CVE-2021-29533: Denial of Service in TensorFlow

Background

TensorFlow is an end-to-end open source platform for machine learning. It provides a comprehensive ecosystem of tools, libraries, and resources for developing and deploying machine learning models. TensorFlow is widely used in various domains, including research, industry, and academia. It supports a wide range of platforms and devices, making it accessible and versatile for different applications. However, a vulnerability has been discovered in TensorFlow that can lead to a denial of service attack.

Vulnerability Detail

The vulnerability in TensorFlow (CVE-2021-29533) allows an attacker to trigger a denial of service by exploiting a CHECK failure in the tf.raw_ops.DrawBoundingBoxes function. The implementation of this function uses CHECK_* assertions instead of OP_REQUIRES to validate user-controlled inputs. While OP_REQUIRES allows returning an error condition to the user, the CHECK_* macros result in a program crash if the condition is false. In this case, passing an empty image to tf.raw_ops.DrawBoundingBoxes results in a CHECK failure due to a false assertion. The height of the image is expected to be non-zero, but when it is zero, the assertion fails and the program execution is aborted. This vulnerability can be exploited by an attacker to disrupt the availability of TensorFlow applications.

pip install --upgrade "tensorflow-cpu==2.21.0"

Workaround

As a temporary workaround, users can validate the input images before passing them to the tf.raw_ops.DrawBoundingBoxes function. By checking the height of the image and ensuring it is non-zero, the vulnerability can be mitigated. Here is an example of how to perform this validation in Python:

image_height = get_image_height(image)
if image_height > 0:
    tf.raw_ops.DrawBoundingBoxes(image, ...)
else:
    # Handle the case of an empty image
    ...

Conclusion

In conclusion, the vulnerability (CVE-2021-29533) in TensorFlow can be exploited to trigger a denial of service attack. The fix for this vulnerability will be included in TensorFlow 2.5.0 and backported to earlier versions. Users are advised to update their installations to the latest patched version as soon as it becomes available. In the meantime, a temporary workaround is available by validating the input images before passing them to the vulnerable function. For further support on vulnerability remediation, please contact DevNack.com.

See every vulnerability recorded for tensorflow-cpu →

FAQ