{"items":[{"id":"NVI-00010610","kind":"vulnerability","title":"vLLM: Forged multimodal UUID + P0/P1 processor-cache eviction drift: unauthenticated remote crash of the whole vLLM engine","revision":3,"unknowns":["Customer inventory, deployment, reachability and exploitation have not been inspected.","NVD lastModified does not establish freshness of every associated CPE record."],"lifecycle":"active","identifiers":["GHSA-p92p-rxj5-7p2x","CVE-2026-114050"],"plain_summary":"GitHub lists this reviewed advisory for vllm. Published version ranges and source prerequisites are shown below; this is not an inspection of a visitor’s application.","review_status":"source_listed","control_topics":[],"public_actions":[{"summary":"Identify the exact component and deployed version, examine the linked publisher’s conditions, and retain evidence of any mitigation.","source_ids":["sr-03b4ccd9231d468ba8c0841adb9f4976"],"action_type":"investigate","limitations":"Confirm current vendor support, compatibility and deployment evidence. This report has not tested a fix in your system.","current_supported_target":null}],"schema_version":"1.1.0","source_metrics":[{"kind":"CVSS","as_of":"2026-10-23T20:43:45.000Z","value":"6.5 medium","meaning":"GitHub source severity; exploitability and local consequence require separate evidence.","source_id":"sr-6a8f230bd09637a8ce7902d114d620f5"},{"kind":"CVSS","as_of":"2026-10-22T15:44:59.783Z","value":"7.5 HIGH (3.1)","meaning":"disclosure@vulncheck.com severity; no combined Null risk score.","source_id":"sr-3c09058eac283a13790cc44ab55fa36c"},{"kind":"CVSS","as_of":"2026-10-22T15:44:59.783Z","value":"7.1 HIGH (4.0)","meaning":"disclosure@vulncheck.com severity; no combined Null risk score.","source_id":"sr-3c09058eac283a13790cc44ab55fa36c"},{"kind":"EPSS","as_of":"2026-11-04T12:03:12.000Z","value":"0.207%","meaning":"FIRST EPSS model v2026.06.15; exploitation probability over 30 days, percentile 0.10924. Missing scores are unknown.","source_id":"sr-03b4ccd9231d468ba8c0841adb9f4976"}],"why_it_matters":"Published affected context helps an owner decide what evidence to gather. This report has not examined your application.","related_sources":[{"id":"sr-6a8f230bd09637a8ce7902d114d620f5","url":"https://github.com/advisories/GHSA-p92p-rxj5-7p2x","basis":"vendor_advisory","notes":"GitHub Advisory Database; CC-BY-4.0. Each advisory links its original attribution. Null normalizes fields.","title":"GHSA-p92p-rxj5-7p2x","publisher":"GitHub Advisory Database","observed_at":"2026-10-31T22:07:08.657Z"},{"id":"sr-3c09058eac283a13790cc44ab55fa36c","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-114050","basis":"official_catalog","notes":"This product uses data from the NVD API but is not endorsed or certified by the NVD.","title":"CVE-2026-114050","publisher":"NIST National Vulnerability Database","observed_at":"2026-11-03T09:58:52.895Z"},{"id":"sr-03b4ccd9231d468ba8c0841adb9f4976","url":"https://www.first.org/epss/data.html","basis":"official_catalog","notes":"FIRST EPSS; probability of exploitation in the next 30 days, not local compromise. Dataset date/model retained; FIRST services terms apply.","title":"CVE-2026-114050","publisher":"FIRST EPSS","observed_at":"2026-11-04T17:58:50.312Z"}],"affected_context":[{"certainty":"source_listed","source_ids":["sr-6a8f230bd09637a8ce7902d114d620f5"],"technology":"vllm","known_exclusions":[],"conditions_summary":"## Summary\n\nMultimodal content parts (`image_url`, `video_url`, `input_audio`, ...) accept a user-supplied `uuid` field that vLLM uses **verbatim as the item's cache identity** (`mm_hash`). The two frontend/engine processor caches (P0 \"sender\" cache in the API process, P1 \"receiver\" cache in the EngineCore process) are independent LRU caches whose evictions are **not synchronized**, and the P1 hit path performs **zero consistency validation** — it silently substitutes the cached item's tensor data for whatever the current request actually contains.\n\nA single unauthenticated client can exploit this to kill the entire engine, deterministically, with default server flags:\n\n1. **Seed**: send a small image with `uuid=\"U\"`. Both caches now map `U` to an item with **N_A** visual tokens (e.g. 4).\n2. **Drift**: send ~70 requests, each carrying a *unique* uuid, a max-size image, and an over-long text prompt. Each request is rejected with 400 — but **after** its image has been inserted into the P0 cache (insertion happens during chat rendering, before prompt-length validation), so these requests fill P0 **without ever reaching the engine/P1**. `U` is evicted from P0 only.\n3. **Kill**: send a ","public_affected_ranges":["< 0.31.0"]}],"null_published_at":"2026-11-06T07:24:11.245Z","source_updated_at":"2026-10-23T20:43:45.000Z","source_disclosed_at":"2026-10-23T20:43:44.000Z","is_synthetic_fixture":false,"null_last_reviewed_at":null,"public_change_summary":"Dated EPSS estimate updated; this is not a new disclosure or proof of exploitation.","null_first_observed_at":"2026-10-31T22:07:08.657Z"},{"id":"NVI-00010610","kind":"vulnerability","title":"vLLM: Forged multimodal UUID + P0/P1 processor-cache eviction drift: unauthenticated remote crash of the whole vLLM engine","revision":2,"unknowns":["Customer inventory, deployment, reachability and exploitation have not been inspected.","NVD lastModified does not establish freshness of every associated CPE record."],"lifecycle":"active","identifiers":["GHSA-p92p-rxj5-7p2x","CVE-2026-114050"],"plain_summary":"GitHub lists this reviewed advisory for vllm. Published version ranges and source prerequisites are shown below; this is not an inspection of a visitor’s application.","review_status":"source_listed","control_topics":[],"public_actions":[{"summary":"Identify the exact component and deployed version, examine the linked publisher’s conditions, and retain evidence of any mitigation.","source_ids":["sr-3c09058eac283a13790cc44ab55fa36c"],"action_type":"investigate","limitations":"Confirm current vendor support, compatibility and deployment evidence. This report has not tested a fix in your system.","current_supported_target":null}],"schema_version":"1.1.0","source_metrics":[{"kind":"CVSS","as_of":"2026-10-23T20:43:45.000Z","value":"6.5 medium","meaning":"GitHub source severity; exploitability and local consequence require separate evidence.","source_id":"sr-6a8f230bd09637a8ce7902d114d620f5"},{"kind":"CVSS","as_of":"2026-10-22T15:44:59.783Z","value":"7.5 HIGH (3.1)","meaning":"disclosure@vulncheck.com severity; no combined Null risk score.","source_id":"sr-3c09058eac283a13790cc44ab55fa36c"},{"kind":"CVSS","as_of":"2026-10-22T15:44:59.783Z","value":"7.1 HIGH (4.0)","meaning":"disclosure@vulncheck.com severity; no combined Null risk score.","source_id":"sr-3c09058eac283a13790cc44ab55fa36c"}],"why_it_matters":"Published affected context helps an owner decide what evidence to gather. This report has not examined your application.","related_sources":[{"id":"sr-6a8f230bd09637a8ce7902d114d620f5","url":"https://github.com/advisories/GHSA-p92p-rxj5-7p2x","basis":"vendor_advisory","notes":"GitHub Advisory Database; CC-BY-4.0. Each advisory links its original attribution. Null normalizes fields.","title":"GHSA-p92p-rxj5-7p2x","publisher":"GitHub Advisory Database","observed_at":"2026-10-31T22:07:08.657Z"},{"id":"sr-3c09058eac283a13790cc44ab55fa36c","url":"https://nvd.nist.gov/vuln/detail/CVE-2026-114050","basis":"official_catalog","notes":"This product uses data from the NVD API but is not endorsed or certified by the NVD.","title":"CVE-2026-114050","publisher":"NIST National Vulnerability Database","observed_at":"2026-11-03T09:58:52.895Z"}],"affected_context":[{"certainty":"source_listed","source_ids":["sr-6a8f230bd09637a8ce7902d114d620f5"],"technology":"vllm","known_exclusions":[],"conditions_summary":"## Summary\n\nMultimodal content parts (`image_url`, `video_url`, `input_audio`, ...) accept a user-supplied `uuid` field that vLLM uses **verbatim as the item's cache identity** (`mm_hash`). The two frontend/engine processor caches (P0 \"sender\" cache in the API process, P1 \"receiver\" cache in the EngineCore process) are independent LRU caches whose evictions are **not synchronized**, and the P1 hit path performs **zero consistency validation** — it silently substitutes the cached item's tensor data for whatever the current request actually contains.\n\nA single unauthenticated client can exploit this to kill the entire engine, deterministically, with default server flags:\n\n1. **Seed**: send a small image with `uuid=\"U\"`. Both caches now map `U` to an item with **N_A** visual tokens (e.g. 4).\n2. **Drift**: send ~70 requests, each carrying a *unique* uuid, a max-size image, and an over-long text prompt. Each request is rejected with 400 — but **after** its image has been inserted into the P0 cache (insertion happens during chat rendering, before prompt-length validation), so these requests fill P0 **without ever reaching the engine/P1**. `U` is evicted from P0 only.\n3. **Kill**: send a ","public_affected_ranges":["< 0.31.0"]}],"null_published_at":"2026-11-04T10:04:35.935Z","source_updated_at":"2026-10-23T20:43:45.000Z","source_disclosed_at":"2026-10-23T20:43:44.000Z","is_synthetic_fixture":false,"null_last_reviewed_at":null,"public_change_summary":"Source facts changed; compare the preserved prior revision.","null_first_observed_at":"2026-10-31T22:07:08.657Z"},{"id":"NVI-00010610","kind":"vulnerability","title":"vLLM: Forged multimodal UUID + P0/P1 processor-cache eviction drift: unauthenticated remote crash of the whole vLLM engine","revision":1,"unknowns":["Customer inventory, deployment, reachability and exploitation have not been inspected."],"lifecycle":"active","identifiers":["GHSA-p92p-rxj5-7p2x","CVE-2026-114050"],"plain_summary":"GitHub lists this reviewed advisory for vllm. Published version ranges and source prerequisites are shown below; this is not an inspection of a visitor’s application.","review_status":"source_listed","control_topics":[],"public_actions":[{"summary":"Identify the exact component and deployed version, examine the linked publisher’s conditions, and retain evidence of any mitigation.","source_ids":["sr-6a8f230bd09637a8ce7902d114d620f5"],"action_type":"investigate","limitations":"Confirm current vendor support, compatibility and deployment evidence. This report has not tested a fix in your system.","current_supported_target":null}],"schema_version":"1.1.0","source_metrics":[{"kind":"CVSS","as_of":"2026-10-23T20:43:45.000Z","value":"6.5 medium","meaning":"GitHub source severity; exploitability and local consequence require separate evidence.","source_id":"sr-6a8f230bd09637a8ce7902d114d620f5"}],"why_it_matters":"Published affected context helps an owner decide what evidence to gather. This report has not examined your application.","related_sources":[{"id":"sr-6a8f230bd09637a8ce7902d114d620f5","url":"https://github.com/advisories/GHSA-p92p-rxj5-7p2x","basis":"vendor_advisory","notes":"GitHub Advisory Database; CC-BY-4.0. Each advisory links its original attribution. Null normalizes fields.","title":"GHSA-p92p-rxj5-7p2x","publisher":"GitHub Advisory Database","observed_at":"2026-10-31T22:07:08.657Z"}],"affected_context":[{"certainty":"source_listed","source_ids":["sr-6a8f230bd09637a8ce7902d114d620f5"],"technology":"vllm","known_exclusions":[],"conditions_summary":"## Summary\n\nMultimodal content parts (`image_url`, `video_url`, `input_audio`, ...) accept a user-supplied `uuid` field that vLLM uses **verbatim as the item's cache identity** (`mm_hash`). The two frontend/engine processor caches (P0 \"sender\" cache in the API process, P1 \"receiver\" cache in the EngineCore process) are independent LRU caches whose evictions are **not synchronized**, and the P1 hit path performs **zero consistency validation** — it silently substitutes the cached item's tensor data for whatever the current request actually contains.\n\nA single unauthenticated client can exploit this to kill the entire engine, deterministically, with default server flags:\n\n1. **Seed**: send a small image with `uuid=\"U\"`. Both caches now map `U` to an item with **N_A** visual tokens (e.g. 4).\n2. **Drift**: send ~70 requests, each carrying a *unique* uuid, a max-size image, and an over-long text prompt. Each request is rejected with 400 — but **after** its image has been inserted into the P0 cache (insertion happens during chat rendering, before prompt-length validation), so these requests fill P0 **without ever reaching the engine/P1**. `U` is evicted from P0 only.\n3. **Kill**: send a ","public_affected_ranges":["< 0.31.0"]}],"null_published_at":"2026-10-31T22:27:09.625Z","source_updated_at":"2026-10-23T20:43:45.000Z","source_disclosed_at":"2026-10-23T20:43:44.000Z","is_synthetic_fixture":false,"null_last_reviewed_at":null,"public_change_summary":"First indexed by Null; source disclosure date is shown separately.","null_first_observed_at":"2026-10-31T22:07:08.657Z"}],"next_cursor":null,"snapshot_revision":"42594"}