CVE-2022-23590
Crash due to erroneous `StatusOr` in Tensorflow
In short
A maliciously modified TensorFlow SavedModel can crash a TensorFlow process by forcing extraction of an error value from a StatusOr object. This affects applications using TensorFlow 2.7 and earlier versions.
Technical detail
An attacker can craft a malicious GraphDef within a SavedModel that triggers improper handling of StatusOr error states, causing denial of service through process termination. The vulnerability exists in the StatusOr value extraction logic where error conditions are not properly validated before forced extraction, affecting TensorFlow versions prior to 2.7.1 and 2.8.0.
Summary generated and translated by AI from the official description.
Tensorflow is an Open Source Machine Learning Framework. A `GraphDef` from a TensorFlow `SavedModel` can be maliciously altered to cause a TensorFlow process to crash due to encountering a `StatusOr` value that is an error and forcibly extracting the value from it. We have patched the issue in multiple GitHub commits and these will be included in TensorFlow 2.8.0 and TensorFlow 2.7.1, as both are affected.
CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:N/I:N/A:H
Affected products
tensorflow · tensorflowWant to know if your infrastructure is exposed to this?
Talk to TrueHacking →References
https://github.com/tensorflow/tensorflow/blob/274df9b02330b790aa8de1cee164b70f72b9b244/tensorflow/core/graph/graph.cc#L560-L567https://github.com/tensorflow/tensorflow/commit/955059813cc325dc1db5e2daa6221271406d4439https://github.com/tensorflow/tensorflow/security/advisories/GHSA-pqrv-8r2f-7278