CVE-2020-28975: Scikit learn Denial of Service Vulnerability

HIGH
Advisory Identifier: GHSA-jxfp-4rvq-9h9m
Share
CVSS 7.5
Published
Advisory Summary

CVE-2020-28975 is a HIGH (CVSS 7.5) denial of service vulnerability (General Security Weakness) affecting Scikit learn libraries in the PyPI ecosystem. The issue involves scikit-learn denial of service. Remediation is available by upgrading affected packages to version 1.0.1.

CVE-2020-28975 at a Glance

CVE Identifier CVE-2020-28975
GHSA Identifier GHSA-jxfp-4rvq-9h9m
Severity HIGH
CVSS v3.1 Base Score 7.5
CWE Classification General Security Weakness
Affected Product Scikit learn
Package Ecosystem PyPI
Published Date
Last Updated

Security Impact

CVSS: 7.5

The security impact metrics below are parsed directly from the CVSS v3.1 vector string (CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H).

Attack Vector Network (AV:N)
Attack Complexity Low (AC:L)
Privileges Required None (PR:N)
User Interaction None (UI:N)
Scope Unchanged (S:U)
Confidentiality None (C:N)
Integrity None (I:N)
Availability High (A:H)

How to Fix CVE-2020-28975

To remediate this vulnerability, upgrade all affected Scikit learn dependencies to a patched release. Both direct and transitive dependencies should be checked, as frameworks or security libraries may pull in vulnerable versions transitively.

scikit-learn fixed: 1.0.1 
pip install scikit-learn>=1.0.1
Verify Direct and Transitive Dependencies

Run the following command in your project directory to verify that all resolved dependencies use the patched release: 

pip show scikit-learn

How to Check if You Are Affected

To determine whether your Python environment is affected by CVE-2020-28975, inspect your installed packages using pip:

pip show scikit-learn

Analyze the terminal output to locate any version of Scikit learn in your dependency tree. If any resolved version matches the vulnerable ranges listed in the table below (for example, versions earlier than 1.0.1), your application is affected.

Note: The pip show scikit-learn command checks currently installed packages in your active Python environment.

Technical Details

Vulnerability Mechanism & Technical Analysis

This vulnerability is classified under General Security Weakness. In Scikit learn, software state validation or input boundaries are circumvented during processing. Review the official source advisory statement below for specific affected components, classes, and reproduction details.

Source Advisory Statement

svm_predict_values in svm.cpp in Libsvm v324, as used in scikit-learn 0.23.2 and other products, allows attackers to cause a denial of service (segmentation fault) via a crafted model SVM (introduced via pickle, json, or any other model permanence standard) with a large value in the _n_support array. NOTE: the scikit-learn vendor's position is that the behavior can only occur if the library's API is violated by an application that changes a private attribute.

CVE-2020-28975 Affected Scikit learn Versions

1 component(s)
Fixed Versions:
1.0.1
Product / Package Ecosystem Affected Versions Fixed Version
scikit-learn
pypi/scikit-learn
PyPI
≥ 0.23.2
< 1.0.1
1.0.1

References

NVD — CVE-2020-28975 NVD
https://nvd.nist.gov/vuln/detail/CVE-2020-28975
Scikit learn Source Repository REPOSITORY
https://github.com/scikit-learn/scikit-learn/issues/18891
Security Fix Commit FIX COMMIT
https://github.com/scikit-learn/scikit-learn/commit/1bf13d567d3cd74854aa8343fd25b61dd768bb85
Scikit learn Source Repository REPOSITORY
https://github.com/cjlin1/libsvm/blob/9a3a9708926dec87d382c43b203f2ca19c2d56a0/svm.cpp#L2501
Scikit learn Source Repository REPOSITORY
https://github.com/pypa/advisory-database/tree/main/vulns/scikit-learn/PYSEC-2020-108.yaml
Scikit learn Source Repository REPOSITORY
https://github.com/scikit-learn/scikit-learn
Scikit learn Security Information SECURITY
https://security.gentoo.org/glsa/202301-03
Scikit learn Security Information SECURITY
http://packetstormsecurity.com/files/160281/SciKit-Learn-0.23.2-Denial-Of-Service.html
seclists.org/fulldisclosure/2020... WEB
http://seclists.org/fulldisclosure/2020/Nov/44

Frequently Asked Questions

What is CVE-2020-28975?
CVE-2020-28975 is a security vulnerability in Scikit learn involving scikit-learn Denial of Service under General Security Weakness.
Which Scikit learn versions are affected?
Affected versions include releases prior to 1.0.1 across affected packages: scikit-learn.
What version fixes CVE-2020-28975?
Upgrading to 1.0.1 (or the latest patched release for your respective branch) resolves the vulnerability.
Is scikit-learn affected?
Yes, scikit-learn is affected in releases prior to 1.0.1. Upgrade your dependency to 1.0.1 or later.
What is the CVSS score of CVE-2020-28975?
CVE-2020-28975 has a CVSS v3.1 base score of 7.5 (HIGH) with vector string CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H.
What CWE is associated with CVE-2020-28975?
CVE-2020-28975 is classified under General Security Weakness.
How can I check PyPI dependencies for CVE-2020-28975?
Run the dependency tree command: 'pip show scikit-learn'. This command lists direct and transitive dependencies in your build configuration so you can verify if a vulnerable version is resolved.
Machine-readable OSV JSON OSV v1.5 Schema
Normalized schema output
View Raw JSON 
{
  "schema_version" : "1.4.0",
  "id" : "GHSA-jxfp-4rvq-9h9m",
  "modified" : "2024-10-23T18:41:25Z",
  "published" : "2022-05-24T17:34:40Z",
  "aliases" : [ "CVE-2020-28975" ],
  "summary" : "scikit-learn Denial of Service",
  "details" : "svm_predict_values in svm.cpp in Libsvm v324, as used in scikit-learn 0.23.2 and other products, allows attackers to cause a denial of service (segmentation fault) via a crafted model SVM (introduced via pickle, json, or any other model permanence standard) with a large value in the _n_support array.\nNOTE: the scikit-learn vendor's position is that the behavior can only occur if the library's API is violated by an application that changes a private attribute.",
  "severity" : [ {
    "type" : "CVSS_V3",
    "score" : "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H"
  } ],
  "affected" : [ {
    "package" : {
      "ecosystem" : "PyPI",
      "name" : "scikit-learn"
    },
    "ranges" : [ {
      "type" : "ECOSYSTEM",
      "events" : [ {
        "introduced" : "0.23.2"
      }, {
        "fixed" : "1.0.1"
      } ]
    } ]
  } ],
  "references" : [ {
    "type" : "ADVISORY",
    "url" : "https://nvd.nist.gov/vuln/detail/CVE-2020-28975"
  }, {
    "type" : "WEB",
    "url" : "https://github.com/scikit-learn/scikit-learn/issues/18891"
  }, {
    "type" : "WEB",
    "url" : "https://github.com/scikit-learn/scikit-learn/commit/1bf13d567d3cd74854aa8343fd25b61dd768bb85"
  }, {
    "type" : "WEB",
    "url" : "https://github.com/cjlin1/libsvm/blob/9a3a9708926dec87d382c43b203f2ca19c2d56a0/svm.cpp#L2501"
  }, {
    "type" : "WEB",
    "url" : "https://github.com/pypa/advisory-database/tree/main/vulns/scikit-learn/PYSEC-2020-108.yaml"
  }, {
    "type" : "PACKAGE",
    "url" : "https://github.com/scikit-learn/scikit-learn"
  }, {
    "type" : "WEB",
    "url" : "https://security.gentoo.org/glsa/202301-03"
  }, {
    "type" : "WEB",
    "url" : "http://packetstormsecurity.com/files/160281/SciKit-Learn-0.23.2-Denial-Of-Service.html"
  }, {
    "type" : "WEB",
    "url" : "http://seclists.org/fulldisclosure/2020/Nov/44"
  } ],
  "database_specific" : {
    "cwe_ids" : [ ],
    "severity" : "HIGH",
    "github_reviewed" : true,
    "github_reviewed_at" : "2024-02-01T20:59:15Z",
    "nvd_published_at" : "2020-11-21T21:15:00Z"
  }
}