Description
vLLM through 0.29.0 fails to properly clean up decode-side metadata for rejected inference requests in prefill/decode disaggregated deployments. Remote attackers can submit requests with max_tokens=0 to exhaust decode-worker memory without bound until the worker restarts.
No analysis available yet.
Remediation
No remediation available yet.
Tracking
Sign in to view the affected projects.
Advisories
No advisories yet.
References
History
Thu, 17 Sep 2026 22:30:00 +0000
| Type | Values Removed | Values Added |
|---|---|---|
| Description | vLLM through 0.29.0 fails to properly clean up decode-side metadata for rejected inference requests in prefill/decode disaggregated deployments. Remote attackers can submit requests with max_tokens=0 to exhaust decode-worker memory without bound until the worker restarts. | |
| Title | vLLM through 0.29.0 Memory Exhaustion via Rejected Requests | |
| First Time appeared |
Vllm
Vllm vllm |
|
| Weaknesses | CWE-401 | |
| CPEs | cpe:2.3:a:vllm:vllm:*:*:*:*:*:*:*:* | |
| Vendors & Products |
Vllm
Vllm vllm |
|
| References |
|
|
| Metrics |
cvssV3_1
|
Status: PUBLISHED
Assigner: VulnCheck
Published:
Updated: 2026-09-17T22:15:31.977Z
Reserved: 2026-09-17T21:50:02.601Z
Link: CVE-2026-93436
No data.
No data.
No data.
OpenCVE Enrichment
No data.
Weaknesses
-
CWE-401
Missing Release of Memory after Effective Lifetime