RUSTSEC-2026-0452 (GHSA-9FM9-C4H4-8G93)
Vulnerability from osv_rustsec – Published: 2026-11-03 12:00 – Updated: 2026-11-06 15:34 – Source websiteInternalKeyRef::from_encoded is a safe constructor whose only length check is debug_assert!(encoded.len() >= 16), which is compiled out in release builds. Its accessors perform unchecked length arithmetic on the slice: user_key() slices &encoded[..len - 16], trailer()/seq_num() pass len - 16 and timestamp() passes len - 8 to read_u64_be, which computes buffer.as_ptr().add(offset) and performs an 8-byte load. For a short slice the subtraction underflows usize, the pointer wraps below the allocation and eight bytes before the buffer are read (undefined behavior; confirmed with AddressSanitizer as a heap-buffer-overflow). In debug builds the same call panics instead.
The same invariant is missing at the on-disk decode boundary: BlockIterator::seek_next_entry rebuilds the current key from block bytes with upper-bound checks only, and BlockIterator::key() wraps it in InternalKeyRef, which surrealdb's kv-surrealkv backend then reads through key_ref.timestamp(). A crafted or corrupted SSTable therefore reaches the same unchecked subtraction without any unsafe at the call site.
1.2.0 rejects internal keys shorter than their 8-byte trailer before any accessor computes an offset and makes the accessors bounds-checked (fix in commit ac642f9, surrealdb 4.0.0 ships surrealkv 1.2.0).
{
"affected": [
{
"database_specific": {
"categories": [
"memory-exposure"
],
"cvss": "CVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:C/C:H/I:N/A:L",
"informational": null
},
"ecosystem_specific": {
"affected_functions": null,
"affects": {
"arch": [],
"functions": [],
"os": []
}
},
"package": {
"ecosystem": "crates.io",
"name": "surrealkv",
"purl": "pkg:cargo/surrealkv"
},
"ranges": [
{
"events": [
{
"introduced": "0.0.0-0"
},
{
"fixed": "1.2.0"
}
],
"type": "SEMVER"
}
],
"versions": []
}
],
"aliases": [
"GHSA-9fm9-c4h4-8g93"
],
"database_specific": {
"license": "CC0-1.0"
},
"details": "`InternalKeyRef::from_encoded` is a safe constructor whose only length check is `debug_assert!(encoded.len() \u003e= 16)`, which is compiled out in release builds. Its accessors perform unchecked length arithmetic on the slice: `user_key()` slices `\u0026encoded[..len - 16]`, `trailer()`/`seq_num()` pass `len - 16` and `timestamp()` passes `len - 8` to `read_u64_be`, which computes `buffer.as_ptr().add(offset)` and performs an 8-byte load. For a short slice the subtraction underflows `usize`, the pointer wraps below the allocation and eight bytes before the buffer are read (undefined behavior; confirmed with AddressSanitizer as a `heap-buffer-overflow`). In debug builds the same call panics instead.\n\nThe same invariant is missing at the on-disk decode boundary: `BlockIterator::seek_next_entry` rebuilds the current key from block bytes with upper-bound checks only, and `BlockIterator::key()` wraps it in `InternalKeyRef`, which `surrealdb`\u0027s `kv-surrealkv` backend then reads through `key_ref.timestamp()`. A crafted or corrupted SSTable therefore reaches the same unchecked subtraction without any `unsafe` at the call site.\n\n`1.2.0` rejects internal keys shorter than their 8-byte trailer before any accessor computes an offset and makes the accessors bounds-checked (fix in commit `ac642f9`, `surrealdb` 4.0.0 ships `surrealkv` 1.2.0).",
"id": "RUSTSEC-2026-0452",
"modified": "2026-11-06T15:34:32Z",
"published": "2026-11-03T12:00:00Z",
"references": [
{
"type": "PACKAGE",
"url": "https://crates.io/crates/surrealkv"
},
{
"type": "ADVISORY",
"url": "https://rustsec.org/advisories/RUSTSEC-2026-0452.html"
},
{
"type": "WEB",
"url": "https://github.com/surrealdb/surrealkv/commit/ac642f945d63c9c9f7ad35f480995fd516f83fb6"
},
{
"type": "WEB",
"url": "https://github.com/surrealdb/surrealkv/pull/497"
}
],
"related": [],
"severity": [
{
"score": "CVSS:3.1/AV:L/AC:L/PR:N/UI:N/S:C/C:H/I:N/A:L",
"type": "CVSS_V3"
}
],
"summary": "Out-of-bounds read when decoding internal keys shorter than 16 bytes"
}
Drag nodes to arrange them. Double-click a related vulnerability to open it, or a source to visit its link. The Physics button on the left tunes the layout and the simulation; the legend at the bottom left hides or shows nodes by kind and by sighting type.
Sightings
| Author | Source | Type | Date | Other |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or observed by the user.
- Confirmed: The vulnerability has been validated from an analyst's perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
- Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
- Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
- Not confirmed: The user expressed doubt about the validity of the vulnerability.
- Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.
Vulnerability timeline
Sightings and recorded dates, grouped by day · UTC
Loading timeline…
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.
Related by attack behaviour
Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.