PostgreSQL Fixes 52 Security Flaws, Including a 22-Year-Old Bug That Could Let Users Run Code on the Server
PostgreSQL's maintainers have released updates that fix 52 security flaws. The most serious, tracked as CVE-2026-6460, has a CVSS score of 8.8. It lets a database user who can create objects run code on the server, with the rights of the operating system user that runs PostgreSQL. Versions before 18.7, 17.12, 16.16, 15.20 and 14.25 are affected. Full protection also requires rebuilding any C-language extensions installed on the server.
The flaw involves composite types, which are built from named fields. If a field's type is changed while older values of that type are still held elsewhere in the server, the server can read an old value as if it had the new type. The fix adds version checks so the server can detect a value that no longer matches its type. Researcher Mehmet Ince traces the root of the bug to May 2004, when PostgreSQL gained the ability to change a column's type.
The PostgreSQL Global Development Group has not reported any attacks, and CISA's assessment in the CVE record lists exploitation as none. Ince has published a technical writeup demonstrating a working exploit on PostgreSQL 18.6, though it omits the final step that runs a command. He says he could trigger a crash as a lowest-privileged login role. Red Hat says an attacker needs a login with the ability to create or modify objects. None of the sources says whether an account with only default privileges can go beyond the crash to run code.