Service interruption via unbounded grammar cache growth
Published Mar 19, 2025 · Updated Mar 19, 2025
Resource exhaustion in vLLM before 0.8.0 allows remote authenticated users to interrupt service through guided-decoding requests. The V0 outlines logits processor unconditionally enables the outlines filesystem cache, and every unique schema adds another compiled grammar without a size limit. Any client allowed to call the OpenAI-compatible API can select the outlines backend per request and fill the filesystem, stopping the server and colocated disk-dependent workloads.
Summary
What happened
Resource exhaustion in vLLM before 0.8.0 allows remote authenticated users to interrupt service through guided-decoding requests. The V0 outlines logits processor unconditionally enables the outlines filesystem cache, and every unique schema adds another compiled grammar without a size limit. Any client allowed to call the OpenAI-compatible API can select the outlines backend per request and fill the filesystem, stopping the server and colocated disk-dependent workloads.
The record
- CVE
- CVE-2025-29770
- Published
- Mar 19, 2025
- Updated
- Mar 19, 2025
- Vendor
- vLLM Project
- Product
- vLLM
- Classifications
- CWE-770, T1499.003
- Attack vector
- network
- Privileges
- authenticated
Timeline
How it unfolded
- Mar 19, 2025CVE publishedPublication date reported by the CVE source.
- Mar 19, 2025Record updatedLatest update available in the CVE record.
Exploitability
Present is not the same as exploitable
Compare your product and version with the public record. A matching version still requires validation against your environment.
Is a vulnerable build present?
Compare these published version ranges with your installed build and any vendor patches.
- Affected versionversion=< 0.8.0
What conditions does exploitation require?
What is affected?
Published CVSS scores
CVSS describes severity. EPSS estimates exploitation probability.
Attacks
What attackers are doing with it
Daily unique IPs observed by Shadowserver honeypots for known exploited vulnerabilities (KEVs). Missing observations do not establish an absence of attacks.
Weakness, pattern, technique
Public exploit references
No public exploit references are available in this record.
Labels summarize the accepted research assessment. They do not indicate a test against your environment.
Technologies
Your stack
See the directory against your own environment.
Your stack
Check the software in your environment
Book a demo to see how Hinoki identifies affected software and validates exploitability in your environment.
Book a demo