CVE-2021-29552 CHECK-failure in `UnsortedSegmentJoin`

Low 2.5 Published 14 May 2021 Updated 3 Aug 2024

An attacker can cause a denial of service in TensorFlow by providing an empty num_segments tensor to UnsortedSegmentJoin, triggering a CHECK failure.

Exploited in the wild: Not known
Login needed: Any signed-in user
Attackable: Local or same network only
Victim action needed: Not needed
Fix: Available

What is CVE-2021-29552?

CVE-2021-29552 is a low-severity denial of service vulnerability in Google Tensorflow, published on 14 May 2021. It can only be exploited from the same machine or local network, with no action from the victim.

Official description

TensorFlow is an end-to-end open source platform for machine learning. An attacker can cause a denial of service by controlling the values of `num_segments` tensor argument for `UnsortedSegmentJoin`. This is because the implementation(https://github.com/tensorflow/tensorflow/blob/a2a607db15c7cd01d754d37e5448d72a13491bdb/tensorflow/core/kernels/unsorted_segment_join_op.cc#L92-L93) assumes that the `num_segments` tensor is a valid scalar. Since the tensor is empty the `CHECK` involved in `.scalar<T>()()` that checks that the number of elements is exactly 1 will be invalidated and this would result in process termination. The fix will be included in TensorFlow 2.5.0. We will also cherrypick this commit on TensorFlow 2.4.2, TensorFlow 2.3.3, TensorFlow 2.2.3 and TensorFlow 2.1.4, as these are also affected and still in supported range.

Weakness: CWE-617

Which versions are affected by CVE-2021-29552?

28 open source packages ship an affected version, including tensorflow, tensorflow-cpu and tensorflow-gpu. It affects 1 product.

Package	Affected	Fixed in
`tensorflow` Bitnami	All versions	`2.1.4, 2.2.3, 2.3.3, 2.4.2`
`tensorflow` PyPI	All versions	`2.2.0rc0, 2.3.0rc0, 2.3.4, 2.4.3`
`tensorflow` PyPI	All versions	`2.1.4`
`tensorflow` PyPI	All versions	`2.1.4`
`tensorflow` PyPI	From `2.2.0`	`2.2.3`
`tensorflow` PyPI	From `2.2.0`	`2.2.3`

22 more packages

`tensorflow` PyPI	From `2.3.0`	`2.3.3`
`tensorflow` PyPI	From `2.3.0`	`2.3.3`
`tensorflow` PyPI	From `2.4.0`	`2.4.2`
`tensorflow` PyPI	From `2.4.0`	`2.4.2`
`tensorflow-cpu` PyPI	All versions	`2.2.0rc0, 2.3.0rc0, 2.3.4, 2.4.3`
`tensorflow-cpu` PyPI	All versions	`2.1.4`
`tensorflow-cpu` PyPI	All versions	`2.1.4`
`tensorflow-cpu` PyPI	From `2.2.0`	`2.2.3`
`tensorflow-cpu` PyPI	From `2.2.0`	`2.2.3`
`tensorflow-cpu` PyPI	From `2.3.0`	`2.3.3`
`tensorflow-cpu` PyPI	From `2.3.0`	`2.3.3`
`tensorflow-cpu` PyPI	From `2.4.0`	`2.4.2`
`tensorflow-cpu` PyPI	From `2.4.0`	`2.4.2`
`tensorflow-gpu` PyPI	All versions	`2.2.0rc0, 2.3.0rc0, 2.3.4, 2.4.3`
`tensorflow-gpu` PyPI	All versions	`2.1.4`
`tensorflow-gpu` PyPI	All versions	`2.1.4`
`tensorflow-gpu` PyPI	From `2.2.0`	`2.2.3`
`tensorflow-gpu` PyPI	From `2.2.0`	`2.2.3`
`tensorflow-gpu` PyPI	From `2.3.0`	`2.3.3`
`tensorflow-gpu` PyPI	From `2.3.0`	`2.3.3`
`tensorflow-gpu` PyPI	From `2.4.0`	`2.4.2`
`tensorflow-gpu` PyPI	From `2.4.0`	`2.4.2`

Products include Google Tensorflow.

How do I fix CVE-2021-29552?

Upgrade tensorflow to a fixed version for your release line: 2.1.4, 2.2.3, 2.3.3 or 2.4.2.

Upgrade, then put limits on request size, rate and nesting depth so one client cannot exhaust the service. More about Denial of service.

Is CVE-2021-29552 being exploited?

There is no known exploitation of CVE-2021-29552 in the wild and no public exploit code.

All Google Tensorflow vulnerabilities

Sources

Also known as BIT-tensorflow-2021-29552, GHSA-jhq9-wm9m-cf89, PYSEC-2021-189, PYSEC-2021-480, PYSEC-2021-678.