---
id: CVE-2026-17507
title: >-
  In Bouncy Castle for Java before 1.86, the MLS implementation
  (org.bouncycastle.mls) holds RFC 9420's uint32 leaf_index in a signed int, so
  a wire value with the top bit set decodes to a negative number
summary: >-
  In Bouncy Castle for Java before 1.86, the MLS implementation
  (org.bouncycastle.mls) holds RFC 9420's uint32 leaf_index in a signed int, so
  a wire value with the top bit set decodes to a negative number. That is a
  legitimate encoding rat…
severity: high
cvss: 7.5
cvssVector: 'CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H'
cwe:
  - CWE-195
vendor: bouncycastle
product: bc-java
affected:
  - 'bc-java >= 1.73, < 1.86'
patched:
  - bc-java 1.86
published: '2026-10-02'
updated: '2026-10-14'
sourceUpdated: '2026-10-14T18:59:38.483'
source: NVD
sourceUrl: 'https://nvd.nist.gov/vuln/detail/CVE-2026-17507'
references:
  - url: >-
      https://github.com/bcgit/bc-java/commit/1eb83471eb01727bfdb944a5a46355cc77d55f44
    label: 91579145-5d7b-4cc5-b925-a0262ff19630
  - url: 'https://github.com/bcgit/bc-java/wiki/CVE%E2%80%902026%E2%80%9017507'
    label: 91579145-5d7b-4cc5-b925-a0262ff19630
  - url: >-
      https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-17507.json
  - url: 'https://access.redhat.com/security/cve/CVE-2026-17507'
  - url: 'https://www.cve.org/CVERecord?id=CVE-2026-17507'
tags:
  - nvd
  - cve.org
  - csaf
  - vex
  - red-hat
epss: 0.0042
epssPercentile: 0.3646
ssvc:
  exploitation: none
  automatable: 'yes'
  technicalImpact: partial
  timestamp: '2026-10-02T17:43:42.891835Z'
scores:
  nvd: 7.5
  cna: 8.7
  vendor: 7.5
ingestedAt: '2026-10-02T08:14:06.224Z'
---

## Overview

In Bouncy Castle for Java before 1.86, the MLS implementation (org.bouncycastle.mls) holds RFC 9420's uint32 leaf_index in a signed int, so a wire value with the top bit set decodes to a negative number. That is a legitimate encoding rather than malformed input, and it must still decode, since the MLS interop test vectors round-trip the full range. GroupKeySet.SecretTree.hasLeaf and Group.validateRemove compared the decoded value directly against the tree's leaf count, and a signed comparison treats any negative int as less than a positive bound, so an out-of-range sender passed the membership check. In the hasLeaf case the SenderData of an unprotected PrivateMessage could then drive LeafIndex.directPath through NodeIndex.parent() arithmetic that never reaches the tree root, growing the resulting node list without bound until the JVM exhausted its heap. A single small message from any current group member could therefore deny service to every other member of the group. Both comparisons now interpret the value as unsigned via Integer.toUnsignedLong, rejecting an out-of-range sender however it was encoded; well-formed leaf indices are unaffected.

## Affected

- `bc-java >= 1.73, < 1.86`

## Remediation

Upgrade past the affected range:

- `bc-java 1.86`

## Vendor advisories

- **Red Hat VEX** · updated 2026-10-02 · [vex](https://security.access.redhat.com/data/csaf/v2/vex/2026/cve-2026-17507.json)