Error determining wiki structure: 500

[alrex66608]

使用docker build -f Dockerfile-ollama-local -t deepwiki:ollama-local .构建镜像并运行，提示如下报错

2025-11-06 08:58:21,461 - WARNING - api.rag - rag.py:286 - Document 156 has empty embedding vector, skipping
2025-11-06 08:58:21,461 - WARNING - api.rag - rag.py:286 - Document 157 has empty embedding vector, skipping
2025-11-06 08:58:21,462 - WARNING - api.rag - rag.py:286 - Document 158 has empty embedding vector, skipping
2025-11-06 08:58:21,462 - ERROR - api.rag - rag.py:296 - No valid embeddings found in any documents
2025-11-06 08:58:21,462 - ERROR - api.simple_chat - simple_chat.py:117 - No valid embeddings found: No valid documents with embeddings found. Cannot create retriever.
INFO: 127.0.0.1:48564 - "POST /chat/completions/stream HTTP/1.1" 500 Internal Server Error

使用本地传统部署提示报错，似乎都是向量创建的问题
INFO: 127.0.0.1:43736 - "GET /auth/status HTTP/1.1" 200 OK INFO: 127.0.0.1:43736 - "GET /auth/status HTTP/1.1" 200 OK 2025-11-05 16:35:25,863 - INFO - api.simple_chat - simple_chat.py:85 - Request size: 2633 tokens 2025-11-05 16:35:25,876 - INFO - api.ollama_patch - ollama_patch.py:48 - Ollama model 'nomic-embed-text' is available 2025-11-05 16:35:25,877 - ERROR - api.simple_chat - simple_chat.py:123 - Error preparing retriever: 'embedder_ollama' INFO: 127.0.0.1:43736 - "POST /chat/completions/stream HTTP/1.1" 500 Internal Server Error

SOS/(ㄒoㄒ)/~~ 官方

[zhfly021]

我也是同样的问题，请问你解决了吗

[Dustyposa]

本地 ollama 能通嘛？

[alrex66608]

本地 ollama 能通嘛？

可以通，现在跑通了，但是似乎页面在提问的时候总是Error: WebSocket connection failed. Falling back to HTTP...，好慢！！！

[alrex66608]

我也是同样的问题，请问你解决了吗

修改embedder.json

{
    "embedder": {
        "client_class": "OllamaClient",
        "model_kwargs": {
            "model": "nomic-embed-text"
        }
    },
    "embedder_ollama": {
        "client_class": "OllamaClient",
        "model_kwargs": {
            "model": "nomic-embed-text"
        }
    },
    "retriever": {
        "top_k": 20
    },
    "text_splitter": {
        "split_by": "word",
        "chunk_size": 350,
        "chunk_overlap": 100
    }
}

你大概率还会遇到其他问题（后端正常运行，前端提示500），可以看这个issues #366

[9BestAdmin]

Такая же проблема, выбрана модель Gemini

[NuoLanC]

我也是同样的问题，请问你解决了吗

修改embedder.json

{
"embedder": {
"client_class": "OllamaClient",
"model_kwargs": {
"model": "nomic-embed-text"
}
},
"embedder_ollama": {
"client_class": "OllamaClient",
"model_kwargs": {
"model": "nomic-embed-text"
}
},
"retriever": {
"top_k": 20
},
"text_splitter": {
"split_by": "word",
"chunk_size": 350,
"chunk_overlap": 100
}
}
你大概率还会遇到其他问题（后端正常运行，前端提示500），可以看这个issues #366

大佬有部部署文档，我跑了一晚上没跑起来