Skip to main content

Improper Neutralization of Special Elements used in a Command ('Command Injection')

CVE-2024-5480

High
10/10

Summary

A vulnerability in PyTorch's "torch.distributed.rpc" framework allows for Remote Code Execution (RCE). The framework, which is used in distributed training scenarios, does not properly verify the functions being called during RPC (Remote Procedure Call) operations. This oversight permits attackers to execute arbitrary commands by leveraging built-in Python functions such as "eval" during multi-cpu RPC communication. The vulnerability arises from the lack of restriction on function calls when a worker node serializes and sends a PythonUDF (User Defined Function) to the master node, which then deserializes and executes the function without validation. This flaw can be exploited to compromise master nodes initiating distributed training, potentially leading to the theft of sensitive AI-related data.

  • LOW
  • NETWORK
  • HIGH
  • CHANGED
  • NONE
  • NONE
  • HIGH
  • HIGH

CWE-77 - Command Injection

A command injection attack involves injecting an operating system command through the data input, which gets executed on the host operating system with the privileges of the victimized application. The impact of a command injection attack may range from loss of data confidentiality and integrity to unauthorized remote access to the hosting system. The attack may cause serious data breaches and system takeover.

References

Advisory Timeline

  • Published