vLLM Vulnerable to Remote Code Execution via Mooncake Integration
Description
vLLM is a high-throughput and memory-efficient inference and serving engine for LLMs. Versions starting from 0.6.5 and prior to 0.8.5, having vLLM integration with mooncake, are vulnerable to remote code execution due to using pickle based serialization over unsecured ZeroMQ sockets. The vulnerable sockets were set to listen on all network interfaces, increasing the likelihood that an attacker is able to reach the vulnerable ZeroMQ sockets to carry out an attack. vLLM instances that do not make use of the mooncake integration are not vulnerable. This issue has been patched in version 0.8.5.
AI Insight
LLM-synthesized narrative grounded in this CVE's description and references.
Affected packages
Versions sourced from the GitHub Security Advisory.
| Package | Affected versions | Patched versions |
|---|---|---|
vllmPyPI | >= 0.6.5, < 0.8.5 | 0.8.5 |
Affected products
4- osv-coords3 versionspkg:apk/chainguard/py3.10-vllm-cuda-12.6pkg:apk/chainguard/tritonserver-backend-vllm-24.04pkg:pypi/vllm
< 0.8.5-r1+ 2 more
- (no CPE)range: < 0.8.5-r1
- (no CPE)range: < 24.04-r17
- (no CPE)range: >= 0.6.5, < 0.8.5
Patches
Vulnerability mechanics
References
7- github.com/advisories/GHSA-hj4w-hm2g-p6w5ghsaADVISORY
- nvd.nist.gov/vuln/detail/CVE-2025-32444ghsaADVISORY
- github.com/pypa/advisory-database/tree/main/vulns/vllm/PYSEC-2025-42.yamlghsaWEB
- github.com/vllm-project/vllm/blob/32b14baf8a1f7195ca09484de3008063569b43c5/vllm/distributed/kv_transfer/kv_pipe/mooncake_pipe.pyghsax_refsource_MISCWEB
- github.com/vllm-project/vllm/commit/a5450f11c95847cf51a17207af9a3ca5ab569b2cghsax_refsource_MISCWEB
- github.com/vllm-project/vllm/security/advisories/GHSA-hj4w-hm2g-p6w5ghsax_refsource_CONFIRMWEB
- github.com/vllm-project/vllm/security/advisories/GHSA-x3m8-f7g5-qhm7ghsax_refsource_MISCWEB
News mentions
0No linked articles in our index yet.