---
id: CVE-2026-118391
title: >-
  Unbounded query cache growth from interpolated JSON paths and splice lists in
  Ecto
summary: >-
  Allocation of Resources Without Limits or Throttling vulnerability in
  elixir-ecto ecto allows an unauthenticated remote attacker to exhaust memory
  in the application and in the database server by sending requests with
  distinct JSON keys …
severity: high
cvss: 8.9
cvssVector: 'CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:N/VI:N/VA:H/SC:N/SI:N/SA:H'
cvssSource: cna
cwe:
  - CWE-770
vendor: elixir-ecto
product: ecto
affected:
  - ecto >= 3.4.0 < 3.14.3
  - >-
    elixir-ecto/ecto >= cb2f03964dac324a9771f84f5088e395326e6de8 <
    eef014ad090d27d1a5b69c0e32a50eabd44ab3bb
ssvc:
  exploitation: none
  automatable: 'yes'
  technicalImpact: partial
  timestamp: '2026-11-04T20:48:32.374051Z'
published: '2026-11-04'
updated: '2026-11-05'
sourceUpdated: '2026-11-05T19:07:34.294Z'
source: CVEORG
sourceUrl: 'https://www.cve.org/CVERecord?id=CVE-2026-118391'
references:
  - url: >-
      https://github.com/elixir-ecto/ecto/security/advisories/GHSA-cxjx-vmm2-qr8r
    label: GHSA-cxjx-vmm2-qr8r
  - url: 'https://cna.erlef.org/cves/CVE-2026-118391.html'
    label: EEF CNA record for CVE-2026-118391
  - url: 'https://osv.dev/vulnerability/EEF-CVE-2026-118391'
    label: OSV record EEF-CVE-2026-118391
  - url: >-
      https://github.com/elixir-ecto/ecto/commit/d0e97d8a6e274c81c25fbf08087da8d8b9bae0ea
    label: >-
      Do not cache queries with interpolated JSON paths, splices, and constants
      (main)
  - url: 'https://ecto.hexdocs.pm/Ecto.Query.API.html#json_extract_path/2'
    label: Ecto.Query.API.json_extract_path/2
  - url: 'https://ecto.hexdocs.pm/Ecto.Query.API.html#splice/1'
    label: Ecto.Query.API.splice/1
  - url: >-
      https://github.com/elixir-ecto/ecto/commit/eb03f45b999e2bf67ffae92811233eb6e56dba55
    label: 'Add splice to fragment (#4228), v3.11.0'
  - url: 'https://github.com/elixir-ecto/ecto/issues/4470'
    label: Should we allow disabling of Ecto cache on specific queries? (#4470)
  - url: 'https://github.com/elixir-ecto/ecto/issues/1031'
    label: String interpolation will grow cached unboundedly (#1031)
  - url: >-
      https://github.com/elixir-ecto/ecto/commit/cb2f03964dac324a9771f84f5088e395326e6de8
    label: Introducing commit cb2f039 in elixir-ecto/ecto
  - url: >-
      https://github.com/elixir-ecto/ecto/commit/eef014ad090d27d1a5b69c0e32a50eabd44ab3bb
    label: Fix commit eef014a in elixir-ecto/ecto
tags:
  - cve.org
epss: 0.00494
epssPercentile: 0.42666
ingestedAt: '2026-11-05T22:38:31.508Z'
---

## Overview

Allocation of Resources Without Limits or Throttling vulnerability in elixir-ecto ecto allows an unauthenticated remote attacker to exhaust memory in the application and in the database server by sending requests with distinct JSON keys or list lengths where the application interpolates them into json_extract_path/2, field[^key] or splice/1. Ecto caches prepared queries per repository in an ETS table with no size limit and no eviction. Interpolated JSON path elements become part of the query expression instead of its parameters, and splice/1 expands into one placeholder per element, so every distinct key or list length creates a new cache entry. With ecto_sql each entry is also a new prepared statement on the database connection that runs it. None of this memory is released until the processes restart.

This issue affects ecto: from 3.4.0 before 3.14.3.

## Affected

- `ecto >= 3.4.0 < 3.14.3`
- `elixir-ecto/ecto >= cb2f03964dac324a9771f84f5088e395326e6de8 < eef014ad090d27d1a5b69c0e32a50eabd44ab3bb`

## Remediation

Upgrade Ecto to 3.14.3 or later. Queries with interpolated JSON path elements or an interpolated splice/1 list are no longer cached, the same as running them with query_cache: false, so distinct keys and list lengths don't add cache entries.

### Workarounds

Pass the path as a query parameter so it is not part of the cache key. With PostgreSQL: fragment("? #> ?", i.meta, type(^path, {:array, :string})). Alternatively, disable the cache for these queries with Repo.all(query, query_cache: false), or restrict keys and list lengths to a known set before building the query.