fix:error when adding the ollama embedding model

[Howe829]

Checklist:

Important

Please review the checklist below before submitting your pull request.

• Please open an issue before creating a PR or link to an existing issue
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas
• I ran dev/reformat(backend) and cd web && npx lint-staged(frontend) to appease the lint gods

Description

Describe the big picture of your changes here to communicate to the maintainers why we should accept this pull request. If it fixes a bug or resolves a feature request, be sure to link to that issue. Close issue syntax: Fixes #<issue number>, see documentation for more details.

Fixes #8209

Type of Change

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• This change requires a documentation update, included: Dify Document
• Improvement, including but not limited to code refactoring, performance optimization, and UI/UX improvement
• Dependency upgrade

Testing Instructions

Please describe the tests that you ran to verify your changes. Provide instructions so we can reproduce. Please also list any relevant details for your test configuration

• Test A
• Test B

[AAEE86]

You can Fixes #8209 in Description.

[crazywoola]

It should be "true", otherwise it will report 500 error.

[crazywoola]

Maybe fix #6493

[Howe829]

It should be "true", otherwise it will report 500 error.

Hi, actually I think "true" will cause 500 error

@crazywoola

@AAEE86
Please check the ollama source code, you will see the UseMmap Option Field is of type *bool, which means it can be nil,true or false.it can't be a string.
https://github.com/ollama/ollama/blob/735a0ca2e480b40fc714751b73848c08cf4eed43/api/types.go#L242

[AAEE86]

Looks like it should be True.

You can create a new PR.

[kksmi]

I changed that place in [api/core/model_runtime/model_providers/ollama/text_embedding/text_embedding.py] to True, but it still reports an error when adding the embedded model.

[Howe829]

I changed that place in [api/core/model_runtime/model_providers/ollama/text_embedding/text_embedding.py] to True, but it still reports an error when adding the embedded model.

Could you please paste the error here? I think you can check whether you can connect to the Ollama API server or if the Base URL of the Ollama server is correct

[kksmi]

I changed that place in [api/core/model_runtime/model_providers/ollama/text_embedding/text_embedding.py] to True, but it still reports an error when adding the embedded model.

Could you please paste the error here? I think you can check whether you can connect to the Ollama API server or if the Base URL of the Ollama server is correct

The URL and name are correct and it can be added as LLM.

[AAEE86]

I changed that place in [api/core/model_runtime/model_providers/ollama/text_embedding/text_embedding.py] to True, but it still reports an error when adding the embedded model.

Could you please paste the error here? I think you can check whether you can connect to the Ollama API server or if the Base URL of the Ollama server is correct

The URL and name are correct and it can be added as LLM.

payload = {"input": inputs, "model": model, "options": {"use_mmap": True}}
# Make the request to the Ollama API
response = requests.post(endpoint_url, headers=headers, data=json.dumps(payload), timeout=(10, 300))

Are you sure it was modified according to the above code?

[Howe829]

I changed that place in [api/core/model_runtime/model_providers/ollama/text_embedding/text_embedding.py] to True, but it still reports an error when adding the embedded model.

Could you please paste the error here? I think you can check whether you can connect to the Ollama API server or if the Base URL of the Ollama server is correct

The URL and name are correct and it can be added as LLM.

Could you show me the code you changed? I guess you are running it with docker and the error you showed above is already fixed by this pr. The True or "true" problem is an other issue.

[kksmi]

我把[api/core/model_runtime/model_providers/ollama/text_embedding/text_embedding.py]中那地方改成了True，但是在添加嵌入模型的时候还是报错。

您能把错误粘贴到这里吗？我认为您可以检查是否可以连接到 Ollama API 服务器，或者 Ollama 服务器的基本 URL 是否正确

URL 和名称正确，可以添加为 LLM。

你能给我看一下你修改的代码吗？我猜你是用docker运行的，上面显示的错误已经被这个pr修复了。True或“true”问题是另一个问题。
Thank you very much for your guidance, I pulled the repository multiple times, updated the dify and as a result did not implement the update of that code, by manually modifying the file but got a 500 service error, finally modified True and he finally worked.

[liuxiaohao-xn]

i have change the file ,But when I uploaded the knowledge base，Still the following error occurs：

[kksmi]

我已经修改了文件，但是当我上传知识库时，仍然出现以下错误：

我也遇到了这个问题，刚刚发现

[AAEE86]

我觉得可以尝试将 "options": {"use_mmap": True} 删除，我不太清楚它在这里起什么作用？

[crazywoola]

我觉得可以尝试将 "options": {"use_mmap": True} 删除，我不太清楚它在这里起什么作用？

Try to fix this one. #8128

[kksmi]

i have change the file ,But when I uploaded the knowledge base，Still the following error occurs：

你去掉之后还会提示错误吗，我还是会，有部分文件不会，但是感觉没有索引成功

[Howe829]

i have change the file ,But when I uploaded the knowledge base，Still the following error occurs：

你去掉之后还会提示错误吗，我还是会，有部分文件不会，但是感觉没有索引成功

Did you restart your celery worker?
I can successfully index and do the retrieval

[kksmi]

我已经修改了文件，但是当我上传知识库时，仍然出现以下错误：

你去掉之后还会提示错误但是吗，我还是会，有部分文件不会，感觉没有索引成功

你重启了芹菜工人吗？ 我可以成功索引和搜索

I've found that I can't synchronize the code files within the api by updating according to the docs' update dedify prompts, so it often changes back to the original code.

[liuxiaohao-xn]

我已经更改了文件，但是当我上传知识库时，仍然出现以下错误：

你去掉之后还会提示错误吗，我还是会，有部分文件不会，但是感觉没有索引成功

你重新启动你的芹菜工人了吗？我可以成功索引并进行检索

我全部镜像都手动重启了，但是在创建知识库的时候还是会报这样的错误

[kksmi]

我已经更改了文件，但是当我上传知识库时，仍然出现以下错误：

你去掉之后还会提示错误吗，我还是会，有部分文件不会，但是感觉没有索引成功

你重新启动你的芹菜工人了吗？我可以成功索引并进行检索

我全部镜像都手动重启了，但是在创建知识库的时候还是会报这样的错误

手动重启之后，如果你是windows，需要修改worker和api下该文件，去掉该文件的的options就可以跑了

[kksmi]

我觉得可以尝试将 "options": {"use_mmap": True} 删除，我不太清楚它在这里起什么作用？

Try to fix this one. #8128

It turns out that removing the option is the only way to run it, the error may be related to his version of ollama, there is no need to include options, I hope to remove options in the next api image upgrade, it's not convenient for windows users, thank you very much.

[Howe829]

我觉得可以尝试将 "options": {"use_mmap": True} 删除，我不太清楚它在这里起什么作用？

Try to fix this one. #8128

It turns out that removing the option is the only way to run it, the error may be related to his version of ollama, there is no need to include options, I hope to remove options in the next api image upgrade, it's not convenient for windows users, thank you very much.

I think you can check the Ollama logs to see if there are any errors. I highly recommend upgrading Ollama to the latest version.

[verigle]

添加文档的时候依然会在80%左右的时候出现 ollama embedding 500 错误