vllm
Affected
- introduced
0, fixed0.28.0(ecosystem)
Fixed versions
0.28.0
Fix
- Fixed version
0.28.0pip install --upgrade vllm==0.28.0
Advisory
Also known as: CVE-2026-105753, PYSEC-2026-4197
vLLM: Mirrored multimodal IPC caches desync after a rejected request — a later request reusing the same media hash trips a receiver assertion in the engine core
Affected
vllmSummary
vLLM's default multimodal cache (mm_processor_cache_type="lru") mirrors state across two processes: the frontend (P0) holds only metadata (MultiModalProcessorSenderCache) while the engine core (P1) holds the real payload (MultiModalReceiverCache). The design invariant is that get_and_update() runs on P0 and P1 in lockstep for every request, so eviction order stays mirrored and P0 can answer "is this cached in P1?" without talking to P1.
That invariant breaks when a request is rejected after P0 has rendered and hashed the multimodal input (populating the P0 cache) but before P1 receives the item — for example, an oversized chat prompt rejected on max_model_len *after* rendering. P0 now believes the media is cached while P1 never got it. A later request reusing the same media hash gets a P0 hit, so P0 sends None instead of the payload, and P1 — which has nothing cached — trips assert mm_item is not None, f"Expected a cached item for {mm_hash=}".
This is a remotely reachable, request-controlled cache-mirroring desync on the standard multimodal inference path. It requires only the default cache configuration.
Affected code
Links pinned to the confirmed commit `752a3a504485` (v0.25.1):
MultiModalProcessorSenderCache at `vllm/multimodal/cache.py#L379`; get_and_update_item at `#L410-L421`, commit assertion at `#L418`.MultiModalReceiverCache at `vllm/multimodal/cache.py#L630`; get_and_update_item at `#L652-L663`, with the failing assert mm_item is not None, f"Expected a cached item for {mm_hash=}" at `#L660`.processor_only, disabled-caching, and shm paths are not affected.render_chat_request), and the length check raises after the render in `vllm/entrypoints/serve/utils/api_utils.py#L171-L189`._merge_mm_kwargs) commits the P0 sender entry during render.LRUCache.touch() guarding if key in self:) — a different mechanism that does not touch the sender/receiver commit-ordering protocol and does not remediate this assertion.The P1 receiver sink — the assertion that fires when P1 receives None for a hash it never cached:
# vllm/multimodal/cache.py Lines 651-663
@override
def get_and_update_item(
self,
mm_item: MultiModalKwargsItem | None,
mm_hash: str,
) -> MultiModalKwargsItem:
if (cached_item := self._cache.get(mm_hash)) is not None:
return cached_item
assert mm_item is not None, f"Expected a cached item for {mm_hash=}"
self._cache[mm_hash] = mm_item
return mm_item
The P0 sender — on a hit it drops the payload (returns None) and, on a miss during render, unconditionally commits the metadata entry with no rollback tied to admission:
# vllm/multimodal/cache.py Lines 409-422
@override
def get_and_update_item(
self,
mm_item: MultiModalProcessorCacheInItem,
mm_hash: str,
) -> MultiModalProcessorCacheOutItem:
if (cached_item := self._cache.get(mm_hash)) is not None:
return None, cached_item.prompt_updates
assert mm_item is not None, f"Expected a cached item for {mm_hash=}"
self._cache[mm_hash] = MultiModalProcessorCacheItemMetadata(*mm_item)
return mm_item
Impact
A remote client submitting multimodal requests can poison a cache identity — render a media item successfully, then have that request rejected — so a later request reusing the same media hash fails on the P1 receiver assertion. This is an availability failure against a shared serving instance. No code execution, memory corruption, or data disclosure is claimed.
On this revision the failure is scoped as a request-level preprocessing error (the engine core catches around preprocess_add_request); public reports show the same assertion cascading into further engine-loop assertions on other revisions. It applies to multimodal models running the default mirrored lru cache.
Suggested Fix
Two complementary changes:
assert mm_item is not None into a checked, request-scoped error (fetch-on-miss from P0) so a desync degrades a single request rather than asserting in the engine loop.Medium Basis: The source's own band: MODERATE
0, fixed 0.28.0 (ecosystem)0.28.0
0.28.0pip install --upgrade vllm==0.28.0Fix commands are the export's own generic commands. Ojuju never runs them. Check them against your setup first.
Check your own computer: guard packages check
Source: GitHub Advisory Database (reviewed advisories) · CC BY 4.0