After the stream output is stuck, spit out data at one time #35

Closed
opened 2026-02-16 10:17:49 -05:00 by yindo · 10 comments
Owner

Originally created by @teasong on GitHub (Mar 3, 2025).

Self Checks

  • This is only for bug report, if you would like to ask a question, please head to Discussions.
  • I have searched for existing issues search for existing issues, including closed ones.
  • I confirm that I am using English to submit this report (我已阅读并同意 Language Policy).
  • [FOR CHINESE USERS] 请务必使用英文提交 Issue,否则会被关闭。谢谢!:)
  • Please do not modify this template :) and fill in all the required fields.

Dify version

0.15.3

Cloud or Self Hosted

Self Hosted (Docker)

Steps to reproduce

Image

Image

✔️ Expected Behavior

data one by one

Actual Behavior

spit out data at one time

Originally created by @teasong on GitHub (Mar 3, 2025). ### Self Checks - [x] This is only for bug report, if you would like to ask a question, please head to [Discussions](https://github.com/langgenius/dify/discussions/categories/general). - [x] I have searched for existing issues [search for existing issues](https://github.com/langgenius/dify/issues), including closed ones. - [x] I confirm that I am using English to submit this report (我已阅读并同意 [Language Policy](https://github.com/langgenius/dify/issues/1542)). - [x] [FOR CHINESE USERS] 请务必使用英文提交 Issue,否则会被关闭。谢谢!:) - [x] Please do not modify this template :) and fill in all the required fields. ### Dify version 0.15.3 ### Cloud or Self Hosted Self Hosted (Docker) ### Steps to reproduce ![Image](https://github.com/user-attachments/assets/5bcd9b0f-a7bb-4555-8cf5-16e758c55cff) ![Image](https://github.com/user-attachments/assets/3b71fd81-881f-464b-8e19-114014e8a2e8) ### ✔️ Expected Behavior data one by one ### ❌ Actual Behavior spit out data at one time
yindo added the good first issue label 2026-02-16 10:17:49 -05:00
yindo closed this issue 2026-02-16 10:17:49 -05:00
Author
Owner

@dosubot[bot] commented on GitHub (Mar 3, 2025):

Hello @teasong! 👋 I'm here to help you with bugs, questions, and more while you wait for a human maintainer. Let's squash those bugs together!

In Dify version 0.15.3, when self-hosted using Docker, workflows that include conditional branches, such as If-Else nodes before an LLM node, cannot stream output and instead switch to a blocking response. This behavior is expected and confirmed by the project maintainers. To address this issue, you can either wait for a future official Docker release that includes a fix or manually update the code in base_stream_processor.py within the Docker container and restart the container [1][2].

To continue talking to Dosu, mention @dosu.


Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Bug Report | Other

@dosubot[bot] commented on GitHub (Mar 3, 2025): <!-- Greeting --> Hello @teasong! 👋 I'm here to help you with bugs, questions, and more while you wait for a human maintainer. Let's squash those bugs together! <!-- Answer --> In Dify version 0.15.3, when self-hosted using Docker, workflows that include conditional branches, such as If-Else nodes before an LLM node, cannot stream output and instead switch to a blocking response. This behavior is expected and confirmed by the project maintainers. To address this issue, you can either wait for a future official Docker release that includes a fix or manually update the code in `base_stream_processor.py` within the Docker container and restart the container <sup>[[1]](https://github.com/langgenius/dify/issues/14010)</sup><sup>[[2]](https://github.com/langgenius/dify/issues/12362)</sup>. <!-- Dosu Comment Footer --> *To continue talking to [Dosu](https://dosu.dev), mention @dosu.* --- <sup>Help Dosu learn! Give it feedback: [Great Response](https://app.dosu.dev/response-feedback/3b26b4b6-2b21-4843-a8d2-b3ddaf9493c2?feedback_type=great_response) | [Irrelevant Answer](https://app.dosu.dev/response-feedback/3b26b4b6-2b21-4843-a8d2-b3ddaf9493c2?feedback_type=irrelevant_answer) | [Incorrect Sources](https://app.dosu.dev/response-feedback/3b26b4b6-2b21-4843-a8d2-b3ddaf9493c2?feedback_type=incorrect_sources) | [Too Verbose](https://app.dosu.dev/response-feedback/3b26b4b6-2b21-4843-a8d2-b3ddaf9493c2?feedback_type=too_verbose) | [Hallucination](https://app.dosu.dev/response-feedback/3b26b4b6-2b21-4843-a8d2-b3ddaf9493c2?feedback_type=hallucination) | [Bug Report](https://app.dosu.dev/response-feedback/3b26b4b6-2b21-4843-a8d2-b3ddaf9493c2?feedback_type=bug_report) | [Other](https://app.dosu.dev/response-feedback/3b26b4b6-2b21-4843-a8d2-b3ddaf9493c2?feedback_type=other)</sup>
Author
Owner

@crazywoola commented on GitHub (Mar 3, 2025):

Judging by the response content, it is a stream event.

@crazywoola commented on GitHub (Mar 3, 2025): Judging by the response content, it is a stream event.
Author
Owner

@teasong commented on GitHub (Mar 3, 2025):

When he thinks, he outputs slowly, but when he comes to real data, he spits out all the data at once.like the pic 2,all data in red circle

@teasong commented on GitHub (Mar 3, 2025): When he thinks, he outputs slowly, but when he comes to real data, he spits out all the data at once.like the pic 2,all data in red circle
Author
Owner

@cuckootan commented on GitHub (Mar 14, 2025):

same question.

Image
@cuckootan commented on GitHub (Mar 14, 2025): same question. <img width="1079" alt="Image" src="https://github.com/user-attachments/assets/95215e18-a4a1-4f6f-bd0e-fd24959f1d9e" />
Author
Owner

@cuckootan commented on GitHub (Mar 14, 2025):

0.15.3

@cuckootan commented on GitHub (Mar 14, 2025): 0.15.3
Author
Owner

@cowcomic commented on GitHub (Mar 17, 2025):

1.0.1 There is also the same problem

The ollama model is configured to support function calls but still doesn't work

Use the OpenAI-API-compatible mode to connect to ollama, configured Function calling=tool call,Stream function calling=support, but it still doesn't work

@cowcomic commented on GitHub (Mar 17, 2025): 1.0.1 There is also the same problem The ollama model is configured to support function calls but still doesn't work Use the OpenAI-API-compatible mode to connect to ollama, configured Function calling=tool call,Stream function calling=support, but it still doesn't work
Author
Owner

@ml4165837 commented on GitHub (Apr 23, 2025):

Surprisingly, it hasn't been resolved after crossing three versions

@ml4165837 commented on GitHub (Apr 23, 2025): Surprisingly, it hasn't been resolved after crossing three versions
Author
Owner

@alterxyz commented on GitHub (Jun 17, 2025):

It seems like this only happens with certain LLMs, not all. We might have to look into those model plugin one by one.

@alterxyz commented on GitHub (Jun 17, 2025): It seems like this only happens with certain LLMs, not all. We might have to look into those model plugin one by one.
Author
Owner

@alterxyz commented on GitHub (Jun 17, 2025):

Still exist like DeepSeek (Provider) on V1.x

@alterxyz commented on GitHub (Jun 17, 2025): Still exist like DeepSeek (Provider) on V1.x
Author
Owner

@dosubot[bot] commented on GitHub (Aug 21, 2025):

Hi, @teasong. I'm Dosu (https://dosu.dev), and I'm helping the dify-official-plugins team manage their backlog and am marking this issue as stale.

Issue Summary:

  • You reported that in version 0.15.3 (self-hosted via Docker), stream output gets stuck and then releases all data at once instead of incrementally.
  • It was clarified that this behavior is expected when workflows include conditional branches before an LLM node, with a suggested manual code update as a workaround.
  • Other users confirmed the issue persists in later versions (including 1.0.1) and with certain LLMs or models like Ollama and DeepSeek.
  • Maintainers acknowledge the problem may be model-specific and requires further investigation.

Next Steps:

  • Please let me know if this issue is still relevant with the latest version of dify-official-plugins by commenting here to keep the discussion open.
  • Otherwise, this issue will be automatically closed in 5 days.

Thank you for your understanding and contribution!

@dosubot[bot] commented on GitHub (Aug 21, 2025): Hi, @teasong. I'm Dosu (https://dosu.dev), and I'm helping the dify-official-plugins team manage their backlog and am marking this issue as stale. **Issue Summary:** - You reported that in version 0.15.3 (self-hosted via Docker), stream output gets stuck and then releases all data at once instead of incrementally. - It was clarified that this behavior is expected when workflows include conditional branches before an LLM node, with a suggested manual code update as a workaround. - Other users confirmed the issue persists in later versions (including 1.0.1) and with certain LLMs or models like Ollama and DeepSeek. - Maintainers acknowledge the problem may be model-specific and requires further investigation. **Next Steps:** - Please let me know if this issue is still relevant with the latest version of dify-official-plugins by commenting here to keep the discussion open. - Otherwise, this issue will be automatically closed in 5 days. Thank you for your understanding and contribution!
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: langgenius/dify-official-plugins#35