Patch nowAll advisoriesPackagesCompaniesWhat changedMethodSubscribe

Security advisories in the AI and data stack

Severe, 6 weeks2979Projects319
doclingGHSA-3cr3-8m4c-fpxw

docling: resource exhaustion

Medium4.3CVE-2026-105747 · Published Oct 6, 2026

### Summary When docling detects the input format of a gzip-compressed tar archive (METS-GBS), and later when the METS-GBS backend opens it, it calls `tarfile.TarFile.getmembers()`. That builds the full member list in memory before the `max_member_count` limit is checked. A small archive with a very large number of empty members therefore makes docling allocate memory in proportion to the member count, and the limit has no effect. ### Details `docling/datamodel/document.py` (format detection) and `docling/backend/mets_gbs_backend.py` both iterate `tar.getmembers()` and count members inside the loop. `getmembers()` reads every header up front. Measured on Python 3.12: an archive of 1,000,000 empty members compresses to about 6.2 MB and makes `getmembers()` hold about 408 MB (about 66 times the input size). Format detection runs for any `application/gzip` input before `allowed_formats` is applied, so the allocation happens even when METS-GBS is not an allowed format. This check was introduced by the fix for CVE-2026-44018 (GHSA-r3xg-rg9j-67fv) in 2.91.0. The eager enumeration it relies on has been there since METS-GBS detection was added in 2.45.0. ### Impact Memory exhaustion...

GitHub advisory

Affected versions

PackageAffectedFixed in
docling
PyPI
>= 2.45.0, < 2.131.02.131.0
Details and references

### Summary When docling detects the input format of a gzip-compressed tar archive (METS-GBS), and later when the METS-GBS backend opens it, it calls `tarfile.TarFile.getmembers()`. That builds the full member list in memory before the `max_member_count` limit is checked. A small archive with a very large number of empty members therefore makes docling allocate memory in proportion to the member count, and the limit has no effect. ### Details `docling/datamodel/document.py` (format detection) and `docling/backend/mets_gbs_backend.py` both iterate `tar.getmembers()` and count members inside the loop. `getmembers()` reads every header up front. Measured on Python 3.12: an archive of 1,000,000 empty members compresses to about 6.2 MB and makes `getmembers()` hold about 408 MB (about 66 times the input size). Format detection runs for any `application/gzip` input before `allowed_formats` is applied, so the allocation happens even when METS-GBS is not an allowed format. This check was introduced by the fix for CVE-2026-44018 (GHSA-r3xg-rg9j-67fv) in 2.91.0. The eager enumeration it relies on has been there since METS-GBS detection was added in 2.45.0. ### Impact Memory exhaustion of the converting process, proportional to the size of the input archive. Confidentiality and integrity are not affected. ### Patches Fixed in docling 2.131.0 by [#4412](https://github.com/docling-project/docling/pull/4412). Format detection and the METS-GBS backend now read archive members one at a time and stop as soon as `max_member_count` is exceeded, including when looking up page files. ### Workarounds Upgrade to 2.131.0. For older versions: Reject gzip or tar inputs before they reach docling when METS-GBS support is not needed, and run conversions of untrusted input with memory limits.

CVSS 3.1
CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:N/I:N/A:L
Severity from
GitHub (reviewed advisory)
Weakness
CWE-409, CWE-770
Also known as
CVE-2026-105747

More docling advisories

All docling
Advisory
Docling: `enable_local_fetch` is not enforced in HTML browser-rendering mode
Medium5.9Oct 6
Docling: Arbitrary local file read via draw:image xlink:href in the OpenDocument backend
MediumOct 5
Docling: Unsafe URI and Path Handling in HTML Backend
High7.1Jun 3
Docling: Potential Path Traversal via LaTeX \includegraphics and \input Commands
Medium5.5Jun 3
Docling: Unsafe XML Entity Expansion in USPTO Patent Backend
High7.5Jun 3
Docling: Unsafe Archive Extraction and XML Parsing in METS-GBS Backend
Medium5.5Jun 3