Access LLM by Ollama with error: [Errno -2] Name or service not known

[mobguang]

Hello @Aries-ckt ,

I launched DB-GPT 0.7.3 by docker compose on M2 ultra machine. And I added a LLM by ollama, please refer to the screenshot for the setting.

Then I tried to chat with LLM by "Chat Normal", and I got the error:

ERROR!Model server error!code=1, error msg is LLMServer Generate Error, Please CheckErrorInfo.: [Errno -2] Name or service not known

Please kindly provide the solution. Thanks in advance.

The docker log detail message as following (the error message at the end of the log):

2025-09-16 20:25:42.004 | 2025-09-16 12:25:42 b5d3d41263c5 dbgpt.model.cluster.worker.default_worker[1] INFO model_name: gpt-oss:120b, model_path: None, model_param_class: <class 'dbgpt.model.proxy.llms.ollama.OllamaDeployModelParameters'>
2025-09-16 20:25:42.004 | 2025-09-16 12:25:42 b5d3d41263c5 dbgpt.model.cluster.worker.manager[1] INFO Init empty instances list for gpt-oss:120b@llm
2025-09-16 20:25:42.004 | 2025-09-16 12:25:42 b5d3d41263c5 dbgpt.model.cluster.worker.manager[1] INFO Begin start all worker, apply_req: model='gpt-oss:120b' apply_type=<WorkerApplyType.START: 'start'> worker_type=<WorkerType.LLM: 'llm'> params={} apply_user=None
2025-09-16 20:25:42.004 | 2025-09-16 12:25:42 b5d3d41263c5 dbgpt.model.cluster.worker.manager[1] INFO Apply req: model='gpt-oss:120b' apply_type=<WorkerApplyType.START: 'start'> worker_type=<WorkerType.LLM: 'llm'> params={} apply_user=None, apply_func: <function LocalWorkerManager._start_all_worker.._start_worker at 0xffff56b1cd60>
2025-09-16 20:25:42.005 | 2025-09-16 12:25:42 b5d3d41263c5 dbgpt.model.cluster.worker.default_worker[1] INFO Begin load model, model params:
2025-09-16 20:25:42.005 |
2025-09-16 20:25:42.005 | =========================== OllamaDeployModelParameters ===========================
2025-09-16 20:25:42.005 |
2025-09-16 20:25:42.005 | name: gpt-oss:120b
2025-09-16 20:25:42.005 | provider: proxy/ollama
2025-09-16 20:25:42.005 | verbose: False
2025-09-16 20:25:42.005 | concurrency: 5
2025-09-16 20:25:42.005 | backend: None
2025-09-16 20:25:42.005 | prompt_template: None
2025-09-16 20:25:42.005 | context_length: None
2025-09-16 20:25:42.005 | reasoning_model: None
2025-09-16 20:25:42.005 | api_base: http://host.docker.internal:11434
2025-09-16 20:25:42.005 |
2025-09-16 20:25:42.005 | ======================================================================
2025-09-16 20:25:42.005 |
2025-09-16 20:25:42.005 |
2025-09-16 20:25:42.005 | 2025-09-16 12:25:42 b5d3d41263c5 dbgpt.model.adapter.proxy_adapter[1] INFO Load model from params:
2025-09-16 20:25:42.005 |
2025-09-16 20:25:42.005 | =========================== OllamaDeployModelParameters ===========================
2025-09-16 20:25:42.005 |
2025-09-16 20:25:42.005 | name: gpt-oss:120b
2025-09-16 20:25:42.005 | provider: proxy/ollama
2025-09-16 20:25:42.005 | verbose: False
2025-09-16 20:25:42.005 | concurrency: 5
2025-09-16 20:25:42.005 | backend: None
2025-09-16 20:25:42.005 | prompt_template: None
2025-09-16 20:25:42.005 | context_length: None
2025-09-16 20:25:42.005 | reasoning_model: None
2025-09-16 20:25:42.005 | api_base: http://host.docker.internal:11434
2025-09-16 20:25:42.005 |
2025-09-16 20:25:42.005 | ======================================================================
2025-09-16 20:25:42.005 |
2025-09-16 20:25:42.005 | llm client class: <class 'dbgpt.model.proxy.llms.ollama.OllamaLLMClient'>
2025-09-16 20:25:42.055 | INFO: 127.0.0.1:39858 - "POST /api/controller/models HTTP/1.1" 200 OK
2025-09-16 20:25:42.092 | 2025-09-16 12:25:42 b5d3d41263c5 dbgpt.model.cluster.worker.manager[1] INFO Model gpt-oss:120b startup successfully
2025-09-16 20:25:42.092 | 2025-09-16 12:25:42 b5d3d41263c5 dbgpt.model.cluster.worker.manager[1] INFO There has model storage, save model storage
2025-09-16 20:25:42.101 | 2025-09-16 12:25:42 b5d3d41263c5 dbgpt.model.cluster.worker.manager[1] INFO Save model storage successfully
2025-09-16 20:25:42.101 | INFO: 192.168.65.1:19177 - "POST /api/v2/serve/model/models HTTP/1.1" 200 OK
2025-09-16 20:25:42.104 | 2025-09-16 12:25:42 b5d3d41263c5 dbgpt.model.cluster.controller.controller[1] INFO Get all instances with WorkerManager@service, healthy_only: True
2025-09-16 20:25:42.104 | 2025-09-16 12:25:42 b5d3d41263c5 dbgpt.model.cluster.controller.controller[1] INFO Get all instances with None, healthy_only: False
2025-09-16 20:25:42.104 | INFO: 192.168.65.1:19177 - "GET /api/v2/serve/model/models HTTP/1.1" 200 OK
2025-09-16 20:25:47.448 | 2025-09-16 12:25:47 b5d3d41263c5 dbgpt.model.cluster.worker.manager[1] INFO Begin shutdown model, shutdown_req: host='172.18.0.3' port=5670 model='gpt-oss:120b' worker_type=<WorkerType.LLM: 'llm'> params={} delete_after=False worker_name=None sys_code=None user_name=None
2025-09-16 20:25:47.448 | 2025-09-16 12:25:47 b5d3d41263c5 dbgpt.model.cluster.worker.manager[1] INFO Apply req: model='gpt-oss:120b' apply_type=<WorkerApplyType.STOP: 'stop'> worker_type=<WorkerType.LLM: 'llm'> params={} apply_user=None, apply_func: <function LocalWorkerManager._stop_all_worker.._stop_worker at 0xffff56b1ce00>
2025-09-16 20:25:47.449 | 2025-09-16 12:25:47 b5d3d41263c5 dbgpt.util.model_utils[1] WARNING Torch not installed, skip clear torch cache
2025-09-16 20:25:47.497 | INFO: 127.0.0.1:39878 - "DELETE /api/controller/models?model_name=gpt-oss%3A120b%40llm&host=172.18.0.3&port=5670&weight=1.0&check_healthy=true&healthy=false&enabled=true&prompt_template=&last_heartbeat=&remove_from_registry=false HTTP/1.1" 200 OK
2025-09-16 20:25:47.497 | 2025-09-16 12:25:47 b5d3d41263c5 dbgpt.model.cluster.worker.manager[1] INFO Model gpt-oss:120b shutdown successfully
2025-09-16 20:25:47.498 | INFO: 192.168.65.1:20378 - "POST /api/v2/serve/model/models/stop HTTP/1.1" 200 OK
2025-09-16 20:25:47.501 | 2025-09-16 12:25:47 b5d3d41263c5 dbgpt.model.cluster.controller.controller[1] INFO Get all instances with WorkerManager@service, healthy_only: True
2025-09-16 20:25:47.501 | 2025-09-16 12:25:47 b5d3d41263c5 dbgpt.model.cluster.controller.controller[1] INFO Get all instances with None, healthy_only: False
2025-09-16 20:25:47.501 | INFO: 192.168.65.1:20378 - "GET /api/v2/serve/model/models HTTP/1.1" 200 OK
2025-09-16 20:25:51.096 | 2025-09-16 12:25:51 b5d3d41263c5 dbgpt.model.cluster.worker.default_worker[1] INFO model_name: gpt-oss:120b, model_path: None, model_param_class: <class 'dbgpt.model.proxy.llms.ollama.OllamaDeployModelParameters'>
2025-09-16 20:25:51.096 | 2025-09-16 12:25:51 b5d3d41263c5 dbgpt.model.cluster.worker.manager[1] INFO Init empty instances list for gpt-oss:120b@llm
2025-09-16 20:25:51.096 | 2025-09-16 12:25:51 b5d3d41263c5 dbgpt.model.cluster.worker.manager[1] INFO Begin start all worker, apply_req: model='gpt-oss:120b' apply_type=<WorkerApplyType.START: 'start'> worker_type=<WorkerType.LLM: 'llm'> params={} apply_user=None
2025-09-16 20:25:51.096 | 2025-09-16 12:25:51 b5d3d41263c5 dbgpt.model.cluster.worker.manager[1] INFO Apply req: model='gpt-oss:120b' apply_type=<WorkerApplyType.START: 'start'> worker_type=<WorkerType.LLM: 'llm'> params={} apply_user=None, apply_func: <function LocalWorkerManager._start_all_worker.._start_worker at 0xffff56b1c680>
2025-09-16 20:25:51.097 | 2025-09-16 12:25:51 b5d3d41263c5 dbgpt.model.cluster.worker.default_worker[1] INFO Begin load model, model params:
2025-09-16 20:25:51.097 |
2025-09-16 20:25:51.097 | =========================== OllamaDeployModelParameters ===========================
2025-09-16 20:25:51.097 |
2025-09-16 20:25:51.097 | name: gpt-oss:120b
2025-09-16 20:25:51.097 | provider: proxy/ollama
2025-09-16 20:25:51.097 | verbose: False
2025-09-16 20:25:51.097 | concurrency: 5
2025-09-16 20:25:51.097 | backend: None
2025-09-16 20:25:51.097 | prompt_template: None
2025-09-16 20:25:51.097 | context_length: None
2025-09-16 20:25:51.097 | reasoning_model: None
2025-09-16 20:25:51.097 | api_base: http://host.docker.internal:11434
2025-09-16 20:25:51.097 |
2025-09-16 20:25:51.097 | ======================================================================
2025-09-16 20:25:51.097 |
2025-09-16 20:25:51.097 |
2025-09-16 20:25:51.097 | 2025-09-16 12:25:51 b5d3d41263c5 dbgpt.model.adapter.proxy_adapter[1] INFO Load model from params:
2025-09-16 20:25:51.097 |
2025-09-16 20:25:51.097 | =========================== OllamaDeployModelParameters ===========================
2025-09-16 20:25:51.097 |
2025-09-16 20:25:51.097 | name: gpt-oss:120b
2025-09-16 20:25:51.097 | provider: proxy/ollama
2025-09-16 20:25:51.097 | verbose: False
2025-09-16 20:25:51.097 | concurrency: 5
2025-09-16 20:25:51.097 | backend: None
2025-09-16 20:25:51.097 | prompt_template: None
2025-09-16 20:25:51.097 | context_length: None
2025-09-16 20:25:51.097 | reasoning_model: None
2025-09-16 20:25:51.097 | api_base: http://host.docker.internal:11434
2025-09-16 20:25:51.097 |
2025-09-16 20:25:51.097 | ======================================================================
2025-09-16 20:25:51.097 |
2025-09-16 20:25:51.097 | llm client class: <class 'dbgpt.model.proxy.llms.ollama.OllamaLLMClient'>
2025-09-16 20:25:51.151 | INFO: 127.0.0.1:56450 - "POST /api/controller/models HTTP/1.1" 200 OK
2025-09-16 20:25:51.188 | 2025-09-16 12:25:51 b5d3d41263c5 dbgpt.model.cluster.worker.manager[1] INFO Model gpt-oss:120b startup successfully
2025-09-16 20:25:51.188 | 2025-09-16 12:25:51 b5d3d41263c5 dbgpt.model.cluster.worker.manager[1] INFO There has model storage, save model storage
2025-09-16 20:25:51.191 | 2025-09-16 12:25:51 b5d3d41263c5 dbgpt.model.cluster.worker.manager[1] INFO Save model storage successfully
2025-09-16 20:25:51.191 | INFO: 192.168.65.1:20378 - "POST /api/v2/serve/model/models/start HTTP/1.1" 200 OK
2025-09-16 20:25:51.194 | 2025-09-16 12:25:51 b5d3d41263c5 dbgpt.model.cluster.controller.controller[1] INFO Get all instances with WorkerManager@service, healthy_only: True
2025-09-16 20:25:51.194 | 2025-09-16 12:25:51 b5d3d41263c5 dbgpt.model.cluster.controller.controller[1] INFO Get all instances with None, healthy_only: False
2025-09-16 20:25:51.194 | INFO: 192.168.65.1:20378 - "GET /api/v2/serve/model/models HTTP/1.1" 200 OK
2025-09-16 20:25:54.558 | 2025-09-16 12:25:54 b5d3d41263c5 dbgpt.model.cluster.worker.manager[1] INFO Begin shutdown model, shutdown_req: host='172.18.0.3' port=5670 model='gpt-oss:120b' worker_type=<WorkerType.LLM: 'llm'> params={} delete_after=False worker_name=None sys_code=None user_name=None
2025-09-16 20:25:54.558 | 2025-09-16 12:25:54 b5d3d41263c5 dbgpt.model.cluster.worker.manager[1] INFO Apply req: model='gpt-oss:120b' apply_type=<WorkerApplyType.STOP: 'stop'> worker_type=<WorkerType.LLM: 'llm'> params={} apply_user=None, apply_func: <function LocalWorkerManager._stop_all_worker.._stop_worker at 0xffff56b1c680>
2025-09-16 20:25:54.559 | 2025-09-16 12:25:54 b5d3d41263c5 dbgpt.util.model_utils[1] WARNING Torch not installed, skip clear torch cache
2025-09-16 20:25:54.614 | INFO: 127.0.0.1:56482 - "DELETE /api/controller/models?model_name=gpt-oss%3A120b%40llm&host=172.18.0.3&port=5670&weight=1.0&check_healthy=true&healthy=false&enabled=true&prompt_template=&last_heartbeat=&remove_from_registry=false HTTP/1.1" 200 OK
2025-09-16 20:25:54.615 | 2025-09-16 12:25:54 b5d3d41263c5 dbgpt.model.cluster.worker.manager[1] INFO Model gpt-oss:120b shutdown successfully
2025-09-16 20:25:54.615 | INFO: 192.168.65.1:20378 - "POST /api/v2/serve/model/models/stop HTTP/1.1" 200 OK
2025-09-16 20:25:54.618 | 2025-09-16 12:25:54 b5d3d41263c5 dbgpt.model.cluster.controller.controller[1] INFO Get all instances with WorkerManager@service, healthy_only: True
2025-09-16 20:25:54.618 | 2025-09-16 12:25:54 b5d3d41263c5 dbgpt.model.cluster.controller.controller[1] INFO Get all instances with None, healthy_only: False
2025-09-16 20:25:54.618 | INFO: 192.168.65.1:20378 - "GET /api/v2/serve/model/models HTTP/1.1" 200 OK
2025-09-16 20:26:03.824 | 2025-09-16 12:26:03 b5d3d41263c5 dbgpt.model.cluster.worker.default_worker[1] INFO model_name: gpt-oss:120b, model_path: None, model_param_class: <class 'dbgpt.model.proxy.llms.ollama.OllamaDeployModelParameters'>
2025-09-16 20:26:03.825 | 2025-09-16 12:26:03 b5d3d41263c5 dbgpt.model.cluster.worker.manager[1] INFO Init empty instances list for gpt-oss:120b@llm
2025-09-16 20:26:03.826 | 2025-09-16 12:26:03 b5d3d41263c5 dbgpt.model.cluster.worker.manager[1] INFO Begin start all worker, apply_req: model='gpt-oss:120b' apply_type=<WorkerApplyType.START: 'start'> worker_type=<WorkerType.LLM: 'llm'> params={} apply_user=None
2025-09-16 20:26:03.827 | 2025-09-16 12:26:03 b5d3d41263c5 dbgpt.model.cluster.worker.manager[1] INFO Apply req: model='gpt-oss:120b' apply_type=<WorkerApplyType.START: 'start'> worker_type=<WorkerType.LLM: 'llm'> params={} apply_user=None, apply_func: <function LocalWorkerManager._start_all_worker.._start_worker at 0xffff56b1c7c0>
2025-09-16 20:26:03.827 | 2025-09-16 12:26:03 b5d3d41263c5 dbgpt.model.cluster.worker.default_worker[1] INFO Begin load model, model params:
2025-09-16 20:26:03.827 |
2025-09-16 20:26:03.827 | =========================== OllamaDeployModelParameters ===========================
2025-09-16 20:26:03.827 |
2025-09-16 20:26:03.827 | name: gpt-oss:120b
2025-09-16 20:26:03.827 | provider: proxy/ollama
2025-09-16 20:26:03.827 | verbose: False
2025-09-16 20:26:03.827 | concurrency: 5
2025-09-16 20:26:03.827 | backend: None
2025-09-16 20:26:03.827 | prompt_template: None
2025-09-16 20:26:03.827 | context_length: None
2025-09-16 20:26:03.827 | reasoning_model: None
2025-09-16 20:26:03.827 | api_base: http://host.docker.internal:11434
2025-09-16 20:26:03.827 |
2025-09-16 20:26:03.827 | ======================================================================
2025-09-16 20:26:03.827 |
2025-09-16 20:26:03.827 |
2025-09-16 20:26:03.828 | 2025-09-16 12:26:03 b5d3d41263c5 dbgpt.model.adapter.proxy_adapter[1] INFO Load model from params:
2025-09-16 20:26:03.828 |
2025-09-16 20:26:03.828 | =========================== OllamaDeployModelParameters ===========================
2025-09-16 20:26:03.828 |
2025-09-16 20:26:03.828 | name: gpt-oss:120b
2025-09-16 20:26:03.828 | provider: proxy/ollama
2025-09-16 20:26:03.828 | verbose: False
2025-09-16 20:26:03.828 | concurrency: 5
2025-09-16 20:26:03.828 | backend: None
2025-09-16 20:26:03.828 | prompt_template: None
2025-09-16 20:26:03.828 | context_length: None
2025-09-16 20:26:03.828 | reasoning_model: None
2025-09-16 20:26:03.828 | api_base: http://host.docker.internal:11434
2025-09-16 20:26:03.828 |
2025-09-16 20:26:03.828 | ======================================================================
2025-09-16 20:26:03.828 |
2025-09-16 20:26:03.828 | llm client class: <class 'dbgpt.model.proxy.llms.ollama.OllamaLLMClient'>
2025-09-16 20:26:03.867 | INFO: 127.0.0.1:40434 - "POST /api/controller/models HTTP/1.1" 200 OK
2025-09-16 20:26:03.902 | 2025-09-16 12:26:03 b5d3d41263c5 dbgpt.model.cluster.worker.manager[1] INFO Model gpt-oss:120b startup successfully
2025-09-16 20:26:03.902 | 2025-09-16 12:26:03 b5d3d41263c5 dbgpt.model.cluster.worker.manager[1] INFO There has model storage, save model storage
2025-09-16 20:26:03.906 | 2025-09-16 12:26:03 b5d3d41263c5 dbgpt.model.cluster.worker.manager[1] INFO Save model storage successfully
2025-09-16 20:26:03.906 | INFO: 192.168.65.1:61412 - "POST /api/v2/serve/model/models/start HTTP/1.1" 200 OK
2025-09-16 20:26:03.909 | 2025-09-16 12:26:03 b5d3d41263c5 dbgpt.model.cluster.controller.controller[1] INFO Get all instances with WorkerManager@service, healthy_only: True
2025-09-16 20:26:03.909 | 2025-09-16 12:26:03 b5d3d41263c5 dbgpt.model.cluster.controller.controller[1] INFO Get all instances with None, healthy_only: False
2025-09-16 20:26:03.910 | INFO: 192.168.65.1:61412 - "GET /api/v2/serve/model/models HTTP/1.1" 200 OK
2025-09-16 20:26:21.925 | INFO: 192.168.65.1:44556 - "GET /chat?scene=chat_normal&id=1af26d1c-92f8-11f0-96f8-4e7ee5565a65&model=gpt-oss:120b HTTP/1.1" 307 Temporary Redirect
2025-09-16 20:26:21.935 | INFO: 192.168.65.1:44556 - "GET /chat/?scene=chat_normal&id=1af26d1c-92f8-11f0-96f8-4e7ee5565a65&model=gpt-oss:120b HTTP/1.1" 200 OK
2025-09-16 20:26:22.116 | 2025-09-16 12:26:22 b5d3d41263c5 dbgpt_app.openapi.api_v1.api_v1[1] INFO /controller/model/types
2025-09-16 20:26:22.116 | 2025-09-16 12:26:22 b5d3d41263c5 dbgpt.model.cluster.controller.controller[1] INFO Get all instances with None, healthy_only: True
2025-09-16 20:26:22.118 | INFO: 192.168.65.1:44556 - "GET /api/v1/model/types HTTP/1.1" 200 OK
2025-09-16 20:26:22.122 | INFO: 192.168.65.1:40863 - "GET /api/v1/question/list?is_hot_question=true HTTP/1.1" 200 OK
2025-09-16 20:26:22.123 | INFO: 192.168.65.1:42591 - "GET /api/v1/chat/dialogue/list HTTP/1.1" 200 OK
2025-09-16 20:26:22.125 | 2025-09-16 12:26:22 b5d3d41263c5 dbgpt_serve.agent.app.controller[1] INFO app_detail:chat_normal,chat_normal
2025-09-16 20:26:22.132 | INFO: 192.168.65.1:40863 - "GET /api/v1/chat/dialogue/messages/history?con_uid=1af26d1c-92f8-11f0-96f8-4e7ee5565a65 HTTP/1.1" 200 OK
2025-09-16 20:26:22.133 | INFO: 192.168.65.1:42591 - "GET /api/v1/app/info?chat_scene=chat_normal&app_code=chat_normal HTTP/1.1" 200 OK
2025-09-16 20:26:25.779 | 2025-09-16 12:26:25 b5d3d41263c5 dbgpt_app.openapi.api_v1.api_v1[1] INFO chat_completions:chat_normal,,gpt-oss:120b, timestamp=1758025585778
2025-09-16 20:26:25.787 | 2025-09-16 12:26:25 b5d3d41263c5 dbgpt_app.openapi.api_v1.api_v1[1] INFO get_chat_instance:conv_uid='1af26d1c-92f8-11f0-96f8-4e7ee5565a65' user_input='hi' user_name='001' chat_mode='chat_normal' app_code='chat_normal' temperature=0.6 max_new_tokens=4000 select_param='' model_name='gpt-oss:120b' incremental=False sys_code=None prompt_code=None ext_info={}
2025-09-16 20:26:25.788 | 2025-09-16 12:26:25 b5d3d41263c5 dbgpt.core._private.prompt_registry[1] INFO Get prompt template of scene_name: chat_normal with model_name: gpt-oss:120b, proxyllm_backend: None, language: en
2025-09-16 20:26:25.801 | 2025-09-16 12:26:25 b5d3d41263c5 dbgpt.core.awel.runner.local_runner[1] INFO Begin run workflow from end operator, id: bd8c762e-9add-4fdf-8d9c-049b12cab7c7, runner: <dbgpt.core.awel.runner.local_runner.DefaultWorkflowRunner object at 0xffff6ff8a490>
2025-09-16 20:26:25.801 | INFO: 192.168.65.1:42591 - "POST /api/v1/chat/completions HTTP/1.1" 200 OK
2025-09-16 20:26:25.802 | 2025-09-16 12:26:25 b5d3d41263c5 dbgpt.core.awel.runner.local_runner[1] INFO Begin run workflow from end operator, id: 20c99281-7ffb-476e-b66a-db869381f0ec, runner: <dbgpt.core.awel.runner.local_runner.DefaultWorkflowRunner object at 0xffff6ff8a490>
2025-09-16 20:26:25.805 | 2025-09-16 12:26:25 b5d3d41263c5 dbgpt.model.utils.token_utils[1] INFO tiktoken installed, using it to count tokens, tiktoken will download tokenizer from network, also you can download it and put it in the directory of environment variable TIKTOKEN_CACHE_DIR
2025-09-16 20:26:25.807 | 2025-09-16 12:26:25 b5d3d41263c5 dbgpt_app.scene.base_chat[1] INFO payload request:
2025-09-16 20:26:25.807 | ModelRequest(model='gpt-oss:120b', messages=[ModelMessage(role='system', content='You are a helpful AI assistant.', round_index=0), ModelMessage(role='human', content='hi', round_index=1), ModelMessage(role='human', content='你好', round_index=2), ModelMessage(role='human', content='hi', round_index=0)], temperature=0.6, top_p=None, max_new_tokens=4000, stop=None, stop_token_ids=None, context_len=None, echo=False, span_id='ffdd00176b6524d53bb755761aae9940:8daa740f5e24edd9', context=ModelRequestContext(stream=True, cache_enable=False, user_name='001', sys_code=None, conv_uid=None, span_id='ffdd00176b6524d53bb755761aae9940:8daa740f5e24edd9', chat_mode='chat_normal', chat_param=None, extra={}, request_id=None, is_reasoning_model=False))
2025-09-16 20:26:25.809 | 2025-09-16 12:26:25 b5d3d41263c5 dbgpt.core.awel.runner.local_runner[1] INFO Begin run workflow from end operator, id: a319c470-e39f-4870-b918-34db2b84902e, runner: <dbgpt.core.awel.runner.local_runner.DefaultWorkflowRunner object at 0xffff6ff8a490>
2025-09-16 20:26:25.810 | 2025-09-16 12:26:25 b5d3d41263c5 dbgpt.core.awel.operators.common_operator[1] INFO branch_input_ctxs 0 result None, is_empty: False
2025-09-16 20:26:25.810 | 2025-09-16 12:26:25 b5d3d41263c5 dbgpt.core.awel.operators.common_operator[1] INFO Skip node name llm_model_cache_node
2025-09-16 20:26:25.810 | 2025-09-16 12:26:25 b5d3d41263c5 dbgpt.core.awel.operators.common_operator[1] INFO branch_input_ctxs 1 result True, is_empty: False
2025-09-16 20:26:25.810 | 2025-09-16 12:26:25 b5d3d41263c5 dbgpt.core.awel.runner.local_runner[1] INFO Skip node name llm_model_cache_node, node id 2d13e2f0-dcfd-4b5c-8b45-2ec6663184aa
2025-09-16 20:26:25.812 | 2025-09-16 12:26:25 b5d3d41263c5 dbgpt.model.adapter.base[1] INFO Message version is v2
2025-09-16 20:26:25.814 | 2025-09-16 12:26:25 b5d3d41263c5 dbgpt.model.cluster.worker.default_worker[1] INFO current generate stream function is synchronous generate stream function
2025-09-16 20:26:25.815 | 2025-09-16 12:26:25 b5d3d41263c5 dbgpt.model.cluster.worker.default_worker[1] INFO llm_adapter: <_DynProxyLLMModelAdapter model_name=gpt-oss:120b model_path=None>
2025-09-16 20:26:25.815 |
2025-09-16 20:26:25.815 | model prompt:
2025-09-16 20:26:25.815 |
2025-09-16 20:26:25.815 | system: You are a helpful AI assistant.
2025-09-16 20:26:25.815 | human: hi
2025-09-16 20:26:25.815 | human: 你好
2025-09-16 20:26:25.815 | human: hi
2025-09-16 20:26:25.815 |
2025-09-16 20:26:25.815 | generate stream output:
2025-09-16 20:26:25.815 |
2025-09-16 20:26:26.089 | 2025-09-16 12:26:26 b5d3d41263c5 dbgpt.model.cluster.worker.default_worker[1] ERROR Model inference error, detail: Traceback (most recent call last):
2025-09-16 20:26:26.089 | File "//opt/.uv.venv/lib/python3.11/site-packages/httpx/_transports/default.py", line 101, in map_httpcore_exceptions
2025-09-16 20:26:26.089 | yield
2025-09-16 20:26:26.089 | File "//opt/.uv.venv/lib/python3.11/site-packages/httpx/_transports/default.py", line 250, in handle_request
2025-09-16 20:26:26.089 | resp = self._pool.handle_request(req)
2025-09-16 20:26:26.089 | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
2025-09-16 20:26:26.089 | File "//opt/.uv.venv/lib/python3.11/site-packages/httpcore/_sync/connection_pool.py", line 256, in handle_request
2025-09-16 20:26:26.089 | raise exc from None
2025-09-16 20:26:26.089 | File "//opt/.uv.venv/lib/python3.11/site-packages/httpcore/_sync/connection_pool.py", line 236, in handle_request
2025-09-16 20:26:26.089 | response = connection.handle_request(
2025-09-16 20:26:26.089 | ^^^^^^^^^^^^^^^^^^^^^^^^^^
2025-09-16 20:26:26.089 | File "//opt/.uv.venv/lib/python3.11/site-packages/httpcore/_sync/connection.py", line 101, in handle_request
2025-09-16 20:26:26.089 | raise exc
2025-09-16 20:26:26.089 | File "//opt/.uv.venv/lib/python3.11/site-packages/httpcore/_sync/connection.py", line 78, in handle_request
2025-09-16 20:26:26.089 | stream = self._connect(request)
2025-09-16 20:26:26.089 | ^^^^^^^^^^^^^^^^^^^^^^
2025-09-16 20:26:26.089 | File "//opt/.uv.venv/lib/python3.11/site-packages/httpcore/_sync/connection.py", line 124, in _connect
2025-09-16 20:26:26.089 | stream = self._network_backend.connect_tcp(**kwargs)
2025-09-16 20:26:26.089 | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
2025-09-16 20:26:26.089 | File "//opt/.uv.venv/lib/python3.11/site-packages/httpcore/_backends/sync.py", line 207, in connect_tcp
2025-09-16 20:26:26.089 | with map_exceptions(exc_map):
2025-09-16 20:26:26.089 | File "/usr/lib/python3.11/contextlib.py", line 155, in exit
2025-09-16 20:26:26.089 | self.gen.throw(typ, value, traceback)
2025-09-16 20:26:26.089 | File "//opt/.uv.venv/lib/python3.11/site-packages/httpcore/_exceptions.py", line 14, in map_exceptions
2025-09-16 20:26:26.089 | raise to_exc(exc) from exc
2025-09-16 20:26:26.089 | httpcore.ConnectError: [Errno -2] Name or service not known
2025-09-16 20:26:26.089 |
2025-09-16 20:26:26.089 | The above exception was the direct cause of the following exception:
2025-09-16 20:26:26.089 |
2025-09-16 20:26:26.089 | Traceback (most recent call last):
2025-09-16 20:26:26.089 | File "/app/packages/dbgpt-core/src/dbgpt/model/cluster/worker/default_worker.py", line 178, in generate_stream
2025-09-16 20:26:26.089 | for output in generate_stream_func(
2025-09-16 20:26:26.089 | File "/app/packages/dbgpt-core/src/dbgpt/model/proxy/llms/ollama.py", line 56, in ollama_generate_stream
2025-09-16 20:26:26.089 | for r in client.sync_generate_stream(request):
2025-09-16 20:26:26.089 | File "/app/packages/dbgpt-core/src/dbgpt/model/proxy/llms/ollama.py", line 136, in sync_generate_stream
2025-09-16 20:26:26.089 | for chunk in stream:
2025-09-16 20:26:26.089 | File "//opt/.uv.venv/lib/python3.11/site-packages/ollama/_client.py", line 163, in inner
2025-09-16 20:26:26.089 | with self._client.stream(*args, **kwargs) as r:
2025-09-16 20:26:26.089 | File "/usr/lib/python3.11/contextlib.py", line 137, in enter
2025-09-16 20:26:26.089 | return next(self.gen)
2025-09-16 20:26:26.089 | ^^^^^^^^^^^^^^
2025-09-16 20:26:26.089 | File "//opt/.uv.venv/lib/python3.11/site-packages/httpx/_client.py", line 868, in stream
2025-09-16 20:26:26.089 | response = self.send(
2025-09-16 20:26:26.089 | ^^^^^^^^^^
2025-09-16 20:26:26.089 | File "//opt/.uv.venv/lib/python3.11/site-packages/httpx/_client.py", line 914, in send
2025-09-16 20:26:26.089 | response = self._send_handling_auth(
2025-09-16 20:26:26.089 | ^^^^^^^^^^^^^^^^^^^^^^^^^
2025-09-16 20:26:26.089 | File "//opt/.uv.venv/lib/python3.11/site-packages/httpx/_client.py", line 942, in _send_handling_auth
2025-09-16 20:26:26.089 | response = self._send_handling_redirects(
2025-09-16 20:26:26.089 | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
2025-09-16 20:26:26.089 | File "//opt/.uv.venv/lib/python3.11/site-packages/httpx/_client.py", line 979, in _send_handling_redirects
2025-09-16 20:26:26.089 | response = self._send_single_request(request)
2025-09-16 20:26:26.089 | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
2025-09-16 20:26:26.089 | File "//opt/.uv.venv/lib/python3.11/site-packages/httpx/_client.py", line 1014, in _send_single_request
2025-09-16 20:26:26.089 | response = transport.handle_request(request)
2025-09-16 20:26:26.089 | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
2025-09-16 20:26:26.089 | File "//opt/.uv.venv/lib/python3.11/site-packages/httpx/_transports/default.py", line 249, in handle_request
2025-09-16 20:26:26.089 | with map_httpcore_exceptions():
2025-09-16 20:26:26.089 | File "/usr/lib/python3.11/contextlib.py", line 155, in exit
2025-09-16 20:26:26.089 | self.gen.throw(typ, value, traceback)
2025-09-16 20:26:26.089 | File "//opt/.uv.venv/lib/python3.11/site-packages/httpx/_transports/default.py", line 118, in map_httpcore_exceptions
2025-09-16 20:26:26.089 | raise mapped_exc(message) from exc
2025-09-16 20:26:26.089 | httpx.ConnectError: [Errno -2] Name or service not known
2025-09-16 20:26:26.089 |
2025-09-16 20:26:26.115 | Traceback (most recent call last):
2025-09-16 20:26:26.115 | File "/app/packages/dbgpt-app/src/dbgpt_app/scene/base_chat.py", line 486, in stream_call
2025-09-16 20:26:26.115 | ai_response_text, view_message = await self._handle_final_output(
2025-09-16 20:26:26.115 | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
2025-09-16 20:26:26.115 | File "/app/packages/dbgpt-app/src/dbgpt_app/scene/base_chat.py", line 572, in _handle_final_output
2025-09-16 20:26:26.115 | parsed_output = self.prompt_template.output_parser.parse_model_nostream_resp(
2025-09-16 20:26:26.115 | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
2025-09-16 20:26:26.115 | File "/app/packages/dbgpt-core/src/dbgpt/core/interface/output_parser.py", line 134, in parse_model_nostream_resp
2025-09-16 20:26:26.115 | raise ValueError(
2025-09-16 20:26:26.115 | ValueError: Model server error!code=1, error msg is LLMServer Generate Error, Please CheckErrorInfo.: [Errno -2] Name or service not known
2025-09-16 20:26:26.115 |
2025-09-16 20:26:26.116 | 2025-09-16 12:26:26 b5d3d41263c5 dbgpt_app.scene.base_chat[1] ERROR model response parse failed！Model server error!code=1, error msg is LLMServer Generate Error, Please CheckErrorInfo.: [Errno -2] Name or service not known
2025-09-16 20:26:26.167 | INFO: 192.168.65.1:42591 - "GET /api/v1/chat/dialogue/list HTTP/1.1" 200 OK

[Aries-ckt]

can you try to set your llm config in DB-GPT/configs/dbgpt-proxy-ollama.toml

[mobguang]

@Aries-ckt

Yes. I set this config file already as following:

[system]

language = "${env:DBGPT_LANG:-en}"
api_keys = []
encrypt_key = "your_secret_key"

[service.web]
host = "0.0.0.0"
port = 5670

[service.web.database]
type = "mysql"
host = "${env:MYSQL_HOST:-127.0.0.1}"
port = "${env:MYSQL_PORT:-3306}"
database = "${env:MYSQL_DATABASE:-dbgpt}"
user = "${env:MYSQL_USER:-root}"
password ="${env:MYSQL_PASSWORD:-aa123456}"

[service.model.worker]
host = "127.0.0.1"

[rag.storage]
[rag.storage.vector]
type = "chroma"
persist_path = "pilot/data"

[models]
[[models.llms]]
name = "gpt-oss:120b"
provider = "proxy/ollama"
api_base = "${env:OLLAMA_API_BASE:-http://localhost:11434}"
api_key = ""

[[models.embeddings]]
name = "bge-m3:latest"
provider = "proxy/ollama"
api_url = "${env:OLLAMA_API_BASE:-http://localhost:11434}"
api_key = ""