快速开始失败

[splashthelight]

1. chat出问题

2. 后端测试出问题

[skygazer42]

.env 是不是没有做 LLM_API_KEY 配置，我本地测没出现这种问题

[splashthelight]

我前端代理ip改为127.0.0.1就行了。
但是有另外一个问题

mysql里面是有数据的，milvus我不会看

附一张容器图

这是后端的提示信息

2026-04-04 07:55:44.982 | INFO     | gustobot.application.agents.kg_sub_graph.agentic_rag_agents.workflows.multi_agent.multi_tool:local_search:590 - 🔍 [优先] 查询 PostgreSQL pgvector 结构化数据库...

2026-04-04 07:55:45.000 | INFO     | gustobot.infrastructure.knowledge.embeddings:__init__:53 - Initialised OpenAI-compatible embeddings client (model=%s, base_url=%s)

2026-04-04 07:55:45.010 | INFO     | gustobot.infrastructure.knowledge.vector_store:_initialize:60 - Connected to Milvus at milvus:19530

2026-04-04 07:55:45.021 | INFO     | gustobot.infrastructure.knowledge.vector_store:_initialize:65 - Loaded existing collection: recipes

2026-04-04 07:55:45.030 | INFO     | gustobot.infrastructure.knowledge.reranker:__init__:34 - Reranker initialized: provider=%s, model=%s, base_url=%s

2026-04-04 07:55:45.031 | INFO     | gustobot.infrastructure.knowledge.knowledge_service:__init__:56 - KnowledgeService initialised (chunk_size=%s, chunk_overlap=%s)

2026-04-04 07:55:45.592 | INFO     | gustobot.infrastructure.knowledge.vector_store:search:217 - Found 0 results

INFO:     127.0.0.1:50330 - "POST /api/v1/knowledge/search HTTP/1.1" 200 OK

2026-04-04 07:55:45.593 | INFO     | gustobot.application.agents.kg_sub_graph.agentic_rag_agents.workflows.multi_agent.multi_tool:local_search:645 - ✅ PostgreSQL 返回 0 条结果，过滤后保留 0 条

2026-04-04 07:55:45.594 | WARNING  | gustobot.application.agents.kg_sub_graph.agentic_rag_agents.workflows.multi_agent.multi_tool:local_search:700 - ⚠️ 未选择任何可用的知识库工具

2026-04-04 07:55:45,599 INFO sqlalchemy.engine.Engine SELECT chat_messages.id AS chat_messages_id, chat_messages.session_id AS chat_messages_session_id, chat_messages.message_type AS chat_messages_message_type, chat_messages.content AS chat_messages_content, chat_messages.message_metadata AS chat_messages_message_metadata, chat_messages.order_index AS chat_messages_order_index, chat_messages.created_at AS chat_messages_created_at 

FROM chat_messages 

WHERE chat_messages.session_id = %(session_id_1)s ORDER BY chat_messages.order_index DESC 

 LIMIT %(param_1)s

2026-04-04 07:55:45,599 INFO sqlalchemy.engine.Engine [cached since 1263s ago] {'session_id_1': '41db0a40-520b-40bc-b120-5456f4485ee5', 'param_1': 1}

2026-04-04 07:55:45,601 INFO sqlalchemy.engine.Engine INSERT INTO chat_messages (session_id, message_type, content, message_metadata, order_index) VALUES (%(session_id)s, %(message_type)s, %(content)s, %(message_metadata)s, %(order_index)s)

2026-04-04 07:55:45,601 INFO sqlalchemy.engine.Engine [cached since 1263s ago] {'session_id': '41db0a40-520b-40bc-b120-5456f4485ee5', 'message_type': 'agent_response', 'content': '抱歉，菜谱文化知识库暂未找到相关记载，请尝试描述得更具体一些或稍后再试。', 'message_metadata': '{"session_id": "41db0a40-520b-40bc-b120-5456f4485ee5", "route": "kb-query"}', 'order_index': 8}

2026-04-04 07:55:45,603 INFO sqlalchemy.engine.Engine COMMIT

2026-04-04 07:55:45,610 INFO sqlalchemy.engine.Engine BEGIN (implicit)

2026-04-04 07:55:45,610 INFO sqlalchemy.engine.Engine SELECT chat_messages.id, chat_messages.session_id, chat_messages.message_type, chat_messages.content, chat_messages.message_metadata, chat_messages.order_index, chat_messages.created_at 

FROM chat_messages 

WHERE chat_messages.id = %(pk_1)s

2026-04-04 07:55:45,611 INFO sqlalchemy.engine.Engine [cached since 1263s ago] {'pk_1': 10}

2026-04-04 07:55:45,612 INFO sqlalchemy.engine.Engine SELECT chat_sessions.id AS chat_sessions_id, chat_sessions.user_id AS chat_sessions_user_id, chat_sessions.title AS chat_sessions_title, chat_sessions.created_at AS chat_sessions_created_at, chat_sessions.updated_at AS chat_sessions_updated_at, chat_sessions.is_active AS chat_sessions_is_active 

FROM chat_sessions 

WHERE chat_sessions.id = %(id_1)s 

 LIMIT %(param_1)s

2026-04-04 07:55:45,613 INFO sqlalchemy.engine.Engine [cached since 1263s ago] {'id_1': '41db0a40-520b-40bc-b120-5456f4485ee5', 'param_1': 1}

2026-04-04 07:55:45,615 INFO sqlalchemy.engine.Engine UPDATE chat_sessions SET updated_at=%(updated_at)s WHERE chat_sessions.id = %(chat_sessions_id)s

2026-04-04 07:55:45,615 INFO sqlalchemy.engine.Engine [cached since 1263s ago] {'updated_at': datetime.datetime(2026, 4, 4, 7, 55, 45, 614758), 'chat_sessions_id': '41db0a40-520b-40bc-b120-5456f4485ee5'}

2026-04-04 07:55:45,616 INFO sqlalchemy.engine.Engine COMMIT

2026-04-04 07:55:45,623 INFO sqlalchemy.engine.Engine BEGIN (implicit)

2026-04-04 07:55:45,624 INFO sqlalchemy.engine.Engine SELECT chat_sessions.id, chat_sessions.user_id, chat_sessions.title, chat_sessions.created_at, chat_sessions.updated_at, chat_sessions.is_active 

FROM chat_sessions 

WHERE chat_sessions.id = %(pk_1)s

2026-04-04 07:55:45,624 INFO sqlalchemy.engine.Engine [cached since 1263s ago] {'pk_1': '41db0a40-520b-40bc-b120-5456f4485ee5'}

2026-04-04 07:55:45,627 INFO sqlalchemy.engine.Engine ROLLBACK

INFO:     172.19.0.1:43414 - "POST /api/v1/chat/ HTTP/1.1" 200 OK

[skygazer42]

接口看没问题

[skygazer42]

可能路由到没数据的分支了，你换更好的模型试试比如 qwen-max