Cloudanix VulnDB

CVE-2015-5607

High · 8.8

CVSS

8.8

EPSS

1.2%

Published

Sep 20, 2017

CISA due date

—

Cloudanix analysis

Schedule a fix SSVC: Track

A flaw in older IPython software could let attackers trick logged-in users into performing actions they did not intend, such as running code or changing files, simply by visiting a malicious website.

Exploit Available
Exploitation
Requires User Interaction
Attacker needs
User Interaction
Complexity
Low
Impact
Data Tampering
Component
Application
Class
CSRF
Remediation
Upgrade
Effort
Medium
Patch risk
Low

What it actually does

The vulnerability permits an attacker to submit unauthorized HTTP requests to the IPython REST API by leveraging an authenticated user's browser session without their knowledge or consent.

Why it matters to the business

Successful exploitation can allow remote attackers to execute arbitrary code, modify notebooks, or exfiltrate data within shared scientific computing environments, potentially disrupting research workflows and exposing sensitive intellectual property across organizations using affected IPython deployments.

How an attack unfolds

An attacker hosts a malicious web page containing crafted API calls; when a victim who is authenticated to an IPython server visits the page, their browser automatically submits the forged requests, bypassing authentication and enabling actions such as code execution or configuration changes.

How to fix it

Upgrade IPython to a version beyond 3.x that addresses the REST API CSRF issue; alternatively, place the service behind a reverse proxy that enforces CSRF tokens, disable the REST API if unused, or restrict access via network controls until patches are applied.

How to confirm the fix worked

Review release notes and changelogs for IPython 4+ or equivalent Jupyter components to confirm CSRF protections were added; test by attempting cross-origin requests to the /api endpoint from an external origin while authenticated and verify that requests are rejected.

Typical attack path

  1. Identify an IPython 2 or 3 instance exposing its REST API to authenticated users
  2. Craft a malicious HTML page or email with embedded JavaScript that triggers REST endpoints
  3. Lure an authenticated user to visit the attacker-controlled page
  4. Browser submits forged requests using the user's session cookies or tokens
  5. Attacker achieves unauthorized actions such as running code or altering notebooks

How to detect it

Examine web server or application logs for REST API requests originating from unexpected referrers or lacking proper headers; look for patterns of API calls that coincide with users visiting untrusted external domains shortly before suspicious notebook or process activity.

What your security team should do

  • Immediately inventory all IPython 2/3 deployments and prioritize upgrade or decommissioning
  • Apply network segmentation and authentication requirements for any remaining instances
  • Monitor for unexpected API activity from internal users accessing external sites
  • Consider migrating to maintained Jupyter environments with modern security controls

The record

Cross-site request forgery in the REST API in IPython 2 and 3.

Affected packages

Ecosystem Package Introduced Fixed
PyPI ipython 0 3.2.1
PyPI ipython 0 a05fe052a18810e92d9be8c1185952c13fe4e5b0
PyPI ipython 0.12 2.4.1
PyPI ipython 3.0.0 3.2.3
Debian:11 ipython 0 2.4.1-1
Debian:12 ipython 0 2.4.1-1
Debian:13 ipython 0 2.4.1-1
Mageia:4 ipython 0 2.3.0-1.mga4
Mageia:5 ipython 0 2.3.0-2.2.mga5
Ubuntu:16.04:LTS ipython 0 2.4.1-1
Ubuntu:18.04:LTS ipython 0 5.5.0-1
Ubuntu:Pro:14.04:LTS ipython 0 1.2.1-2ubuntu0.1~esm1
Ubuntu:Pro:18.04:LTS ipython 0 5.5.0-1ubuntu0.1~esm1
Ubuntu:Pro:20.04:LTS ipython 0 7.13.0-1ubuntu0.1~esm1
pypi ipython 0.12 No fix
pypi ipython 3.0.0 No fix

Related vulnerabilities