#211 · open · 4 comments
**Describe the bug** main分支，按照readme的步骤拉起来docker服务后，执行"python -m src.client.agent_test --config configs/agents/api_agents.yaml --agent vllm-14b"正常对话。下一步执行python -m src.assigne没有进度，agent也没有调通。结果如下： <img width="1587" height="579" alt="Image" src="https://github.com/user-attachments/assets/c08d3e38-4b83-4f4b-bda0-bfc50e244193" /> **Desktop (please complete the following information):** - OS: [e.g. Ubuntu 22.04] - Python: [e.g. 3.9]
src\client\task.py worker["status"] 和 WorkerStatus.ALIVE.name 的类型不一致，做下兼容就行 <img width="718" height="376" alt="Image" src="https://github.com/user-attachments/assets/1f90d2a4-5943-48a9-83ff-16dde8616163" />
I suspect this is caused by a controller port mismatch in the FC (AgentRL) setup. The stack uses an AgentRL controller on **:5020**, but the default assignment definition points TaskClient to , so sees zero workers and appears stuck. I opened PR #215 to add an FC assignment preset ( + ) that uses , plus a README note and a clearer warning when the controller is unreachable.
I suspect this is caused by a controller port mismatch in the FC (AgentRL) setup. The extra/docker-compose.yml stack uses an AgentRL controller on :5020, but the default assignment definition points TaskClient to http://localhost:5000/api, so src.assigner sees zero workers and appears stuck. I opened PR #215 to add an FC assignment preset (configs/assignments/default_fc.yaml + definition_fc.yaml) that uses http://localhost:5020/api, plus a README note and a clearer warning when the controller is unreachable.
遇到同样的问题，请问怎么解决呢？我只想跑 os task，用的是 gpt-4o-mini，系统是 MacOS docker compose -f extra/docker-compose.yml up -d controller os_interaction-std redis python -m src.assigner --config configs/assignments/os_gpt4o_mini.yaml <img width="929" height="783" alt="Image" src="https://github.com/user-attachments/assets/41500706-5738-4625-87a7-bf3e8e65bef1" />