035)
featureAdd EuroBERT and Jina v5 ops to graph validation allowlist #3015
featureBetter error handling regarding quantiles state documents #2894
featureBetter handling of invalid JSON state documents #2895
featureBetter messaging regarding OOM process termination #2841
featureDowngrade log severity for a batch of recoverable errors #2889
featureHarden pytorch_inference with TorchScript model graph validation #3008 (issue: #2890)
featureImprove adherence to memory limits for the bucket gatherer #2848
featureReport the actual memory usage of the autodetect process #2846
featureRestrict file system access for pytorch models #2851
featureUpdate the PyTorch library to version 2.7.1 #2863
featureAdd logging tracking to _xpack/usage #148087
fixAggs: Account aggs uncompressed size on coordinator reduction #148885 (issue: #147190)
fixAggs: Fix auto_date_histogram/date_histogram inside a global agg #147427 (issue: #92570)
fixFix ArrayIndexOutOfBoundsException in date_histogram with hard_bounds outside data #148765 (issue: #148763)
fixFix CCR follow to handle indexing_complete race #145304
fixFix FsBlobContainer.moveBlobAtomic fallback on CIFS filesystems #148777 (issue: #148811)
fixDisallow casting to snapshot-only type #148813 (issue: #147867)
fixFix "optimized incorrectly" error from project reorder #149053 (issue: #148612)
fixFix ArrayIndexOutOfBoundsException in centroid grouping evaluateFinal #148815 (issue: #141318)
fixFix FROM *_logs index name validation #149395 (issue: #146073)
fixFix InvalidMappedField equals/hashCode and lazy error message #146117 (issue: #145907)
fixFix errors in blockloaders when filtering nulls on histograms #148983 (issue: #147854)
fixLucene TermsEnum and PostingsEnum may only be used from acquiring thread #149297
fixNode-reduce driver should not release search contexts on failure #145960 (issue: #145509)
fixNormalize _timeseries to JSON for stored source #148716 (issue: #148414)
fixUse search threadpool for coordination in ES|QL #149034
fixHonor task settings for Amazon Bedrock and Google Vertex chat completion #149268 (issue: #148792)
fix[Inference API] Fix NPE for Inference usage when referencing a embedding task type #149636
fix[Inference API] Fixing potential NPE if InferenceContext is null #148921
fix[Inference API] Use image/jpeg instead of image/jpg for inference validation MIME type #148793
fixPin server-launcher native -march to x86-64-v2 #148542 (issue: #148326)
fixUse toTruncatedString in chunked objects toString #147860 (issue: #143694)
fixMonitor exceptions in processing queries and log them as failures #148362
fixFix flaky CIoManagerTest/testFileIoGood test #3017
fixDefault PromQL index to metrics-* #148489
fixDefault Prometheus discovery APIs to metrics-* #148770
fixFix REPLACE with empty pattern #148616
fixDefault exclude_inference_fields to exclude_vectors at REST #148527 (issue: #146425)
fixFix UOE in CanMatch empty-shards skipped-by-cluster map #148754
+7 more items in the release notesfeatureAdd EuroBERT and Jina v5 ops to graph validation allowlist #3015
featureBetter error handling regarding quantiles state documents #2894
featureBetter handling of invalid JSON state documents #2895
featureBetter messaging regarding OOM process termination #2841
featureDowngrade log severity for a batch of recoverable errors #2889
featureHarden pytorch_inference with TorchScript model graph validation #3008 (issue: #2890)
featureImprove adherence to memory limits for the bucket gatherer #2848
featureReport the actual memory usage of the autodetect process #2846
featureRestrict file system access for pytorch models #2851
featureUpdate the PyTorch library to version 2.7.1 #2863
featureBump bundled JDK to Java 26.0.1 #147424
featureUpdate bundled JDK to Java 26 #146167
featureUpdate elastic-apm-agent-java8 to 1.55.6 #148271
fixUpdate failure store redirect logic to exclude backpressure exceptions #148154
fixBugfix - Block Loader Pushdown + Union Types #147940
fixDisallow empty lists in named params, only #147748 (issue: #147448)
fixES|QL query approximation: enforce minimum number of sampled source rows #147886
fixImplement RerankingInferenceService for Elastic service #148365
fix[Inference API] Fix inference initialization thread exhaustion #147063
fixFix flaky CIoManagerTest/testFileIoGood test #3017
fix[DiskBBQ] Check that precondition should not be overwritten on update #148111 (issue: #148004)
fixFix race in TickerScheduleTriggerEngine by checking watcher to node allocation #147678 (issue: #146874)
featureIt now provides 3x or more better search performance on very restrictive filters (prefilters on centroids)
featureProvides a way to condition non iid vectors (expert API for now)
featureGives more bit options (1, 2, 4, and 7 bits!)
featureMore native code improvements for overall performance
featureAccepts the standard Prometheus range query parameters (query, start, end, step, optional index)
featureTranslates the PromQL expression into an ES|QL PROMQL command and executes it via EsqlQueryAction
featureConverts the columnar ES|QL response into the Prometheus matrix JSON format and returns it to the caller
featureBump heap usage limits for INLINE STATS #144679
featureInject circuit breaker into forked SynonymMapBuilder #144800
featureSupport custom rulesets in analysis-icu/icu-transform plugin #143060
featureAdd Clone API Key endpoint #142633 (issue: #59304)
featureUpdate View CRUD Actions to be Index Actions #141570
feature[Entity Store] Add permissions for Entity Store datastream #145981
featureCPS and project routing support for templated searches #139446
featureDo not mark bulk indexing requests as retried after primary relocations #142157 (issue: #141586)
featureAdd dynamic bloom filter sizing based on document count #141342
featureAdd panama simd implementation of contains function for BinaryDocValuesContainsTermQuery #143922
featureAllow loading BYTE_LENGTH without decompressing Zstd byte ref blocks #141322
featureES819 Binary doc values: compact doc offsets using bit packing #142772
featureEnable large blocks for binary doc values by default. This mainly affects fields of type wildcard, ignored source, values hitting ignore above threshold and ignore malformed numbers and dates. #145216
featureFast codePointCount implementation for BytesRef #140388
featurePush contains binary doc values query down to es819 codec #143898
featureRewrite *substring* wildcard queries to contains term queries for binary doc values keywords #143433
featureTrack bloom filter disk usage in IndexDiskUsageAnalyzer #142106
featureUpgrade zstd to version 1.5.7 #140530
featureUse DirectAccessInput in ZstdDecompressor to avoid intermediate heap copy #145658
featureUse max instead of median for merged bloom filter size #143302
featureCodePointCount implementation using Panama vectors API #140693 (issue: #140567)
featureAdd 'logs.otel' and 'logs.ecs' stream types #141564 (issue: #141040)
featureEnsure DLM only runs one general loop at a time #143883
featureSupport Failure Stores in Cross Cluster Search #139316
featureBatch index creation #144074
featureBatch snapshot update tasks after external change #142091
featureEnsure that synthetic _id is usable after restarts/relocations #138678
featureHealth reports GREEN when provisionally unassigned replica #144773
featureIncrease the per-index limit for merges to half the CPUs #141389
featureOpt-in persistent task reassignment on node shutdown #143306
featureCollect dimensions only once per tsid when downsampling #145089
featureRate calculation for downsampled counters becomes aware of counter resets when the aggregate sampling method is used. #143381 (issue: #136178)
featureUse the tdigest type and compression from TDigest in downsampling #143247
+473 more items in the release notessecurityFixes CVE-2026-72683: A flaw in Elasticsearch allows an authenticated user with the privileges required to invoke the simulate pipeline API endpoint (https://www.elastic.co/docs/api/doc/elasticsearch/operation/operation…
securityFixes CVE-2026-63263: Uncontrolled Resource Consumption (CWE-400) in Elasticsearch can lead to denial of service via Exponential Data Expansion (CAPEC-197). An authenticated user may submit a specially crafted query to…
securityFixes CVE-2026-63144: Uncontrolled Recursion (CWE-674) in Elasticsearch can lead to denial of service via a specially crafted search request submitted by a low-privileged authenticated user. A user with read-level index…
securityFixes CVE-2026-63140: Reachable Assertion (CWE-617) in Elasticsearch can lead to denial of service via Input Data Manipulation (CAPEC-153). A specially crafted search request containing a null value in a specific query…
securityFixes CVE-2026-102407: Incorrect Authorization (CWE-863) in Elasticsearch can lead to unauthorized data stream modification via Accessing Functionality Not Properly Constrained by ACLs (CAPEC-1). An authenticated user wi…
securityFixes CVE-2026-78607: Missing Authorization (CWE-862) in the Elasticsearch custom inference service can lead to information disclosure via Privilege Abuse (CAPEC-122). A user holding only inference execution privileges…
featureUpgrade OpenTelemetry in ms-graph-authz #151968
featureImplement watchdog timeout for GROK #152851
featureUpdate the commons-text library version to match commons.lang3 #152092
featureUpgrade Jackson #152163
featureBetter messaging regarding OOM process termination #2841
featureImprove adherence to memory limits for the bucket gatherer #2848
featureReport the actual memory usage of the autodetect process #2846
featureRestrict file system access for pytorch models #2851
featureUpdate the PyTorch library to version 2.7.1 #2863
fixAvoid out-of-memory errors by accounting for memory during the cardinality aggregation reduction phase #152773 (issue: #150290)
fixFix non-monotonic T-Digest centroids caused by floating-point errors #151979
fixFix null pointer exception when aggregation parameter scripts return null #152152 (issue: #136639)
fixFix data stream privilege checking #148179
fixImprove authorization for the simulate bulk action #152148
fixAvoid dense run-length allocation in HyperLogLog #152035
fixFix infinite loop in GeoLineDecomposer dateline crossing checks #152096 (issue: #152066)
fixTranslate tools and tool_choice to Anthropic format for Google Model Garden Anthropic chat completion #152427
fixInclude ingest metadata in self-reference checks #152932
fixAutomatically repair ML anomaly results aliases pointing at a .reindexed-v7 index #147688 (issue: #147686)
fixExtend the anomaly detection results template for reindexed indices and generalize healing beyond job_id mapping #153755 (issue: #147686)
fixFix ML jobs stuck in starting state on Serverless trial projects #150362
fixRoll reindexed ML state indices in daily maintenance #149555
fixAdd defensive protections when parsing query_string #152385
fixApply search timeout to the DFS phase query rewrite #153479
fixApply search timeout to the query rewrite step #153082 (issue: #104187)
fixFork ensureShardSearchActive callbacks off the refresh thread #152509 (issue: #97280)
fixReturn a 400 (Bad Request) for invalid tokens in a script query #152476
fixTrip request breaker on large completion suggest size/shardSize #152552
fixFix snapshot creation on CIFS shares when an access-denied exception occurs #153219 (issue: #152053)
fixIgnore OTLP number data points without a value #152733
fixTighten retry startup logic #152803
securityFixes CVE-2026-56145: Uncontrolled Resource Consumption (CWE-400) in Elasticsearch can lead to denial of service via Excessive Allocation (CAPEC-130). A low-privileged authenticated user with permission to execute EQL s…
securityFixes CVE-2026-56144: Incorrect Authorization (CWE-863) in Elasticsearch can allow an authenticated user with limited index privileges to exploit insufficient authorization controls in the ingest simulation feature. By…
feature[Inference] Add requires org.apache.commons.lang3 to module-info #151794
featureBetter messaging regarding OOM process termination #2841
featureImprove adherence to memory limits for the bucket gatherer #2848
featureReport the actual memory usage of the autodetect process #2846
featureRestrict file system access for pytorch models #2851
featureUpdate the PyTorch library to version 2.7.1 #2863
fixResolve surrogates in union type field resolution before plan serialization #151633 (issue: #151475)
fixPass through includeSourceInError #142335
fixFix bug causing incorrect error on force deleting already deleted model #107188 (issue: #105420)
fixRevert "[ML] Add extra validation in trained_model creation" #152000
securityFixes CVE-2026-56148: Uncontrolled Recursion (CWE-674) in Elasticsearch can lead to a denial of service via Excessive Allocation (CAPEC-130). An authenticated user can submit a specially crafted query that causes excess…
securityFixes CVE-2026-56149: Allocation of Resources Without Limits or Throttling (CWE-770) in Elasticsearch can lead to a denial of service via Excessive Allocation (CAPEC-130). A user with elevated privileges can submit a sp…
featureUpdate the built-in kibana_system role to grant manage, create_index, read, index, write, and delete privileges on the axonius.alert_finding and axonius.incident indices, replacing the previous axonius.alert_and_incident index #149797
feature[otel-data] Explicitly map http.response.status_code as long #149631
featureUpgrade commons-lang3 version for the inference plugin #150242
featureBetter messaging regarding OOM process termination #2841
featureImprove adherence to memory limits for the bucket gatherer #2848
featureReport the actual memory usage of the autodetect process #2846
featureRestrict file system access for pytorch models #2851
featureUpdate the PyTorch library to version 2.7.1 #2863
featureUpgrade netty to 4.1.135.Final #151099
featureOptimize literal action automaton builds #151093 (issues: #123872, #105723)
fixFix Nynorsk stemmer UnsupportedOperationException. The light_nynorsk and minimal_nynorsk stemmers no longer fail with an HTTP 500 error during index creation due to passing an immutable map to Lucene's analysis factory. #150345
fixMove SAML metadata resolution to background thread #144381 (issue: #138031)
fixCatch StackOverflowError in deeply nested RLIKE patterns #150238 (issue: #149838)
fixEnforce limit for max nested functions #149971
fixFix wrong warning in expressions with unrolled multivalues #145968
fix[ESQL] Refactor Greatest and Least functions to use evaluator map #128429 (issue: #114036)
fixTruncate error for ILM's step_info instead of string #150413
fixValidate inference embedding model before checking for existing uses #150150 (issues: #147062, #150084)
fixWhen a streaming inference request is canceled, immediately release apache client networking resources #149987
fix[Inference API] Fix Streaming publisher shutdown race condition #150789 (issue: #150742)
fixDon't apply time zones to epoch-based timestamps #148663
fixFilter _source field names by code point #151146
fixFix OS stats for cgroup paths containing colons #151095
fixAdd extra validation in trained_model creation #150227
fixEQL/SQL: Enforce max expression depth checks #150003
fixAdd IT covering search during rolling restart #149667 (issue: #86927)
fixMake Automatons wildcard code-point aware #151143
fixFix OTLP histogram handling for single-count histograms without bucket boundaries #151411
fixHonor ClusterHealth timeout when waiting for transform internal index shards #149462 (issue: #149400)
featureUpgrade to log4j 2.26.0 #132166 (issue: #132035)
featureBetter messaging regarding OOM process termination #2841
featureImprove adherence to memory limits for the bucket gatherer #2848
featureReport the actual memory usage of the autodetect process #2846
featureRestrict file system access for pytorch models #2851
featureUpdate the PyTorch library to version 2.7.1 #2863
featureBump bundled JDK to Java 26.0.1 #147424
featureAdd logging tracking to _xpack/usage #148087
featureUpdate elastic-apm-agent-java8 to 1.55.6 #148271
fixAggs: Account aggs uncompressed size on coordinator reduction #148885 (issue: #147190)
fixAggs: Fix auto_date_histogram/date_histogram inside a global agg #147427 (issue: #92570)
fixFix ArrayIndexOutOfBoundsException in date_histogram with hard_bounds outside data #148765 (issue: #148763)
fixFix CCR follow to handle indexing_complete race #145304
fixUpdate failure store redirect logic to exclude backpressure exceptions #148154
fixFix FsBlobContainer.moveBlobAtomic fallback on CIFS filesystems #148777 (issue: #148811)
fixFix race in FsBlobContainer.moveBlobAtomic by replacing move op with hard link #147405 (issue: #147171)
fixFix management of PIT close failures after successful query #146634 (issues: #146263, #146187)
fixDisallow empty lists in named params, only #147748 (issue: #147448)
fixDo not enable doc-partitioning for count #143544 (issue: #134512)
fixFix ArrayIndexOutOfBoundsException in centroid grouping evaluateFinal #148815 (issue: #141318)
fixFix FROM *_logs index name validation #149395 (issue: #146073)
fixFix _index LIKE with ? wildcard #147462 (issue: #146364)
fixFix deriv aggregation function intermediate state #146160 (issue: #141305)
fixFix errors in blockloaders when filtering nulls on histograms #148983 (issue: #147854)
fixFix nullify under join and enrich #145743 (issue: #141827)
fixNode-reduce driver should not release search contexts on failure #145960 (issue: #145509)
fixNormalize _timeseries to JSON for stored source #148716 (issue: #148414)
fixReport correct location for SET errors #147582 (issue: #145873)
fixRestore TEXT field size estimate for Enrich and LookupJoin #147357
fixUse search threadpool for coordination in ES|QL #149034
fixHonor task settings for Amazon Bedrock and Google Vertex chat completion #149268 (issue: #148792)
fixImplement RerankingInferenceService for Elastic service #148365
fixMake similarity required for SageMaker ElasticTextEmbeddingPayload #146885
fix[Inference API] Fix inference initialization thread exhaustion #147063
fix[Inference API] Fixing potential NPE if InferenceContext is null #148921
fix[Inference API] Use image/jpeg instead of image/jpg for inference validation MIME type #148793
fixFix potential OOME on SearchResponse#toString() for extemely large responses #147267
fixUse toTruncatedString in chunked objects toString #147860 (issue: #143694)
fixPreserve zero-allocation trained-model deployments in cluster state during ML-node shutdown rebalances #146910 (issue: #146806)
fixFix REPLACE with empty pattern #148616
+13 more items in the release notes