CVE-2023-37941

Apache Superset metadata db write access leads to remote code execution

• CVSS 6.6
• CWE-502 Deserialization of Untrusted Data
• Input Validation and Sanitization
• Remote

If an attacker gains write access to the Apache Superset metadata database, they could persist a specifically crafted Python object that may lead to remote code execution on Superset's web backend. The Superset metadata db is an 'internal' component that is typically only accessible directly by the system administrator and the superset process itself. Gaining access to that database should be difficult and require significant privileges. This vulnerability impacts Apache Superset versions 1.5.0 up to and including 2.1.0. Users are recommended to upgrade to version 2.1.1 or later.

CVE

CVE-2023-37941

CWE

CWE-502 Deserialization of Untrusted Data

CVSS base score

6.6

Published

2023-09-06

OWASP

A08 Software and Data Integrity Failures

Orthogonal defect classification

Function

Code defect classification

Incorrect Functionality

Category

Input Validation and Sanitization

Subcategory

Insecure Parsing or Deserialization

Accessibility scope

Remote

Impact

Arbitrary Code Execution

Fixed by upgrading

Yes

Solution

Update Apache Superset to version 2.1.1 or later.

import pickle
import sqlalchemy

# Assuming `session` is your SQLAlchemy session connected to the Superset metadata database

def persist_vulnerable_object(obj):
    # Serialize the object using pickle (vulnerable to arbitrary code execution)
    data = pickle.dumps(obj)
    
    # Store the serialized data in the database
    new_entry = YourModel(data=data)  # Replace YourModel with your actual model
    session.add(new_entry)
    session.commit()

# Example usage
user_input = {'key': 'value', '__reduce__': (os.system, ('echo Vulnerable!',))}  # Malicious input
persist_vulnerable_object(user_input)

import pickle
import sqlalchemy

# Assuming `session` is your SQLAlchemy session connected to the Superset metadata database

def safe_load(data):
    # Only allow loading of safe types
    allowed_classes = (dict, list, str, int, float, bool)
    obj = pickle.loads(data)
    
    if not isinstance(obj, allowed_classes):
        raise ValueError("Unsafe object type detected!")
    
    return obj

def persist_safe_object(obj):
    # Serialize the object safely
    safe_data = pickle.dumps(obj)
    
    # Store the safe_data in the database
    new_entry = YourModel(data=safe_data)  # Replace YourModel with your actual model
    session.add(new_entry)
    session.commit()

# Example usage
try:
    user_input = {'key': 'value'}  # Replace with actual user input
    persist_safe_object(user_input)
except Exception as e:
    print(f"Error: {e}")

@misc{vaitp:cve202337941,
  title        = {{Apache Superset metadata db write access leads to remote code execution}},
  author       = {Bogaerts, Fr\'ed\'eric and Ivaki, Naghmeh and Fonseca, Jos\'e},
  year         = {2023},
  note         = {VAITP Python Vulnerability Dataset, entry CVE-2023-37941},
  howpublished = {\url{https://netpack.pt/vaitp/vulnerability/CVE-2023-37941/}}
}