Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain
-
AccessObtain read access to one index
-
DeliveryCraft deeply nested aggregation request
-
ExploitSubmit to search API endpoint
-
ExecutionTrigger unbounded recursive processing
-
ImpactExhaust node resources and crash node
Vulnerability Assessment
| Weakness |
CWE
ASVS 5.0
V15 - Secure Coding and Architecture Recursion depth limits
|
| Exploitation Conditions | Exploitation requires an authenticated account with at least read access to a single index - this is the hard prerequisite and the primary limiting factor (PR:L), so fully unauthenticated attackers cannot trigger it. The attacker must reach the Elasticsearch search API over the network and submit a request to the aggregation feature containing deeply nested aggregation definitions. No user interaction, non-default configuration, or specific platform is required beyond having the aggregation-capable search API exposed to the authenticated principal; the aggregation feature is standard functionality. The main defensive levers are restricting who holds index read/search privileges, network-isolating the cluster from untrusted users, and applying the fixed versions referenced in ESA-2026-196 (8.19.21 / 9.4.6 / 9.5.2). |
| Impact | This is a moderate-severity availability-only flaw whose signals are internally consistent rather than conflicting. The CVSS 3.1 vector (AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H, base 6.5) indicates network reachability and low attack complexity but requires low privileges (PR:L) - the description confirms an authenticated user with read access to just one index, a low bar in many multi-tenant or broadly-provisioned clusters. Impact is confined to availability (C:N/I:N/A:H): a single crafted request can crash a node that does not self-recover, requiring manual intervention. There is no CISA KEV listing and no EPSS score or public POC supplied in the input, so exploitation status is 'no public exploit identified at time of analysis'; the absence of these signals is not evidence of safety but does argue against widespread opportunistic exploitation today. Net assessment: this is a real but mid-tier priority - not a critical RCE, but the low privilege requirement, trivial complexity, and non-self-healing node termination make it a credible DoS risk worth prompt patching in exposed or multi-tenant Elasticsearch deployments, below any RCE/KEV items but above purely theoretical high-CVSS issues. |
| Exploit ScenarioDerived | An authenticated user holding only read access to one index sends a single search API call whose aggregation definition is deeply nested, triggering unbounded recursion in Elasticsearch's aggregation parser. The affected node exhausts its resources and terminates, and because it does not recover automatically, an operator must manually restart it to restore service. No public exploit code was identified in the provided data, though the crafted request is simple to construct given low attack complexity. |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv1 – Oct 06, 2026
Denial of service in Elasticsearch allows an authenticated user holding read access to just a single index to terminate a cluster node by submitting a search request whose aggregation definitions are deeply and maliciously nested. Processing the request drives unbounded recursive execution that exhausts the node's resources and crashes the affected process, which does not recover on its own and requires manual intervention to restore service. The flaw is scored CVSS 6.5 (AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H), is availability-only, and no public exploit code or confirmed active exploitation was identified at time of analysis; the practical barrier is the authenticated read privilege prerequisite rather than remote reachability.
RemediationAI
Apply the vendor-released patch by upgrading to Elasticsearch 8.19.21, 9.4.6, or 9.5.2 as documented in ESA-2026-196 (https://discuss.elastic.co/t/elasticsearch-8-19-21-9-4-6-9-5-2-security-update-esa-2026-196/390870), choosing the fixed release on the branch you currently run. Where immediate upgrade is not possible, the most effective compensating control is to shrink the pool of principals who can reach the search API with read privileges: audit and reduce index read/search roles so that only trusted, non-multi-tenant users can issue arbitrary aggregation queries, accepting that this restricts legitimate analytics and dashboard functionality for the removed roles. Network-isolate the cluster so the search API is not reachable by untrusted or shared tenants, for example by placing Elasticsearch behind a reverse proxy or security group that only admits known application clients, at the cost of breaking access for legitimate remote users until rules are relaxed. Because the node does not recover automatically, configure process supervision and orchestration (systemd restart policies, Kubernetes probes, or cluster-level monitoring and alerting on node loss) so a crashed node is detected and restarted quickly; this shortens outage duration but does not prevent the crash itself. Ensure heap sizing and node monitoring are in place to detect abnormal resource exhaustion early, and treat repeated unexplained node terminations as a signal to prioritize the upgrade.
Affected VersionsENISA
Affected Packages
Source: OSV (Open Source Vulnerabilities)
Affects your stack
Affects 1 package in euvd
Checking this CVE against your own stacks and servers is part of the Basic plan and up.
Technical ContextAI · show details
The weakness is CWE-674, Uncontrolled Recursion, located in Elasticsearch's search aggregation pipeline. Aggregations are a standard, always-on feature of the search API that compile a user-supplied tree of aggregation definitions into nested collector structures; in affected versions there is no effective bound on nesting depth, so a crafted request causes the server to recurse without limit while building or executing the aggregation tree. This is a resource-exhaustion class bug rather than memory corruption or data exposure: recursion and the associated object/stack allocation consume the node's available resources until the JVM process or node terminates. Because the crash takes down the node process rather than being handled gracefully, the cluster does not self-heal and an operator must restart or recover the node. The attack surface is the aggregation-capable search endpoint exposed to any principal with index read privileges, and no user interaction, alternate configuration, or specific platform is required beyond that exposure. The CVSS vector confirms network reachability (AV:N), low attack complexity (AC:L), and a hard authentication prerequisite (PR:L), with high availability impact (A:H) and no confidentiality or integrity impact.
Know when Elasticsearch gets its next vulnerability
- Watch products and get an email when something new is published
- Check this CVE against your own stack and servers
- Exploit and PoC references, threat intelligence from 22 sources
No credit card · 7-day full trial
Severity by source
AV:N/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H
Network-reachable search API (AV:N) with a simple crafted request (AC:L) but requiring authenticated index read access (PR:L); impact is availability-only node crash (C:N/I:N/A:H).
Primary rating from Vendor (elastic).
CVSS VectorVendor: elastic
DescriptionCVE.org
Uncontrolled Recursion (CWE-674) in Elasticsearch can lead to Denial of Service via a specially crafted, deeply nested request submitted to the aggregation feature of the search API. Elasticsearch contains an uncontrolled recursion weakness in its search aggregation processing. An authenticated user with read access to a single index can submit a specially crafted request containing deeply nested aggregation definitions. Processing this request triggers unbounded recursive execution that exhausts the server process's available resources, causing the affected node to terminate. The node does not recover automatically and requires manual intervention to restore service.
Threat Intelligence Overview
How this decision was reached
The engine reads each signal below and applies a fixed list of rules from top to bottom. The first rule that matches sets the recommended action – like a statement, every line is shown.
What every action level means
SSVC AssessmentENISA
Priority Score
Attack Technique
Denial of Service attacks render applications or systems unavailable by overwhelming resources or triggering failure conditions.
More in Elasticsearch
View allThe Groovy scripting engine in Elasticsearch before 1.3.8 and 1.4.x before 1.4.3 allows remote attackers to bypass the s
The default configuration in Elasticsearch before 1.2 enables dynamic scripting, which allows remote attackers to execut
Directory traversal vulnerability in Elasticsearch before 1.6.1 allows remote attackers to read arbitrary files via unsp
Directory traversal vulnerability in Elasticsearch before 1.4.5 and 1.5.x before 1.5.2, when a site plugin is enabled, a
A flaw was discovered in Elasticsearch, affecting the _search API that allowed a specially crafted query string to cause
Maliciously crafted on-chain account names processed by Klever-Go prior to 1.7.20 are spliced into Elasticsearch _bulk r
All versions of Elastic Cloud Enterprise has the Elasticsearch “anonymous” user enabled by default in deployed clusters.
A memory disclosure vulnerability was identified in Elasticsearch 7.10.0 to 7.13.3 error reporting. Rated medium severit
Remote code execution in Elasticsearch's machine learning component allows an authenticated user with ML model deploymen
Heap corruption and potential remote code execution in Elasticsearch's native ML inference engine can be triggered by an
The fix for CVE-2020-7009 was found to be incomplete. Rated high severity (CVSS 8.8), this vulnerability is remotely exp
Elasticsearch versions from 6.7.0 before 6.8.8 and 7.0.0 before 7.6.2 contain a privilege escalation flaw if an attacker
EUVD-2026-93936
GHSA-crr5-288w-ffvf
NVD
CVE.org
EPSS
Vulnrichment
VulDB
Exploit-DB