CVE Catalog

CVE-2026-103634

MediumCVSS 5.3
Published: Updated: Translated: NVD NIST

Exploitation Probability (EPSS)

Low risk
0.24%

15th percentile - higher than 15% of all known CVEs

Summary

An out-of-bounds read and write in the Count-Min sketch deserialization of Apache DataSketches C++ (versions 4.1.0 to 5.3.0) occurs because the input size check did not include the preamble, allowing a truncated sketch to cause a read past the end. Additionally, the table size is computed in 32-bit arithmetic, which can wrap to zero, leading to an empty table and subsequent out-of-bounds reads and writes, corrupting heap memory.

Risk Assessment

Applications that deserialize Count-Min sketches from untrusted sources may experience memory corruption, crashes, and potentially enable further exploitation. This could lead to denial of service or code execution.

Recommendation

Upgrade Apache DataSketches C++ to version 5.3.0 or later. Avoid deserializing sketches from untrusted sources.

Other vulnerabilities in Apache DataSketches C++

See all
Original NVD description (English source)

Out-of-bounds read and write in the Count-Min sketch deserialization of Apache DataSketches C++ (repo: datasketches-cpp). count_min_sketch::deserialize() did not include the preamble in its input size check, so a truncated sketch could cause a read of up to 16 bytes past the end of the input. In addition, the table size was computed from the serialized number of buckets and number of hash functions in 32-bit arithmetic. A crafted sketch could make it wrap to zero, so that the sketch deserialized with an empty table, and later updates and estimate queries read and wrote outside the heap allocation. This can corrupt heap memory, causing a crash and potentially enabling further exploitation. This issue affects Apache DataSketches C++: from 4.1.0 before 5.3.0. Only applications that deserialize Count-Min sketches from untrusted sources are affected. Users are recommended to upgrade to version 5.3.0, which fixes this issue.

Vulnerability data from NVD (NIST) · CISA KEV · EPSS