Improperly Controlled Sequential Memory Allocation Affecting org.apache.kafka:generator package, versions [2.4.0,4.2.2)[4.3.0,4.3.2)


Severity

Recommended
0.0
high
0
10

CVSS assessment by Snyk's Security Team. Learn more

Threat Intelligence

EPSS
0.56% (47th percentile)

Do your applications use this vulnerable package?

In a few clicks we can analyze your entire application and see what components are vulnerable in your application, and suggest you quick fixes.

Test your applications
  • Snyk IDSNYK-JAVA-ORGAPACHEKAFKA-22150296
  • published3 Nov 2026
  • disclosed28 Oct 2026
  • creditMufeed VH, Vivek R

Introduced: 28 Oct 2026

NewCVE-2026-111115  (opens in a new tab)
CWE-1325  (opens in a new tab)
CWE-770  (opens in a new tab)

How to fix?

Upgrade org.apache.kafka:generator to version 4.2.2, 4.3.2 or higher.

Overview

Affected versions of this package are vulnerable to Improperly Controlled Sequential Memory Allocation via generated flexible-protocol message readers that parse request headers. An attacker can send a request whose flexible header declares millions of unknown tagged fields, causing the readers to retain a RawTaggedField object for each field and exhaust broker heap or trigger process-wide GC stalls. Because the header is parsed on the network thread before authorization and request-time quota accounting, this can occur before authentication when using pre-authentication APIs such as ApiVersions v3.

Notes

  • At the header-parsing step, the declared field count is bounded only by the outer request frame limit, socket.request.max.bytes; lowering that setting reduces the maximum allocations one request can trigger but does not eliminate the attack.

Workarounds

  • Lower socket.request.max.bytes to limit the size of requests and reduce the number of tagged fields an attacker can declare; this reduces, but does not eliminate, exposure.
  • Restrict network access to broker listeners to trusted clients to reduce the chance that untrusted clients can send malicious request headers.

CVSS Base Scores

version 4.0
version 3.1