langgenius/dify-official-plugins
1
0
Fork 0
mirror of https://github.com/langgenius/dify-official-plugins.git synced 2026-07-22 01:55:27 -04:00
Code Issues 54 Packages Projects Releases Wiki Activity

Disconnected from client (via refresh/close) #359

New Issue
Closed
opened 2026-02-16 10:19:07 -05:00 by yindo · 7 comments
yindo commented 2026-02-16 10:19:07 -05:00
Owner
Copy Link
Copy Source

Originally created by @lhtpluto on GitHub (Jun 10, 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 Dify issues & Dify Official Plugins, 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

1.4.1

Plugin version

Xorbits Inference 0.0.3

Cloud or Self Hosted

Self Hosted (Source)

Steps to reproduce

中文：
使用 Xinference 插件时，如使用Qwen3-235B-A22B-4bit 模型时，输入18K的文字，可以正常反馈，但使用Qwen3-235B-A22B-8bit和DeepSeek-R1-0528-4bit模型时，如输入18K的文字，就会主动断开与Xinference服务器的连接，Xinference服务器显示“Disconnected from client (via refresh/close)”

English：
When using the Xinference plugin, with the Qwen3-235B-A22B-4bit model, inputting 18K characters will function normally. However, when using the Qwen3-235B-A22B-8bit and DeepSeek-R1-0528-4bit models, inputting 18K characters will cause the connection to the Xinference server to be actively disconnected, with the Xinference server displaying "Disconnected from client (via refresh/close)".

中文：
以下是Xinference服务器日志截图
English：
Here is a screenshot of the Xinference server logs.

Image

中文：
以下是 dify 日志截图，红色方框内的日志是对端 Xinference 服务器显示 "Disconnected from client (via refresh/close)" 附近时间段对应的 DIFY 的日志。

English：
The following is a screenshot of the Dify logs. The logs within the red box correspond to the Dify logs during the time period when the remote Xinference server displayed "Disconnected from client (via refresh/close)".

Image

✔️ Error log

No response

Originally created by @lhtpluto on GitHub (Jun 10, 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 [Dify issues](https://github.com/langgenius/dify/issues) & [Dify Official Plugins](https://github.com/langgenius/dify-official-plugins/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 1.4.1 ### Plugin version Xorbits Inference 0.0.3 ### Cloud or Self Hosted Self Hosted (Source) ### Steps to reproduce 中文： 使用 Xinference 插件时，如使用Qwen3-235B-A22B-4bit 模型时，输入18K的文字，可以正常反馈，但使用Qwen3-235B-A22B-8bit和DeepSeek-R1-0528-4bit模型时，如输入18K的文字，就会主动断开与Xinference服务器的连接，Xinference服务器显示“Disconnected from client (via refresh/close)” English： When using the Xinference plugin, with the Qwen3-235B-A22B-4bit model, inputting 18K characters will function normally. However, when using the Qwen3-235B-A22B-8bit and DeepSeek-R1-0528-4bit models, inputting 18K characters will cause the connection to the Xinference server to be actively disconnected, with the Xinference server displaying "Disconnected from client (via refresh/close)". 中文： 以下是Xinference服务器日志截图 English： Here is a screenshot of the Xinference server logs. ![Image](https://github.com/user-attachments/assets/e917bfde-78d8-418f-9684-6b95546ce13b) 中文： 以下是 dify 日志截图，红色方框内的日志是对端 Xinference 服务器显示 "Disconnected from client (via refresh/close)" 附近时间段对应的 DIFY 的日志。 English： The following is a screenshot of the Dify logs. The logs within the red box correspond to the Dify logs during the time period when the remote Xinference server displayed "Disconnected from client (via refresh/close)". ![Image](https://github.com/user-attachments/assets/802561b5-4dd6-4b3c-b365-b5d6d0ae8072) ### ✔️ Error log _No response_
yindo added the bug label 2026-02-16 10:19:07 -05:00
yindo closed this issue 2026-02-16 10:19:07 -05:00
yindo commented 2026-02-16 10:19:07 -05:00
Author
Owner
Copy Link
Copy Source

@lhtpluto commented on GitHub (Jun 10, 2025):

I discovered that LM Studio also encountered this issue.

2025-06-11 00:04:25 [INFO]
[LM STUDIO SERVER] Running chat completion on conversation with 1 messages.
2025-06-11 00:04:25 [INFO]
[LM STUDIO SERVER] Streaming response...
2025-06-11 00:04:25 [DEBUG]
[CacheWrapper][INFO] Trimmed 1992 tokens from the prompt cache
2025-06-11 00:09:25 [INFO]
[LM STUDIO SERVER] Client disconnected. Stopping generation... (If the model is busy processing the prompt, it will finish first.)
2025-06-11 00:12:48 [INFO]
Finished streaming response

@lhtpluto commented on GitHub (Jun 10, 2025): I discovered that LM Studio also encountered this issue. 2025-06-11 00:04:25 [INFO] [LM STUDIO SERVER] Running chat completion on conversation with 1 messages. 2025-06-11 00:04:25 [INFO] [LM STUDIO SERVER] Streaming response... 2025-06-11 00:04:25 [DEBUG] [CacheWrapper][INFO] Trimmed 1992 tokens from the prompt cache 2025-06-11 00:09:25 [INFO] [LM STUDIO SERVER] Client disconnected. Stopping generation... (If the model is busy processing the prompt, it will finish first.) 2025-06-11 00:12:48 [INFO] Finished streaming response
yindo commented 2026-02-16 10:19:07 -05:00
Author
Owner
Copy Link
Copy Source

@lhtpluto commented on GitHub (Jun 11, 2025):

中文:
ollama插件（Qwen3-235B-A22B-Q8），输入18K内容后，LLM使用正常。

English：
The ollama plugin （Qwen3-235B-A22B-Q8） works normally with the LLM after inputting 18K content.

Image

@lhtpluto commented on GitHub (Jun 11, 2025): 中文: ollama插件（Qwen3-235B-A22B-Q8），输入18K内容后，LLM使用正常。 English： The ollama plugin （Qwen3-235B-A22B-Q8） works normally with the LLM after inputting 18K content. ![Image](https://github.com/user-attachments/assets/8462e596-1734-4ed3-8b92-7685173948b8)
yindo commented 2026-02-16 10:19:07 -05:00
Author
Owner
Copy Link
Copy Source

@dosubot[bot] commented on GitHub (Aug 18, 2025):

Hi, @lhtpluto. I'm Dosu, and I'm helping the dify-official-plugins team manage their backlog and am marking this issue as stale.

Issue Summary:

  • You reported that using the Xinference plugin with Qwen3-235B-A22B-8bit and DeepSeek-R1-0528-4bit models causes the server to disconnect when inputting 18K characters.
  • The 4bit Qwen3 model works fine under the same conditions.
  • LM Studio also experiences similar disconnection issues with these models at 18K input length.
  • The ollama plugin with Qwen3-235B-A22B-Q8 handles 18K input normally without disconnecting.
  • You provided logs and screenshots to illustrate the problem.

Next Steps:

  • Please let me know if this issue is still relevant with the latest version of the dify-official-plugins repository by commenting here.
  • If I do not hear back within 5 days, I will automatically close this issue.

Thank you for your understanding and contribution!

@dosubot[bot] commented on GitHub (Aug 18, 2025): Hi, @lhtpluto. 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 using the Xinference plugin with Qwen3-235B-A22B-8bit and DeepSeek-R1-0528-4bit models causes the server to disconnect when inputting 18K characters. - The 4bit Qwen3 model works fine under the same conditions. - LM Studio also experiences similar disconnection issues with these models at 18K input length. - The ollama plugin with Qwen3-235B-A22B-Q8 handles 18K input normally without disconnecting. - You provided logs and screenshots to illustrate the problem. **Next Steps:** - Please let me know if this issue is still relevant with the latest version of the dify-official-plugins repository by commenting here. - If I do not hear back within 5 days, I will automatically close this issue. Thank you for your understanding and contribution!
yindo commented 2026-02-16 10:19:07 -05:00
Author
Owner
Copy Link
Copy Source

@lhtpluto commented on GitHub (Aug 28, 2025):

你好，@lhtpluto。我是Dosu，我正在帮助 dify-official-plugins 团队管理他们的积压问题，并将此问题标记为陈旧。

问题摘要：

  • 您报告说，使用带有 Qwen3-235B-A22B-8bit 和 DeepSeek-R1-0528-4bit 型号的 Xinference 插件会导致服务器在输入 18K 个字符时断开连接。
  • 4bit Qwen3型号在相同条件下工作正常。
  • LM Studio 在 18K 输入长度时也遇到了这些模型的类似断开连接问题。
  • 带有 Qwen3-235B-A22B-Q8 的 ollama 插件可以正常处理 18K 输入而无需断开连接。
  • 您提供了日志和屏幕截图来说明问题。

后续步骤：

  • 请在此处发表评论，让我知道该问题是否仍然与 dify-official-plugins 存储库的最新版本有关。
  • 如果我在 5 天内没有收到回复，我将自动关闭此问题。

感谢您的理解和贡献！

问题依旧

@lhtpluto commented on GitHub (Aug 28, 2025): > 你好，[@lhtpluto](https://github.com/lhtpluto)。我是[Dosu](https://dosu.dev)，我正在帮助 dify-official-plugins 团队管理他们的积压问题，并将此问题标记为陈旧。 > > **问题摘要：** > > * 您报告说，使用带有 Qwen3-235B-A22B-8bit 和 DeepSeek-R1-0528-4bit 型号的 Xinference 插件会导致服务器在输入 18K 个字符时断开连接。 > * 4bit Qwen3型号在相同条件下工作正常。 > * LM Studio 在 18K 输入长度时也遇到了这些模型的类似断开连接问题。 > * 带有 Qwen3-235B-A22B-Q8 的 ollama 插件可以正常处理 18K 输入而无需断开连接。 > * 您提供了日志和屏幕截图来说明问题。 > > **后续步骤：** > > * 请在此处发表评论，让我知道该问题是否仍然与 dify-official-plugins 存储库的最新版本有关。 > * 如果我在 5 天内没有收到回复，我将自动关闭此问题。 > > 感谢您的理解和贡献！ 问题依旧
yindo commented 2026-02-16 10:19:07 -05:00
Author
Owner
Copy Link
Copy Source

@dosubot[bot] commented on GitHub (Aug 28, 2025):

@crazywoola The user lhtpluto has indicated that the issue involving Xinference plugins causing disconnections with 18K character inputs is still relevant and needs attention despite being closed as stale. Could you please assist with this?

@dosubot[bot] commented on GitHub (Aug 28, 2025): @crazywoola The user lhtpluto has indicated that the issue involving Xinference plugins causing disconnections with 18K character inputs is still relevant and needs attention despite being closed as stale. Could you please assist with this?
yindo commented 2026-02-16 10:19:07 -05:00
Author
Owner
Copy Link
Copy Source

@Fucheng-Wu commented on GitHub (Sep 1, 2025):

same issue

@Fucheng-Wu commented on GitHub (Sep 1, 2025): same issue
yindo commented 2026-02-16 10:19:07 -05:00
Author
Owner
Copy Link
Copy Source

@dosubot[bot] commented on GitHub (Sep 17, 2025):

Hi, @lhtpluto. I'm Dosu, and I'm helping the dify-official-plugins team manage their backlog and am marking this issue as stale.

Issue Summary:

  • You reported server disconnections when using the Xinference plugin with Qwen3-235B-A22B-8bit and DeepSeek-R1-0528-4bit models on 18K character inputs.
  • The 4bit Qwen3 model and ollama plugin do not have this issue with the same input size.
  • You noted LM Studio also experiences similar disconnection problems with these models.
  • The issue was previously marked stale but reopened after you confirmed it remains unresolved.
  • Another user, Fucheng-Wu, has confirmed experiencing the same problem.

Next Steps:

  • Please let me know if this issue is still relevant with the latest version of the dify-official-plugins repository by commenting here.
  • If I do not hear back within 5 days, I will automatically close this issue.

Thank you for your understanding and contribution!

@dosubot[bot] commented on GitHub (Sep 17, 2025): Hi, @lhtpluto. 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 server disconnections when using the Xinference plugin with Qwen3-235B-A22B-8bit and DeepSeek-R1-0528-4bit models on 18K character inputs. - The 4bit Qwen3 model and ollama plugin do not have this issue with the same input size. - You noted LM Studio also experiences similar disconnection problems with these models. - The issue was previously marked stale but reopened after you confirmed it remains unresolved. - Another user, Fucheng-Wu, has confirmed experiencing the same problem. **Next Steps:** - Please let me know if this issue is still relevant with the latest version of the dify-official-plugins repository by commenting here. - If I do not hear back within 5 days, I will automatically close this issue. Thank you for your understanding and contribution!
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
bug
bug
bug
dependencies
documentation
enhancement
good first issue
invalid
pull-request
Mirrored from GitHub Pull Request
 question
stale
No Label bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: langgenius/dify-official-plugins#359
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?