Live › vulnerability › CVE-2026-103634
Go to dashboardApache DataSketches C++ mishandles Count-Min sketch deserialization: a truncated or crafted sketch can read past the input buffer and, through a 32-bit size overflow, leave an empty table that later updates and queries read and write outside. An attacker who supplies a crafted sketch can crash the process and may be able to corrupt heap memory further.
Summary written from the sources below. Check the originals before acting.
First seen by InfraStack 2026-10-23 21:57 UTC. All times on this page are UTC.
Only applications that deserialize Count-Min sketches from untrusted sources are affected. Services that only load sketches they produced themselves, or that never call Count-Min deserialization, are not exposed.
No product ranges from the sources yet.
4.1.0 before 5.3.0count_min_sketch::deserialize() and confirm whether the input can come from untrusted sources5.3.0 or laterEach row is the first time that source carried this signal. The analyst’s rows are its own findings.
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.
This page as .md · .json · in your agent: get_signal("CVE-2026-103634")
NVD data is provided by NIST and is public domain · EPSS scores by FIRST, used with attribution. Posts quoted belong to their authors.