Description
LightLLM through 1.2.0 mounts reinforcement learning control routes on the public HTTP API without authentication checks. Unauthenticated attackers can call endpoints like /pause_generation, /abort_request, /flush_cache, and /init_weights_update_group to disrupt inference operations and wedge workers on deployments started with --enable_rl.
Affected products
| Vendor | Product | Affected | Fixed |
|---|---|---|---|
| ModelTC | LightLLM | ≤ 1.2.0 | — |
References
- GitHub Issue #1609 issue-tracking
- https://github.com/ModelTC/lightllm/blob/v1.2.0/lightllm/server/api_http.py#L505-L507 technical-description
- https://github.com/ModelTC/lightllm/blob/v1.2.0/lightllm/server/api_http_rl.py#L51-L139 technical-description
- https://github.com/ModelTC/LightLLM product
- VulnCheck Advisory: LightLLM through 1.2.0 Missing Authentication on RL Control Routes third-party-advisory