CVE-2022-36002: TensorFlow Unbatch CHECK Fail Vulnerability

Background

TensorFlow is an open-source machine learning framework developed by Google. It provides a comprehensive ecosystem of tools, libraries, and resources for building and deploying machine learning models. TensorFlow is widely used in various domains, including research, industry, and academia. The vulnerability identified as CVE-2022-36002 affects TensorFlow versions 0 to 2.7.2, 2.8.0 to 2.8.1, and 2.9.0 to 2.9.1, as well as their CPU and GPU variants.

Vulnerability Detail

The vulnerability in TensorFlow lies in the Unbatch function, which fails a CHECK when receiving a nonscalar input id. This flaw can be exploited to trigger a denial of service attack. An attacker can provide a malicious input that causes the CHECK to fail, leading to the termination of the program or service. This vulnerability has a CVSS score of 5.9, indicating a moderate severity level.

pip install --upgrade "tensorflow==2.21.0"

Workaround

As a workaround, users can avoid using the Unbatch function with nonscalar input id until the TensorFlow installation is updated to a patched version. This can be achieved by validating and filtering the input data to ensure it is scalar before passing it to the Unbatch function. Here is an example of how to perform this validation in Python:

import tensorflow as tf

def validate_input_id(id):
    if tf.rank(id) > 0:
        raise ValueError('Input id must be scalar')

# Usage
id = ...  # Input id
validate_input_id(id)
# Call Unbatch function
...  # Unbatch call

Conclusion

The vulnerability in TensorFlow's Unbatch function can lead to a denial of service attack. Users are advised to update their TensorFlow installation to one of the patched versions (2.7.2, 2.8.1, or 2.9.1) to mitigate this vulnerability. Alternatively, users can implement the provided workaround to validate and filter the input data. For further support on vulnerability remediation, please contact DevNack.com.

See every vulnerability recorded for tensorflow →

FAQ