Deserialization of Untrusted Data Affecting vllm package, versions [,0.8.0)

vllm is an A high-throughput and memory-efficient inference and serving engine for LLMs

Affected versions of this package are vulnerable to Deserialization of Untrusted Data in the MooncakePipe class, which relies on pickle for serialization and deserialization in recv_tensor(). An attacker can execute arbitrary code on the host by sending malicious payloads using the ZMQ over TCP interface, which is exposed by the Mooncake integration by default.

Serialization is a process of converting an object into a sequence of bytes which can be persisted to a disk or database or can be sent through streams. The reverse process of creating object from sequence of bytes is called deserialization. Serialization is commonly used for communication (sharing objects between multiple hosts) and persistence (store the object state in a file or a database). It is an integral part of popular protocols like Remote Method Invocation (RMI), Java Management Extension (JMX), Java Messaging System (JMS), Action Message Format (AMF), Java Server Faces (JSF) ViewState, etc.

Deserialization of untrusted data (CWE-502) is when the application deserializes untrusted data without sufficiently verifying that the resulting data will be valid, thus allowing the attacker to control the state or the flow of the execution.

• GitHub Commit
• GitHub PR