Technical impact
The vulnerability resides in the libiconv CP949 decoder within the GNU C Library (glibc). When an invalid CP949 byte sequence – for example, \xa2\xe8 – is processed while the conversion destination is configured with the //IGNORE option, the decoder mistakenly advances its read pointer twice. If the invalid sequence is at the end of the user‑supplied buffer, the routine reads beyond the buffer limits, causing a fault in the calling application. This is a classic buffer over‑read (CWE‑126) that results in a denial‑of‑service condition rather than arbitrary code execution.
Affected Systems
The vulnerability affects any system running glibc version 2.1 or newer that includes the affected libiconv implementation. Systems that use CP949→UTF‑8 conversion paths—such as internationalized input parsers, file name decoders, or user‑supplied data processing—built against an unpatched glibc release are at risk.
Risk and Exploitability
The CVSS score of 5.9 indicates medium severity. The EPSS score is not available, and the flaw is not catalogued in CISA’s KEV. Based on the description, the likely attack vector involves an attacker supplying crafted CP949 data to an application that performs iconv conversion with the //IGNORE option, which may be possible remotely if the application accepts external CP949 input. Successful exploitation leads only to application crashes, providing a denial‑of‑service that can impact service availability.