Vulnerabilities · 2021 · Google
CVE-2021-37650
Google Tensorflow vulnerability
TensorFlow is an end-to-end open source platform for machine learning. In affected versions the implementation for `tf.raw_ops.ExperimentalDatasetToTFRecord` and `tf.raw_ops.DatasetToTFRecord` can trigger heap buffer overflow and segmentation fault. The [implementation](https://github.com/tensorflow/tensorflow/blob/f24faa153ad31a4b51578f8181d3aaab77a1ddeb/tensorflow/core/kernels/data/experimental/to_tf_record_op.cc#L93-L102) assumes that all records in the dataset are of string type. However, there is no check for that, and the example given above uses numeric types. We have patched the issue in GitHub commit e0b6e58c328059829c3eb968136f17aa72b6c876. The fix will be included in TensorFlow 2.6.0. We will also cherrypick this commit on TensorFlow 2.5.1, TensorFlow 2.4.3, and TensorFlow 2.3.4, as these are also affected and still in supported range.
Affected products
Tensorflow. The exact versions are in the references.
What to do now
- Check your exposureFind out whether Google Tensorflow runs anywhere in your environment, including systems of suppliers.
- UpdateInstall the version in which the vendor fixed the vulnerability. The references lead to the advisory.
- Set the orderSystems that can be reached from the internet come first, then the rest.
References
- github.com/tensorflow/tensorflow/commit/e0b6e58c328059829c3eb968136f17aa72b6c876 · Patch
- github.com/tensorflow/tensorflow/security/advisories/GHSA-f8h4-7rgh-q2gm · Third Party Advisory
- nvd.nist.gov/vuln/detail/CVE-2021-37650
- github.com/pypa/advisory-database/tree/main/vulns/tensorflow-cpu/PYSEC-2021-563.yaml
- github.com/pypa/advisory-database/tree/main/vulns/tensorflow-gpu/PYSEC-2021-761.yaml
- github.com/pypa/advisory-database/tree/main/vulns/tensorflow/PYSEC-2021-272.yaml
- github.com/advisories/GHSA-f8h4-7rgh-q2gm
- nvd.nist.gov/vuln/detail/CVE-2021-37650
- cve.org/CVERecord?id=CVE-2021-37650
More vulnerabilities in Google products
Data: CVE list and the NVD. This product uses data from the NVD API but is not endorsed or certified by the NVD.