Security Vulnerabilities
Memory Corruption when processing concurrent DMA buffer allocation and deallocation commands without proper synchronization.
CVSS Score
6.7
EPSS Score
0.001
Published
2026-10-06
Memory corruption when performing concurrent operations on shared memory page lists due to lack of proper synchronization mechanisms.
CVSS Score
7.8
EPSS Score
0.001
Published
2026-10-06
Memory corruption while processing IOCTL command called from user space to the kernel with invalid parameters.
CVSS Score
6.6
EPSS Score
0.001
Published
2026-10-06
Memory corruption when non-secure loader rewrites page tables before secure memory initialization.
CVSS Score
7.8
EPSS Score
0.001
Published
2026-10-06
A flaw was found in SSSD. When configured with the Entra ID identity provider, input lookup names containing single quotes are not properly escaped before being included in Microsoft Graph Open Data Protocol (OData) queries. A low-privileged local user can exploit this flaw by submitting a crafted search request, altering query filters to broaden user or group searches. This can lead to information disclosure by retrieving unintended directory objects, as well as a Denial of Service (DoS) through excessive processing and cache population.
CVSS Score
4.4
EPSS Score
0.001
Published
2026-10-06
vLLM is an inference and serving engine for large language models. Prior to 0.30.0, the /inference/v1/generate endpoint in the disaggregated scale-out path accepts caller-supplied tensors in the features.kwargs_data field, cache identifiers in the features.mm_hashes field, ranges in the features.mm_placeholders field, and wire-selected multimodal field processors without rebinding them to the active model renderer contract. Forged grid geometry, field types, or non-positive placeholder lengths can terminate the shared EngineCore; when an attacker knows or can induce a victim's content hash, forged cache hashes can poison or retrieve cross-request encoder-cache state; and dropped sparse placeholder masks can alter replayed transport semantics. This issue is fixed in version 0.30.0.
CVSS Score
6.5
EPSS Score
0.003
Published
2026-10-05
vLLM is an inference and serving engine for large language models. Prior to 0.30.0, flash late-interaction scoring at the /score and /rerank endpoints derives each worker's query_key value from the caller-controlled X-Request-Id header. A concurrent request that reuses a victim's identifier can overwrite the cached query embedding so the victim's documents are scored against the attacker's query, and shared use counters can also cause a late-interaction cache-miss error. This issue is fixed in version 0.30.0.
CVSS Score
4.2
EPSS Score
0.002
Published
2026-10-05
vLLM is an inference and serving engine for large language models. Prior to 0.30.0, OpenAI-compatible request models accept a non-empty cache_salt value without enforcing the character and length restrictions required by the IPCCacheServerKey consumer in LMCache-MP. On deployments using the LMCache-MP connector, a salt that contains a forbidden character or exceeds the permitted length can raise an uncaught ValueError during scheduler cache lookup, causing EngineCore to terminate and denying service to all concurrent users. This issue is fixed in version 0.30.0.
CVSS Score
6.5
EPSS Score
0.003
Published
2026-10-05
vLLM is an inference and serving engine for large language models. Prior to 0.30.0, structured-output request failures can escape request-scoped validation and reach the EngineCore fatal-error path. A per-request backend mismatch can re-raise a grammar compilation exception, padding produced by the ngram_gpu speculative-decoding mode can pass a negative token to guidance validation, and the Rust frontend can admit empty structured-output values that the Python frontend rejects, allowing ordinary constrained-generation requests to terminate the shared engine. This issue is fixed in version 0.30.0.
CVSS Score
6.5
EPSS Score
0.003
Published
2026-10-05
vLLM is an inference and serving engine for large language models. From 0.24.0 until 0.30.0, the Qwen2VLVideoBackend and Qwen3VLVideoBackend classes accept request-level values for the media_io_kwargs.video.max_frames and media_io_kwargs.video.fps fields without enforcing server-side ceilings. An unauthenticated caller can submit these values to the /tokenize endpoint, causing the sampler to decode every frame selected from attacker-controlled video input, consume disproportionate frontend memory, and potentially terminate the API process before scheduling or admission control. The Rust frontend is not affected because it rejects the media_io_kwargs field. This issue is fixed in version 0.30.0.
CVSS Score
5.3
EPSS Score
0.003
Published
2026-10-05