---
id: "CVE-2026-116128"
kind: "vulnerability"
aliases: ["CVE-2026-116128"]
title: "Starlet (Perl): unbounded chunked request buffering exhausts server memory"
severity: "high"
cvss: null
epss: null
exploited: false
status: "open"
category: "package"
products: []
ecosystems: []
tags: ["package", "perl", "psgi", "self-hosted", "starlet"]
not_affected_if: ["no-internet-exposure", "not-starlet", "version-0.33-or-later"]
published_at: "2026-10-30T13:18:31+00:00"
first_seen_at: "2026-10-30T13:30:27+00:00"
updated_at: "2026-10-30T13:32:32+00:00"
sources: [{"source": "nvd", "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-116128", "first_seen_at": "2026-10-30T13:30:27+00:00"}]
---

# Starlet (Perl): unbounded chunked request buffering exhausts server memory

Starlet, a Perl PSGI web server, buffers chunked request bodies without limit when a chunk-size line never ends. A remote client can make a worker run out of memory with a single request.

_Summary, check and fix written by a model from the sources; labelled as such._

## How to check

- Check the installed Starlet version with `perl -MStarlet -e 'print $Starlet::VERSION'` or `cpanm --info Starlet`.
- Look for Starlet in `cpanfile`, `cpanfile.snapshot` or `Makefile.PL` of any Perl web app.
- Confirm whether the app is reachable from the internet and accepts chunked request bodies.

## Fix

- Starlet versions from `0.20` before `0.33` are affected; move to `0.33` or later.

## Timeline

- 2026-10-30T13:30:27+00:00 nvd https://nvd.nist.gov/vuln/detail/CVE-2026-116128

## From the sources

Starlet versions from 0.20 before 0.33 for Perl allow memory exhaustion via unbounded buffering when decoding chunked request bodies.

The decoder adds each read to one buffer of unconsumed input. Nothing limits the length of a chunk-size line, so a line without a CRLF grows the buffer for as long as the client keeps sending, and the read timeout restarts after every read. The data of a large chunk is also held in the buffer until the whole chunk has arrived.

A remote attacker can exhaust a worker's memory with a single request.

---

_Free to read at a human pace. An API with keys and an MCP server are in preparation; early access: hello@infrastack.ai_