Wiz rejoint Google Cloud : Faire de la magie ensemble

CVE-2026-132275: 
Python Analyse et atténuation de la vulnérabilité

Aperçu

CVE-2026-132275 is a denial-of-service vulnerability in svglib caused by uncontrolled exponential resource consumption when processing nested SVG <use> references. The svg2rlg() renderer recursively deep-copies referenced subtrees without enforcing an overall expansion limit, allowing a small attacker-controlled SVG to trigger exponential CPU and memory exhaustion. All svglib versions up to and including 2.3.0 are affected, as well as xhtml2pdf workflows that depend on svglib for SVG rendering. The vulnerability was disclosed on October 22, 2026, and fixed in svglib 2.3.1. It carries a CVSS v3.1 base score of 7.5 (High) (GitHub Advisory).

Détails techniques

The root cause is classified as CWE-400 (Uncontrolled Resource Consumption) and CWE-776 (Improper Restriction of Recursive Entity References), analogous to the classic XML "billion laughs" attack. In SvgRenderer.renderUse() within src/svglib/svglib.py, the method calls xlink_href_target() which raises CircularRefError only for direct cycles, but permits acyclic directed acyclic graphs (DAGs) with repeated fan-out. Each <use> element triggers a copy.deepcopy() of the referenced subtree followed by recursive renderNode() calls, causing fan ** levels total node renders — for example, fan=10 and levels=5 yields 100,000 node expansions from a ~1.4 KB SVG. The attack is also reachable via xhtml2pdf's pisa.CreatePDF() when user-supplied HTML contains an embedded data:image/svg+xml image element (GitHub Advisory, Fix Commit).

Impact

Successful exploitation causes CPU and memory exhaustion (denial of service) in any application that renders attacker-controlled SVG content using svglib, including thumbnail/preview generators, SVG-to-PDF converters, and report generation pipelines. Benchmarks show a 5-level payload (~1.4 KB) consuming 469 MiB of RAM and 34 seconds of CPU time on svglib 2.3.0; through xhtml2pdf, the same payload consumes 499 MiB and ~79 seconds. There is no confidentiality or integrity impact — the vulnerability is limited to availability, but can crash or hang web application processes handling user-supplied content (GitHub Advisory).

Exploitabilité

A public proof-of-concept Python script is included in the GitHub Security Advisory, demonstrating the vulnerability with a minimal, non-destructive runnable example. The exploit requires no authentication, no privileges, and no user interaction — only the ability to supply an SVG file to an application using svglib. The vulnerability is classified as automatable by NVD SSVC. There is no evidence of in-the-wild exploitation at this time, and it is not listed in the CISA KEV catalog. The EPSS score is 0.0, reflecting low current exploitation probability (GitHub Advisory).

Étapes d’exploitation

  1. Craft the malicious SVG payload: Construct a small SVG document using a fan-out <use> reference structure. Define a base <rect id="l0"/> and a chain of <g> elements where each level i contains fan copies of <use xlink:href="#l{i-1}"/>. A fan of 10 and 5 levels produces ~1.4 KB of SVG that expands to over 100,000 nodes.
  2. Identify the target: Find an application that accepts user-supplied SVG files and renders them with svglib (version ≤ 2.3.0), such as a thumbnail generator, SVG-to-PDF service, or an xhtml2pdf-based HTML-to-PDF endpoint.
  3. Deliver the payload: Submit the crafted SVG directly to the SVG rendering endpoint, or embed it as a data:image/svg+xml URI within an <img> tag in HTML submitted to an xhtml2pdf pisa.CreatePDF() endpoint.
  4. Trigger exponential expansion: The svg2rlg() renderer processes the <use> elements, deep-copying each referenced subtree recursively. Because only direct cycles are rejected, the fan-out DAG is fully expanded, multiplying node count by fan per level.
  5. Achieve denial of service: CPU and memory consumption grow exponentially with nesting depth, causing the rendering process to hang or crash, denying service to legitimate users (GitHub Advisory).

Indicateurs de compromis

  • Logs: Application logs showing SVG rendering operations taking abnormally long (tens of seconds to minutes) for small input files; error messages containing "Too many expansions (limit is 10000), skipping the rest" in applications patched to 2.3.1 that are still receiving attack payloads.
  • Process: Sudden spike in CPU usage (sustained near 100%) and memory consumption (hundreds of MiB) in the process handling SVG or PDF rendering; process hangs or OOM-killer events associated with svglib or xhtml2pdf worker processes.
  • Network: Repeated HTTP requests submitting small SVG files (under 2 KB) or HTML documents with embedded data:image/svg+xml content to rendering or PDF-generation endpoints, especially with deeply nested <use> and <g> elements referencing prior groups.
  • File System: Presence of small SVG files (< 2 KB) containing multiple <use xlink:href> elements with a fan-out pattern across several nesting levels in upload directories or temporary processing folders (GitHub Advisory).

Atténuation et solutions de contournement

Upgrade svglib to version 2.3.1, which introduces a MAX_USE_EXPANSIONS limit of 10,000 total <use> element expansions per document (shared across external SVG references), implemented via commit 7380d3c (Fix Commit, Release Notes). If immediate patching is not possible, restrict the ability to render untrusted SVG files, disable SVG rendering in user-facing workflows, or implement OS-level resource limits (CPU timeouts, memory caps) around SVG rendering processes. For xhtml2pdf users, validate or sanitize embedded data:image/svg+xml content in HTML input before passing it to pisa.CreatePDF().

Ressources supplémentaires

État de correction de la distribution Linux

Corrigez la disponibilité entre les grandes distributions Linux et leurs versions.

RHEL / CentOS

Inconnu

Source: Ce rapport a été généré à l’aide de l’IA

Connexe Python Vulnérabilités :

CVE ID

Gravité

Partition

Technologies

Nom du composant

Exploit CISA KEV

A corrigé

Date de publication

CVE-2026-132275HIGH7.5
  • Python logoPython
  • svglib
NonOuiNov 20, 2026
CVE-2026-132276MEDIUM5.9
  • Python logoPython
  • wcmatch
NonOuiNov 20, 2026
CVE-2026-132277MEDIUM5.3
  • Python logoPython
  • inscriptis
NonOuiNov 20, 2026
CVE-2026-77332LOW2.3
  • Python logoPython
  • memtomem-stm
NonOuiNov 20, 2026
CVE-2026-77331LOW2
  • Python logoPython
  • memtomem-stm
NonOuiNov 20, 2026

Évaluation gratuite des vulnérabilités

Évaluez votre posture de sécurité cloud

Évaluez vos pratiques de sécurité cloud dans 9 domaines de sécurité afin d’évaluer votre niveau de risque et d’identifier les lacunes dans vos défenses.

Évaluation de la demande

Ressources Wiz supplémentaires

Obtenez une démo personnalisée

Prêt(e) à voir Wiz en action ?

"La meilleure expérience utilisateur que j’ai jamais vue, offre une visibilité totale sur les workloads cloud."
David EstlickRSSI
"Wiz fournit une interface unique pour voir ce qui se passe dans nos environnements cloud."
Adam FletcherChef du service de sécurité
"Nous savons que si Wiz identifie quelque chose comme critique, c’est qu’il l’est réellement."
Greg PoniatowskiResponsable de la gestion des menaces et des vulnérabilités