Your RSA-2048 keys break in 2030. Find every one of them before attackers do.
🐍
🐍 PyPI
Not in CISA KEV
MEDIUM severity

GHSA-fhfc-2q7x-929f — tensorflow

MEDIUMFix: tensorflow/tensorflow@c1f4918

GHSA-fhfc-2q7x-929f is a medium-severity (CVSS 5.9) Reachable Assertion vulnerability in tensorflow. A fix is available for tensorflow — see the affected versions and patch details below.

TensorFlow vulnerable to `CHECK` fail in `CollectiveGather`

Also known asBIT-tensorflow-2022-35994CVE-2022-35994PYSEC-2026-3172PYSEC-2026-3320PYSEC-2026-990
Published
Sep 16, 2022
Updated
Jul 13, 2026
Affected
9 pkgs
Patched
9 / 9
Exploits
None indexed
Exploitation data as of Sep 26, 2026 · OSV.dev, NVD, FIRST.org (EPSS)

EPSS Exploitation Probability

via FIRST.org ↗
0.5%probability of exploitation in next 30 days
Lower Risk0.00%
Lower risk than most CVEs39th percentile — riskier than 39% of all scored CVEsHighest risk

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

How urgent is this, really

GHSA-fhfc-2q7x-929f by exploitation likelihood (EPSS) against impact (CVSS). Outside the shaded patch-first corner.

Where this sits among everything scored

Of 379,842 CVEs with a current EPSS score, this one falls in the < 10% band (highlighted). Counts from FIRST.org, log-scaled.

Real-World Exposure

9 pkgs affected
🐍tensorflow🐍tensorflow🐍tensorflow🐍tensorflow-cpu🐍tensorflow-cpu🐍tensorflow-cpu🐍tensorflow-gpu🐍tensorflow-gpu+1 more

Real-time download stats are indexed for npm and PyPI packages. This vulnerability affects PyPI packages — download data is not available via public APIs for these ecosystems.

Description

Impact

When CollectiveGather receives an scalar input input, it gives a CHECK fails that can be used to trigger a denial of service attack.

import tensorflow as tf
arg_0=1
arg_1=1
arg_2=1
arg_3=1
arg_4=(3, 3,3)
arg_5='auto'
arg_6=0
arg_7=''
tf.raw_ops.CollectiveGather(input=arg_0, group_size=arg_1, group_key=arg_2,
                            instance_key=arg_3, shape=arg_4,
                            communication_hint=arg_5, timeout_seconds=arg_6, name=arg_7)

Patches

We have patched the issue in GitHub commit c1f491817dec39a26be3c574e86a88c30f3c4770.

The fix will be included in TensorFlow 2.10.0. We will also cherrypick this commit on TensorFlow 2.9.1, TensorFlow 2.8.1, and TensorFlow 2.7.2, as these are also affected and still in supported range.

For more information

Please consult our security guide for more information regarding the security model and how to contact us with issues and questions.

Attribution

This vulnerability has been reported by 刘力源, Information System & Security and Countermeasures Experiments Center, Beijing Institute of Technology.

Affected Packages

9 total 9 fixed
EcosystemPackageVulnerable rangeFix
🐍PyPItensorflowall versions2.7.2pip install --upgrade 'tensorflow==2.7.2'
🐍PyPItensorflow≥ 2.8.0&&< 2.8.12.8.1pip install --upgrade 'tensorflow==2.8.1'
🐍PyPItensorflow≥ 2.9.0&&< 2.9.12.9.1pip install --upgrade 'tensorflow==2.9.1'
🐍PyPItensorflow-cpuall versions2.7.2pip install --upgrade 'tensorflow-cpu==2.7.2'
🐍PyPItensorflow-cpu≥ 2.8.0&&< 2.8.12.8.1pip install --upgrade 'tensorflow-cpu==2.8.1'
🐍PyPItensorflow-cpu≥ 2.9.0&&< 2.9.12.9.1pip install --upgrade 'tensorflow-cpu==2.9.1'

Affected Products

1 product · 7 configurations
Application
tensorflowgoogle
≥ 2.9.0 && < 2.9.1
1 version
2.10

Detection & mitigation playbook

Open-source dependency
  1. Detect

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

  2. Fix

    Update tensorflow to 2.7.2 or later, then make sure no transitive (indirect) dependency still pins the vulnerable range — O3 confirms GHSA-fhfc-2q7x-929f 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.

Frequently Asked Questions

### Impact When `CollectiveGather` receives an scalar input `input`, it gives a `CHECK` fails that can be used to trigger a denial of service attack. ```python import tensorflow as tf arg_0=1 arg_1=1 arg_2=1 arg_3=1 arg_4=(3, 3,3) arg_5='auto' arg_6=0 arg_7='' tf.raw_ops.CollectiveGather(input=arg_0, group_size=arg_1, group_key=arg_2, instance_key=arg_3, shape=arg_4, communication_hint=arg_5, timeout_seconds=arg_6, name=arg_7) ``` ### Patches We have patched the issue in GitHub commit [c1f491817dec39a26be3c574e86a88c30f3c4770](https://gi
O3 Security · Impact-Aware SCA

Is GHSA-fhfc-2q7x-929f in your dependencies?

Find it across PyPI, including transitive dependencies.